Adobe.after.effects.cs4 visual.effects.and.compositing.studio.techniques

Find the media files for this eBook at:
www.peachpit.com/ebookfiles/0321592018

Adobe ®

After Effects CS4 ®

STUDIO TECHNIQUES
Mark Christiansen


Adobe® After Effects® CS4 Visual Effects and Compositing Studio Techniques
Mark Christiansen

This Adobe Press book is published by Peachpit.
For information on Adobe Press books, contact:

Peachpit
1249 Eighth Street
Berkeley, CA 94710
(510) 524-2178
Fax: (510) 524-2221

To report errors, please send a note to errata@peachpit.com

Peachpit is a division of Pearson Education
Copyright © 2009 Mark Christiansen
For the latest on Adobe Press books, go to www.adobepress.com

Project Editor: Karyn Johnson
Development and Copy Editor: Linda Laflamme
Production Editor: Becky Winter
Technical Editor: Alex Czetwertynski
Proofreader: Rebecca Rider
Composition: David Van Ness
Indexer: Karin Arrigoni
Cover design: Peachpit Press/Charlene Will
Cover illustration: Regina Cleveland

Notice of Rights
All rights reserved. No part of this book may be reproduced or transmitted in any form by any means, elec-
tronic, mechanical, photocopying, recording, or otherwise, without the prior written permission of the pub-
lisher. For information on getting permission for reprints and excerpts, contact permissions@peachpit.com.

Notice of Liability
The information in this book is distributed on an “As Is” basis, without warranty. While every precaution has been
taken in the preparation of the book, neither the author nor Peachpit shall have any liability to any person or
entity with respect to any loss or damage caused or alleged to be caused directly or indirectly by the instructions
contained in this book or by the computer software and hardware products described in it.

Trademarks
Many of the designations used by manufacturers and sellers to distinguish their products are claimed as trade-
marks. Where those designations appear in this book, and Peachpit was aware of a trademark claim, the des-
ignations appear as requested by the owner of the trademark. All other product names and services identified
throughout this book are used in editorial fashion only and for the benefit of such companies with no intention
of infringement of the trademark. No such use, or the use of any trade name, is intended to convey endorse-
ment or other affiliation with this book.

ISBN 13: 978-0-321-59201-9
ISBN 10: 0-321-59201-8

9 8 7 6 5 4 3 2 1

Printed and bound in the United States of America


Contents

Foreword xi

Introduction xix

Section I Working Foundations 1
Chapter 1 Composite in After Effects 3
Workspaces and Panels 4
Order Reduces Effort 8
Project, Footage, and Composition Settings 15
Previews and View Panels 24
Effects & Presets 33
Output via the Render Queue 33
Study a Shot like an Effects Artist 37

Chapter 2 The Timeline 39
Organization 40
Keyframes and the Graph Editor 46
Über-duper 58
Spatial Offsets 61
Motion Blur 64
Manipulate Time 68
Conclusion 75

Chapter 3 Selections: The Key to Compositing 77
Selection Types 79
Compositing: Science and Nature 83
Alpha Channels and Premultiplication 87
Masks 91
Combine Selections 95
Masks in Motion 98
Blending Modes: Compositing Beyond Selections 100
Track Mattes 106
Conclusion 108

Chapter 4 Optimize the Pipeline 109
Multiple Compositions, Multiple Projects 110
Adjustment and Guide Layers 119
Render Pipeline 122
Project Optimization 129
Conclusion 132

iii

Section II Effects Compositing Essentials 133
Chapter 5 Color Correction 135
Optimized Levels 137
Color Matching 161
Conclusion 174

Chapter 6 Color Keying 175
Good Habits and Best Practices 176
Linear Keyers and Hi-Con Mattes 179
Blue and Green Screen Keys 184
Get the Best Out of Keylight 193
Beyond Keylight: Better Mattes 201
Conclusion 209

Chapter 7 Rotoscoping and Paint 211
Articulated Mattes 213
Beyond Built-in Limitations 216
Morph 219
Puppet 225
Paint and Cloning 229
Conclusion 235

Chapter 8 Effective Motion Tracking 237
Point Tracking Essentials 239
Match Multiple Objects 249
Stabilize a Moving Shot 252
Incorporate MochaAE 255
Use Tracking with Expressions 261
Import 3D Tracking Data 263
Conclusion 268

Chapter 9 The Camera and Optics 269
Cameras: Virtual and Real 271
Storytelling and the Camera 284
Camera Blur 290
The Role of Grain 295
Film and Video Looks 301
Conclusion 308

Chapter 10 Expressions 309
What Expressions Are 310
Creating Expressions 312
The Language of Expressions 314
Linking an Effect Parameter to a Property 314

iv

Using a Layer’s Index 316
Looping Keyframes 318
Using Markers 320
Time Remapping Expressions 324
Layer Space Transforms 328
Color Sampling and Conversion 337
Extra Credit 339
Conclusion 344

Chapter 11 32-Bit HDR Compositing
and Color Management 345
Color Management: Why Bother? 347
Film and Dynamic Range 360
Linear Floating Point HDR 369
Conclusion 378

Section III Creative Explorations 379
Chapter 12 Light 381
Source and Direction 382
Color Looks 386
Source, Reﬂection, and Shadow 390
Multipass 3D Compositing 399

Chapter 13 Climate and the Environment 405
Particulate Matter 406
Sky Replacement 410
Fog, Smoke, and Mist 413
Billowing Smoke 417
Wind 422
Precipitation 423

Chapter 14 Pyrotechnics: Heat, Fire, Explosions 429
Firearms 430
Energy Effects 435
Heat Distortion 439
Fire 442
Explosions 446
In a Blaze of Glory 447

Index 448

Scripting Appendix by Jeff Almasol and
After Effects JavaScript Guide by Dan Ebberts
available on the accompanying DVD-ROM

v

About the Author
Mark Christiansen is a San Francisco-based visual effects
supervisor and creative director. He has worked with visual
effects companies, including The Orphanage, on several
Hollywood feature films such as The Day After Tomorrow and
Pirates of the Caribbean 3: At World’s End. As a director, pro-
ducer, designer, and compositor/animator, he has worked
on a diverse slate of commercial, music video, live event,
and television documentary projects for clients as diverse
as Sony, Interscope, HBO, and many of the world’s best-
known Silicon Valley companies. A little music video he
directed and designed in After Effects was featured in the
L.A. Shorts Fest.
Mark has used After Effects since version 2.0 and has
worked directly with the After Effects development and
marketing teams; he was once named the #1 After Effects
beta tester (for version 6.0). A Contributing Editor at DV
Magazine, he is also a founder of ProVideoCoalition. He
has written three previous editions of this book as well
as After Effects 5.5 Magic (with Nathan Moody), and has
contributed to other published efforts including the Class-
room in a Book. He has also created video training and for
the last couple of years has been a professor at fxphd.com.
He is a Phi Beta Kappa graduate of Pomona College. You
can find him at flowseeker.com and christiansen.com, and
on Twitter as Flowseeker.

vi

About the Contributors
Jeff Almasol (Appendix: Scripting) is a
quality engineer on the Adobe After Effects
team by day and crafter of After Effects
scripts at his redefinery.com site by night.
His site provides numerous free scripts,
reference material, and links to other
scripting resources. Prior to Adobe, Jeff worked at Elastic
Reality Inc. and Avid Technology on Elastic Reality, Mar-
quee, AvidProNet, and other products; and at Profound
Effects on Useful Things and Useful Assistants. You might
find him talking in the third person on Twitter (redefin-
ery) and other sites.
Dan Ebberts (Chapter 10: Expressions
and After Effects Javascript Guide) is a
freelance After Effects script author and
animation consultant. His scripting services
have been commissioned for a wide range
of projects, including workflow automation
and complex animation rigging. He is a frequent contribu-
tor to the various After Effects forums and has a special
interest in expressions and complex algorithms. Dan is an
electrical engineer by training, with a BSEE degree from
the University of California, but has spent most of his
career writing software. He can be reached through his
web site at http://motionscript.com.
Stu Maschwitz (Foreword) is a cofounder
and the CTO of The Orphanage, a San
Francisco-based visual effects and film
production company. Maschwitz spent four
years as a visual effects artist at George
Lucas’s Industrial Light & Magic (ILM),
working on such films as Twister and Men in Black, and went
on to create the award-winning Magic Bullet software. At
The Orphanage, Maschwitz has directed numerous
commercials and supervised effects work on films includ-
ing Sin City and The Spirit. Maschwitz is a guerilla filmmaker
at heart and combined this spirit and his effects knowledge
into a book: The DV Rebel’s Guide: An All-Digital Approach to
Making Killer Action Movies on the Cheap (Peachpit Press).

vii

Acknowledgments
This book would never have come about had After Effects
itself not been such a compelling tool that it was worth
becoming an expert at using it. Thank you to the After
Effects team for its integrity and willingness to hear criti-
cism with equanimity, reshaping this tool so that it remains
fully viable more than 15 years into its existence.
Had I not had the opportunity to work in the studio that
was pushing After Effects beyond where anyone else
thought it could go, The Orphanage, I would never have
gained the clear picture of visual effects fundamentals
that became the foundation of this book. And had Stu
Maschwitz, CTO and co-founder of The Orphanage, not
patiently traded a few hundred emails with me regard-
ing the first edition of this book, it would not have been
as strong an initial effort; had his support not remained
in place for the re-edits, I would not have been so able to
improve upon that first try. Ask anyone who is the most
influential user of After Effects: It’s Stu.
If on the other hand, you asked whom to go to for more
expressions and scripting knowledge, you would get this edi-
tion’s two new contributing authors, Dan Ebberts and Jeff
Almasol. Having cited Dan’s motionscript.com site for years
as the place to learn about expressions, I was able to con-
vince him to take on Chapter 10 and rewrite it from scratch;
he put in a lot of effort to provide what I hope you will find
a deep and inspirational treatment of the subject even if you
are already familiar with it. Jeff Almasol took on the task of
taking the topic of scripting, the most technical and nerdy
corner of After Effects short of writing third-party effects,
and making it accessible and friendly to the code-phobic. I
really have to hand it to Jeff; he tackled the chapter, which
is Appendix A on the disc, with real enthusiasm.
Thank you to Brendan Bolles who wrote such an effec-
tive Chapter 11 in the first edition of the book that I have
still kept many of his figures, ideas, and even huge swaths
of his writing intact. Congratulations to Brendan on
having his EXR plug-ins officially licensed by Adobe for
After Effects CS4.

viii

Script coverage in this book is vastly increased thanks also
to Lloyd Alvarez whose site aescripts.com has become a
vital source for some truly fantastic scripts. He knows how
to make them because he makes them for his own needs.
Sean Kennedy and Charles Bordenave (nabscripts.com)
offer TrackerViz, which really boosts the effectiveness of
the After Effects tracker. Thanks also to Sean for offering
examples from his own film projects.
In this edition there are more figure and footage con-
tributors than ever to thank. Thanks for footage to John
Flowers and Case Films, as well as Alex Lindsay and Pixel
Corps; thanks also to Brandon Schilling at Kontent Films
for shooting b-roll with me one afternoon. Thanks as
always to Julie Hill and Artbeats, and my good friends and
colleagues at fxphd.com, John Montgomery and Mike
Seymour, along with the vast worldwide network of art-
ists associated with that site from whom I’ve learned as I
taught. Pete O’Connell contributed samples from his fan-
tastic DVD Advanced Rotoscoping Techniques for After Effects,
available at creativecow.com.
Members of the worldwide community of After Effects
artists offered imagery for this book. Locally, I can thank
partners in crime Kontent Films: Mark Decena for stills
from his feature film Dopamine, Eric Escobar, author of
the great PrepShootPost blog, and from the network of
San Francisco independent filmmakers, Benjamin Morgan,
director of Quality of Life, as well as Matt Ward. Further
afield, thanks to Ross Webb at Mars Productions in Cape
Town, South Africa and Luis Bustamente (4charros)
in Mexico.
Clients gave me the firsthand experience that went into
this book, and some were able to help me secure examples
to share: Christina Crowley, President of The Kenwood
Group; Michael Brynteson at Sony; Rama Dunayevich at
The Orphanage; Coral Petretti at ABC Photography; David
Donegan at Red Bull USA; Tim Fink of Tim Fink Events
and Media; Gary Jaeger and Cameron Baxter at Core Stu-
dio; Jonathan Barson at The Foundry UK; Fred Lewis and
Inhance Digital; Boeing and the Navy UCAV program.

ix

I mined http://flickr.com/creativecommons/ for a few
difficult-to-find source stills; a huge thank you to the gifted
photographers who voluntarily chose to add the Creative
Commons tag to their work: Micah Parker, Jorge L. Peschi-
era, Shuets Udono, Eric E. Yang, and Kevin Miller.
To the people at Adobe who’ve made After Effects what it
is, in particular Dave Simons, Dan Wilk, Erica Schisler, and
Steve Kilisky, and to some of the developers who’ve helped
me understand it better over the years, including Michael
Natkin and Chris Prosser. Huge thanks this time around to
Todd Kopriva who personally delivered his marked-up copy
of the previous edition to my office in San Francisco.
Thanks to the companies that contributed to the book’s
DVD: Peder Norrby, who is Trapcode, Russ Andersson
of Andersson Technologies, Sean Safreed of Red Giant
Software, Andrew Millin of ObviousFX LLC, Marco Paolini
of SilhouetteFX, Pierre Jasmin and Pete Litwinowicz of
RevisionFX, and Philipp Spoth of Frischluft. I don’t include
tools that I wouldn’t endorse.
Other people who were helpful with technical details in
earlier versions of this book include Bruno Nicoletti at
the Foundry UK, Scott Squires (www. effectscorner.com),
Tim Dobbert at The Orphanage, and Don Shay at Cinefex
Magazine.
A huge thank you to Peachpit, who collectively show a
strong commitment to producing the highest quality
books, in particular Karyn Johnson who boldly stepped in
and supplied blunt insight, humor, fun, vitamin C, archival
footage, patience, whatever it took to keep the book on
track. I trust Linda Laflamme to tell me when I’m not mak-
ing sense. Alexandre Czetwertynski is a fellow artist I trust
to give the book an honest technical read.
Thanks to all of the thoughtful folks who have dropped me
a line at aestudiotechniques@gmail.com; your comments
and questions matter.

x

Foreword
I can’t see the point in the theatre. All that
sex and violence. I get enough of that at home.
Apart from the sex, of course.
—Tony Robinson as Baldrick, Blackadder

Who Brings the Sex?
“Make it look real.” That would seem to be the mandate
of the visual effects artist. Spielberg called and he wants
the world to believe, if only for 90 minutes, that dinosaurs
are alive and breathing on an island off the coast of South
America. Your job: make them look real. Right?
Wrong.
I am about to tell you, the visual effects artist, the most
important thing you’ll ever learn in this business: Making
those Velociraptors (or vampires or alien robots or burst-
ing dams) “look real” is absolutely not what you should be
concerned with when creating a visual effects shot.
Movies are not reality. The reason we love them is that they
present us with a heightened, idealized version of reality.
Familiar ideas—say, a couple having an argument—but
turned up to eleven: The argument takes place on the
observation deck of the Empire State building, both he
and she are perfectly backlit by the sun (even though
they’re facing each other), which is at the exact same just-
about-to-set golden-hour position for the entire ten-minute
conversation. The couple are really, really charming and
impossibly good-looking—in fact, one of them is Meg
Ryan. Before the surgery. Oh, and music is playing.
What’s real about that? Nothing at all—and we love it.
Do you think director Alejandro Amenábar took Javier
Aguirresarobe, cinematographer on The Others, aside and
said, “Whatever you do, be sure to make Nicole Kidman
look real?” Heck no. Directors say this kind of stuff to their
DPs: “Make her look like a statue.” “Make him look bullet-
proof.” “Make her look like she’s sculpted out of ice.”

xi

Foreword

Did It Feel Just Like It Should?
Let’s roll back to Jurassic Park. Remember how terrific the
T-Rex looked when she stepped out of the paddock? Man,
she looked good.
She looked good.
The realism of that moment certainly did come in part
from the hard work of Industrial Light and Magic’s fledg-
ling computer graphics department, who developed
groundbreaking technologies to bring that T-Rex to life.
But mostly, that T-Rex felt real because she looked good. She
was wet. It was dark. She had a big old Dean Cundey blue
rim light on her coming from nowhere. In truth, you could
barely see her.
But you sure could hear her. Do you think a T-Rex
approaching on muddy earth would really sound like the
first notes of a new THX trailer? Do you think Spielberg
ever sat with sound designer Gary Rydstrom and said,
“Let’s go out of our way to make sure the footstep sounds
are authentic?” No, he said, “Make that mofo sound like
the Titanic just rear-ended the Hollywood Bowl” (may or
may not be a direct quote).
It’s the sound designer’s job to create a soundscape for a
movie that’s emotionally true. They make things feel right
even if they skip over the facts in the process. Move a gun
half an inch and it sounds like a shotgun being cocked. Get
hung up on? Instant dial tone. Modern computer display-
ing something on the screen? Of course there should be
the sound of an IBM dot-matrix printer from 1978.
Sound designers don’t bring facts. They bring the sex. So
do cinematographers, makeup artists, wardrobe stylists,
composers, set designers, casting directors, and even the
practical effects department.
And yet somehow, we in the visual effects industry are
often forbidden from bringing the sex. Our clients pigeon-
hole us into the role of the prop maker: Build me a T-Rex,
and it better look real. But when it comes time to put that
T-Rex on screen, we are also the cinematographer (with
our CG lights), the makeup artist (with our “wet look”

xii

Foreword

shader), and the practical effects crew (with our rain). And
although he may forget to speak with us in the same flow-
ery terms that he used with Dean on set, Steven wants us to
make that T-Rex looks like a T-Rex should in a movie. Not
just good—impossibly good. Unrealistically blue-rim-light-
outa-nowhere good. Sexy good.
Have you ever argued with a client over aspects of an
effects shot that were immutable facts? For example, you
may have a client that inexplicably requested a little less
motion blur on a shot, or that told you “just a little slower”
for an object after you calculated its exact rate of fall? Do
you ever get frustrated with clients who try to art-direct
reality in this way?
Well, stop it.
Your client is a director, and it’s their job to art-direct real-
ity. It’s not their job to know (or suggest) the various ways
that it may or may not be possible to selectively reduce
motion blur, but it is their job to feel it in their gut that
somehow this particular moment should feel “crisper” than
normal film reality. And you know what else? It’s your job
to predict that they might want this and even propose it.
In fact, you’d better have this conversation early, so you
can shoot the plate with a 45-degree shutter, that both
the actors and the T-Rex might have a quarter the normal
motion blur.

Was It Good for You?
The sad reality is that we, the visual effects industry,
pigeonhole ourselves by being overly preoccupied with real-
ity. We have no one to blame but ourselves. No one else
on the film set does this. If you keep coming back to your
client with defenses such as “That’s how it would really
look” or “That’s how fast it would really fall,” then not only
are you going to get in some arguments that you will lose,
but you’re actually setting back our entire industry by per-
petuating the image of visual effects artists as blind to the
importance of the sex.
On the set, after take one of the spent brass shell falling
to the ground, the DP would turn to the director and say,

xiii

Foreword

“That felt a bit fast. Want me to do one at 48 frames?” And
the director would say yes, and they’d shoot it, and then
months later the editor would choose take three, which
they shot at 72 frames per second “just in case.” That’s
the filmmaking process, and when you take on the task of
creating that same shot in CG, you need to represent, emu-
late, and embody that entire process. You’re the DP, both
lighting the shot and determining that it might look better
overcranked. You’re the editor, confirming that choice in
the context of the cut. And until you show it to your client,
you’re the director, making sure this moment feels right in
all of its glorious unreality.
The problem is that the damage is already done. The
client has worked with enough effects people who have
willingly resigned themselves to not bringing the sex that
they now view all of us as geeks with computers rather
than fellow filmmakers. So when you attempt to break our
self-imposed mold and bring the sex to your client, you
will face an uphill battle. But here’s some advice to ease
the process: Do it without asking. I once had a client who
would pick apart every little detail of a matte painting,
laying down accusations of “This doesn’t look real!”—until
we color corrected the shot cool, steely blue with warm
highlights. Then all the talk of realism went away, and the
shot got oohs and ahs.
Your client reacts to your work emotionally, but they critique
technically. When they see your shot, they react with their
gut. It’s great, it’s getting better, but there’s still something
not right. What they should do is stop there and let you
figure out what’s not right, but instead, they somehow
feel the need to analyze their gut reaction and turn it into
action items: “That highlight is too hot” or “The shadows
under that left foot look too dark.” In fact it would be bet-
ter if they focused on vocalizing their gut reactions: “The
shot feels a bit lifeless,” or “The animation feels too heavy
somehow.” Leave the technical details to the pros.
You may think that those are the worst kind of com-
ments, but they are the best. I’ve seen crews whine on
about “vague” client comments like “give the shot more
oomf.” But trust me, this is exactly the comment you want.

xiv

Foreword

Because clients are like customers at a restaurant, and
you are the chef. The client probably wants to believe that
“more oomf” translates into something really sophisticated,
like volumetric renderings or level-set fluid dynamics, in
the same way that a patron at a restaurant would hope that
a critique like “this dish needs more flavor” would send
the chef into a tailspin of exotic ingredients and tech-
niques. Your client would never admit (or suggest on their
own) that “oomf” is usually some combination of “cheap
tricks” such as camera shake, a lens flare or two, and pos-
sibly some God rays—just like the diner would rather not
know that their request for “more flavor” will probably be
addressed with butter, salt, and possibly MSG.
The MSG analogy is the best: Deep down, you want to go
to a Chinese restaurant that uses a little MSG but doesn’t
admit it. You want the cheap tricks because they work, but
you’d rather not think about it. Your client wants you to
use camera shake and lens flares, but without telling them.
They’d never admit that those cheap tricks “make” a shot,
so let them off the hook and do those things without being
asked. They’ll silently thank you for it. Bringing the sex is
all about cheap tricks.

Lights On or Off?
There are some visual effects supervisors who pride
themselves on being sticklers for detail. This is like being
an architect whose specialty is nails. I have bad news for
the “Pixel F*ckers,” as this type are known: Every shot will
always have something wrong with it. There will forever be
something more you could add, some shortcoming that
could be addressed. What makes a visual effects supervisor
good at their job is knowing which of the infinitely pos-
sible tweaks are important. Anyone can nitpick. A good
supe focuses the crew’s efforts on the parts of the shot that
impact the audience most. And this is always the sex. Audi-
ences don’t care about matte lines or mismatched black
levels, soft elements or variations in grain. If they did, they
wouldn’t have been able to enjoy Blade Runner or Back to the
Future or that one Star Wars movie—what was it called? Oh
yeah: Star Wars. Audiences only care about the sex.

xv

Foreword

On a recent film I was struggling with a shot that was just
kind of sitting there. It had been shot as a pick-up, and it
needed some help fitting into the sequence that had been
shot months earlier. I added a layer of smoke to techni-
cally match the surrounding shots. Still, the shot died on
the screen. Finally, I asked my compositor to softly darken
down the right half of the shot by a full stop, placing half
the plate along with our CG element in a subtle shadow.
Boom, the shot sang.
What I did was, strictly speaking, the job of the cinematog-
rapher, or perhaps the colorist. The colorist, the person
who designs the color grading for a film, is the ultimate
bringer of the sex. And color correction is the ultimate
cheap trick. There’s nothing fancy about what a Da Vinci
2K or an Autodesk Lustre does with color. But what a good
colorist does with those basic controls is bring heaping,
dripping loads of sex to the party. The problem is (and I
mean the problem—the single biggest problem facing our
industry today), the colorist gets their hands on a visual
effects shot only after it has already been approved. In other
words, the film industry is currently shooting itself in
the foot (we, the visual effects artists, being that foot) by
insisting that our work be approved in a sexless environ-
ment. This is about the stupidest thing ever, and until the
industry works this out, you need to fight back by taking
on some of the role of the colorist as you finalize your
shots, just like we did when we made those matte paintings
darker and bluer with warm highlights.
Filmmaking is a battleground between those who bring the
sex and those who don’t. The non-sex-bringing engineers
at Panavision struggle to keep their lenses from flaring,
while ever-sexy cinematographers fight over a limited stock
of 30-year-old anamorphic lenses because they love the
flares. I’ve seen DPs extol the unflinching sharpness of a
priceless Panavision lens right before adding a smear of
nose grease (yes, the stuff on your nose) to the rear ele-
ment to soften up the image to taste. Right now this battle
is being waged on every film in production between the
visual effects department and the colorists of the world.
I’ve heard effects artists lament that after all their hard

xvi

Foreword

work making something look real, a colorist then comes
along and “wonks out the color.” In truth, all that colorist
did was bring the sex that the visual effects should have
been starting to provide on their own. If what the colorist
did to your shot surprised you, then you weren’t thinking
enough about what makes a movie a movie.

In Your Hands
You’re holding a book on visual effects compositing in
Adobe After Effects. There are those who question the
validity of such a thing. Some perpetuate a stigma that
After Effects is for low-end TV work and graphics only. To
do “real” effects work, you should use a program such as
Nuke or Shake. Those techy, powerful applications are
good for getting shots to look technically correct, But they
do not do much to help you sex them up. After Effects may
not be on par with Nuke and Shake in the tech depart-
ment, but it beats them handily in providing a creative
environment to experiment, create, and reinvent a shot.
In that way it’s much more akin to the highly-respected
Autodesk Flame and Inferno systems—it gives you a broad
set of tools to design a shot, and has enough horsepower for
you to ﬁnish it too. It’s the best tool to master if you want
to focus on the creative aspects of visual effects compos-
iting. That’s why this book is unique. Mark’s given you
the good stuff here, both the nitty-gritty details as well as
the aerial view of extracting professional results from an
application that’s as maligned as it is loved. No other book
combines real production experience with a deep under-
standing of the fundamentals, aimed at the most popular
compositing package on the planet.

Bring It
One of the great matte painters of our day once told me
that he spent only the ﬁrst few years of his career strug-
gling to make his work look real, but that he’ll spend the
rest of his life learning new ways of making his work look
good. It’s taken me years of effects supervising, commercial
directing, photography, wandering the halls of museums,
and waking up with hangovers after too much really good

xvii

Foreword

wine to fully comprehend the importance of those words.
I can tell you that it was only after this particular matte
painter made this conscious choice to focus on making
things look good, instead of simply real, that he skyrock-
eted from a new hire at ILM to one of their top talents.
Personally, it’s only after I learned to bring the sex that
I graduated from visual effects supervising to become a
professional director.
So who brings the sex? The answer is simple: The people
who care about it. Those who understand the glorious
unreality of ﬁlm and their place in the process of creating
it. Be the effects artist who breaks the mold and thinks
about the story more than the bit depth. Help turn the
tide of self-inﬂicted prejudice that keeps us relegated to
creating boring reality instead of glorious cinema. Secretly
slip your client a cocktail of dirty tricks and fry it in more
butter than they’d ever use at home.
Bring the sex.

Stu Maschwitz
San Francisco, October 2008

xviii

If you aren’t fired with enthusiasm, you will
be fired—with enthusiasm.
—Vince Lombardi

Why This Book?

A dobe After Effects CS4 Visual Effects and Compositing Studio
Techniques is about creating visual effects—the art and
science of making disparate elements look like they were
taken with a single camera, and of making an ordinary shot
extraordinary yet believable. It goes deep into issues such
as color correction and keying that are only touched on
by books more focused on using After Effects for motion
graphics, while leaving motion-graphics-only tools (Text,
Shapes, and like) more or less alone.
This book does not shy away from strong opinions, even
when they deviate from the official line. These opinions
have been formed through actual work in production at
a few of the finest visual effects facilities in the world, and
they’re valid not only for “high-end” productions but for
any composited shot. Where applicable, the reasoning
behind using one technique over another is provided. I
aim to make you not a better button-pusher but a more
effective artist and technician.
The visual effects industry is historically protective of trade
secrets, often reflexively treating all production informa-
tion as proprietary. Work on a major project, however, and
you will soon discover that even the most complex shot is
made up largely of repeatable techniques and practices;
the art is in how these are applied, combined and custom-
ized, and what is added (or taken away).
Each shot is unique, and yet every shot relies on techniques
that are tried and true. This book offers you as much of the
latter as possible so that you can focus on the former. There’s
not much here in the way of step-by-step instructions; it’s
more important that you grasp how things work so that you
can repurpose the technique for your individual shot.

xx

Introduction

This is emphatically not a book for beginners. Although
the first section is designed to make sure you are making
optimal use of the software, it’s not an effective primer
on After Effects in particular or digital video in general.
If you’re new to After Effects, first spend some time with
its excellent documentation or check out one of the many
books available to help beginners learn to use After Effects.

Some Key Steps
There are a few overall keys to your success as a compositor
and visual effects artist, whether or not you work in After
Effects:
. Get reference. You can’t re-create what you can’t clearly
see. Too many artists skip this step.
. Simplify. This book is about helping you eliminate
needless steps. To paraphrase Einstein, a good solution
is as simple as possible, but no simpler.
. Break it down. As I said above, the most complicated
shot consists of small, comprehensible steps—perhaps
thousands of them—and each image consists of three
or more channels each containing thousands of pixels.
. Don’t expect a perfect result on the first try. My old col-
league Paul Topolos (at this writing employed in the art
department at Pixar) used to say that “recognizing flaws
in your work doesn’t mean you’re a bad artist. It only
means you have taste.”
This is how it’s done at the best studios, and even if you’re
not currently working at one of them, this is how you
should do it, too.

Organization of this Book
Adobe After Effects CS4 Visual Effects and Compositing Studio
Techniques is organized into three sections:
. Section I, “Working Foundations,” is about the soft-
ware. The goal is not to drag you through each menu
and button and be a second manual, but instead to
offer you tips and techniques that will help you into
the coveted state of flow in After Effects, where you are

xxi

Introduction

focused entirely on the job at hand because you no
longer have to think about the tools.
Don’t assume that you’re too advanced to at least skim
this section; I guarantee there’s information in there
you don’t already know.
. Section II, “Effects Compositing Essentials,” focuses
on the core techniques of effects compositing: color
matching, keying, rotoscoping, motion tracking, and
optics, as well as such advanced topics as expressions
and HDR color. This is the heart of the book.
. Section III, “Creative Explorations,” demonstrates actual
shots you are likely to re-create, offering best practices
for techniques every effects artist needs to know.
What you won’t find in these sections are menu-by-menu
descriptions of the interface or step-by-step tutorials that
walk you through projects with little connection to real-
world visual effects needs.

The Unique After Effects Workflow
Some users may be coming to this book unfamiliar with
After Effects but experienced in other compositing soft-
ware. Here’s a brief overview of how the After Effects work-
flow is unique from every other compositing application
out there. Each application is unique, and yet the main
competitors to After Effects—Nuke, Shake, Flame, Fusion
and Toxic, to name a few—are probably more similar to
one another than any of them is to After Effects, which is
in many ways a lot more like Photoshop.
Here are some of the features which can make After Effects
easier for the beginner to use, but can constrain others:
. Render order is established in the Timeline and via
nested compositions: layers, not nodes. After Effects has
Flowchart view but you don’t create your composition
there the way you would with a tree/node interface.
. Transforms, effects, and masks are embedded in every
layer and render in a fixed order.
. After Effects has a persistent concept of an alpha chan-
nel in addition to the three color channels. The alpha

xxii

Introduction

is always treated as if it is straight (not premultiplied)
once an image has been imported and interpreted.
. An After Effects project is not a script, although ver-
sion CS4 introduces a text version of the After Effects
Project (.aep) file, the XML-formatted .aepx file. Most
of its contents are inscrutable other than source file
paths. Actions are not recordable and there is no direct
equivalent to Shake macros.
. Temporal and spatial settings tend to be absolute in
After Effects because it is composition and timeline-
based. This is a boon to projects that involve complex
timing and animation, but it can snare users who aren’t
used to it and suddenly find pre-comps that end prema-
turely or are cropped. Best practices to avoid this are
detailed in Chapter 4.
Of these differences, some are arbitrary, most are a mixed
bag of advantages and drawbacks, and a couple of them
are constantly used by the competition as a metaphorical
stick with which to beat After Effects. The two that come
up the most are the handling of precomposing and the
lack of macros.
This book attempts to shed light on these and other areas
of After Effects that are not explicitly dealt with in its user
interface or documentation. After Effects itself spares you
details that as a casual user, you might never need to know
about, but that as a professional user you should under-
stand thoroughly. This book is here to help.

New in this Edition
After Effects CS4 is the strongest version of the software
yet, but with a few exceptions, including a new user inter-
face with a number of revamped workflows, it does not
change the methodology of most of what was in the previ-
ous edition.
Where After Effects may not have evolved, however, my
own thinking has continued to do so; so even techniques
that would be equally valid in a previous version of the soft-
ware are newly presented or revamped here. In particular, I
have paid attention in this edition to the first few chapters

xxiii

Introduction

in order to better help smart people who don’t yet under-
stand the fundamentals.
Huge changes to the book this time around are thanks to
two new contributors: Dan Ebberts and Jeff Almasol.

What’s on the DVD
Two of the biggest additions to this book unfortunately did
not make it into print due to page constraints.
Jeff Almasol’s Scripting chapter is the Appendix, found on
the disc as a PDF. It is the most accessible resource avail-
able on this complicated and much feared topic, walking
you through three scripts each of which builds upon the
complexity of the previous. Scripting provides the ability
to create incredibly useful extensions to After Effects to
eliminate tedious tasks. Several of these are included in the
redeﬁnery folder as exclusives to this book.
In order to focus on more advanced and applied topics
in the print edition, Dan Ebberts kicked JavaScript funda-
mentals to a special JavaScript addendum also included as
a PDF. This is in many ways the “missing manual” for the
After Effects implementation of JavaScript, omitting all
of the useless web-only scripting commands found in the
best available books, but extending beyond the material in
After Effects help.
If you want to ﬁnd out more about some of the plug-ins
and software mentioned in this book, look no further than
its DVD-ROM. For example, the disc includes demos of
. Andersson Technologies’ SynthEyes (3D tracking
software)
To install the lesson files, footage, . Red Giant Software’s Magic Bullet Looks, Knoll Light
and software demos included on
Factory Unmult, Knoll Light Factory Pro, Key Correct
the DVD, simply copy each chapter
folder in its entirety to your hard Pro, Magic Bullet Colorista, Trapcode Lux, Trapcode
drive. Note that all .aep files are Horizon, Trapcode Form, Trapcode Particular, the
located in the Projects subfolder brand-new Warp and more.
of each chapter folder on the disc,
while .ffx files can be found in the . ReelSmart Motion Blur and PV Feather from
Animation Presets subfolders. Revision FX
. Lenscare and ZBornToy from Frischluft
. Erodilation and CopyImage from ObviousFX

xxiv

Introduction

You’ll also find HD footage from Kontent Films, fxphd.
com, and Pixel Corps with which you can experiment and
practice your techniques. There are dozens of example
files to help you deconstruct the techniques described.
There are also a few useful and free third-party scripts; for
more of these, see the script links listed in the aforemen-
tioned appendix on the disc.

The Bottom Line
Just like the debates about which operating system is best,
debates about which compositing software is tops are
largely meaningless—especially when you consider that the If you have comments or ques-
majority of first-rate, big-budget, movie effects extravagan- tions you’d like to share with the
zas are created with a variety of software applications on author, please e-mail them to
AEStudioTechniques@gmail.com.
a few different platforms. Rarely is it possible to say what
software was used to composite a given shot just by looking
at it, because it’s about the artist, not the tools.
The goal is to understand the logic of the software so that
you can use it to think through your artistic and technical
goals. This book will help you do that.

xxv

This page intentionally left blank

SECTI
SECTIO

I
Working Foundations

Chapter 1 Composite in After Effects 3
Chapter 2 The Timeline 39
Chapter 3 Selections: The Key to Compositing 77
Chapter 4 Optimize the Pipeline 109

CHAPTER

1
Composite in After Effects

Good surfing is power, speed, and flow. The rest of it
doesn’t matter to me at all.
—Gary Elkerton, Australian surfer

Composite in After Effects

T his is a book about visual effects compositing. If you
use Adobe After Effects, the goal is to help you create
believable, fantastic moving images from elements that
were not shot together, and to do it with the least possible
effort. This first section of the book focuses on effortless-
ness, offering a jump-start (if you’re new) or a refresher
(if you’re already an After Effects artist) on the After
Effects workflow.
To be an outstanding compositor, you need to employ your
best skills as both an artist and an engineer. As an artist,
you make creative and aesthetic decisions, but if you are
not also able to understand how to put those decisions
together and how the process works, the artistry of your
work will suffer. Artists and engineers have much in com-
mon: both understand and respect the tools, both know
that the tools themselves don’t make you a great designer,
and in both roles, iteration—multiple refinements—are
often what separates a great effort from mediocrity.
This chapter and the rest of Section I focus on how to get
things done in After Effects as effortlessly as possible. It is
assumed that you already know your way around the basics
If this book opens at too advanced
a level for you, check out Adobe of After Effects and are ready to learn to work smarter.
After Effects CS4 Classroom in a Book
(Adobe Press), a helpful beginner’s
resource. Workspaces and Panels
Figure 1.1 is one way of looking at the most generic of
projects: it shows the Standard workspace that appears
when you first open After Effects CS4, broken down into
its component parts. The interface consists of

4

I: Working Foundations

Figure 1.1 The Standard workspace layout, diagrammed in color. The top areas in red are informational only, while the
purple areas contain tools and settings. The blue areas of the Timeline and Effect panel contain stacks whose order will
change compositing order. Panel tabs, in yellow, can be grabbed to reorder the interface and contain menus at the upper
right to adjust their appearance. The layer stack, in green, is modal and can be swapped for the Graph Editor.

. The main application window contains some panel
groups—six of them by default (the Standard work-
space), as few as two (Minimal workspace), or as many
as 17 (All Panels workspace).
. Each group contains one or more panels each with a
tab in the upper left.
. Separating the panel groups are dividers; panels and
dividers are dragged to customize the workspace (more
on that in a moment).
. Some panels are viewers, with a pop-up menu in the tab
listing available clips.
. At the top is the Tools panel, which can be hidden but
only appears atop the application window (and thus has
All available panels in After Effects
no tab). are listed under the Window menu;
I call these out here to be done identifying them and to the most common include preset
keyboard shortcuts. Use these as
reassure you that this, along with a bunch of menus and a
toggles to keep your workspace
bunch more twirly arrows, is all there is to the After Effects clutter-free.
user interface.

5

Chapter 1 Composite in After Effects

Are you a Zen roshi? Reveal the core essence of After
Effects with the Minimal workspace (via the Workspace
menu in Tools or Window > Workspace). Two of the most
important three panels in After Effects are
. The Composition panel, a viewer where you examine a
shot
. The Timeline panel, the true heart of After Effects,
where elements are layered and timed for individual
compositions (or shots). You will typically have many of
these open at any given time.
For the third, choose Window > Project (Cmd/Ctrl+0)
to add the Project panel. This is the Finder or Windows
Explorer of After Effects—nothing more than files and
folders representing the contents of your project.
At some point even a Zen roshi will assumedly need at least
two more panels. A completed composition typically goes
to the Render Queue (Cmd/Ctrl+Alt/Option+0) for final
Prefer your workspace customiza-
tions to the defaults? Choose New
output (details about this important hub are found at the
Workspace in the Workspace menu end of the chapter), and you are likely to add layer effects
and enter an existing name to over- which are best adjusted in the Effect Controls (with a layer
write it; now After Effects will reset selected, F3 or Cmd/Ctrl+Shift+T), although most the
to your customized version.
same controls can be found by twirling open the layer with
that essential little triangle to the left of the layer name.
Other panels (found in other workspaces, or by selecting
them under the Window menu) contain controls for spe-
cific tools such as paint (Paint and Brush Tips), the most
significant of which are covered in detail later in the book.

Customize the Workspace
Simply by choosing the Minimal workspace and adding the
Project panel, you’ve customized the workspace; switch to
another workspace and back and you’ll see that the Project
panel remains until you choose Reset “Minimal” in the
Workspace menu. Want to keep your new configuration?
Save the Workspace under the same menu.
Figure 1.2 Each of the six possible
drop zones is highlighted in color; You can freely move any panel around the UI. To do so,
drag one panel over another to reveal click and drag its tab around the screen. As you move one
each of these targets. Dropping on
panel over another, purple geometric shapes like those
the center or along the top has an
identical result. in Figure 1.2 appear. These are the drop zones. The docking

6


zones along the edges let you place a panel adjacent—
for example, to the left of the Composition panel. The
grouping zones in the center group panels together in one
frame.
Maximize the Screen
Drag a panel to one edge of the application window and Many After Effects artists like two HD-resolution
aqua colored bands appear along the edge; when dropped displays side by side (Figure 1.3, top), although
the panel will occupy that entire side. a single display can be optimal if it’s large enough
(Figure 1.3, bottom)
Double-arrow icons appear when you move the cursor
between two or more panel groups, allowing you to resize
However, you may notice that a floating panel (Ctrl/
adjacent panels. I don’t do this much thanks to one of my Cmd-drag the tab to make it float) lacks the Zoom
favorite After Effects shortcuts, the Tilde key (~), which button along the top to send the window to full
toggles the panel under the cursor to full-screen. screen. The shortcut Ctrl+ (Cmd+) maximizes
and centers any window. Press it again and even the
Which is best for After Effects, one big monitor or two top menu bar toggles off, filling the entire screen.
smaller ones? I think the jury is somewhat out—my studio
currently includes both setups, and although most artists If you’re stuck with a single small display you can
are used to a pair of 24" screens, many prefer a single 30". press Tilde (~) to maximize a single panel and RAM
Preview in full-screen mode by checking the Full
Screen box in the Preview panel.

Figure 1.3 The preferred After Effects
monitor setup seems to be a pair of
2K or larger displays (top) although a
single 30” display at a high resolution
(bottom), used with the Tilde key to
zoom panels to full screen, is also
quite luxuriant.

7


Order Reduces Effort
To my mind, the ideal workflow is an effortless one. Here’s
how to begin reducing steps you will take many, many
times in a normal After Effects workday.
You can tear off any panel and make
it float by holding down the Ctrl/ Figure 1.4 charts the nondestructive workflow of After
Cmd key as you drag it away; I like Effects: from source files to comps made up of layers,
to tear off the Render Queue and
toggle it on and off via its shortcut which can contain settings, all of which are represented in
(Alt+Ctrl+0/Opt+Cmd+0). a project file that never touches those source files.

Figure 1.4 The After Effects pipeline
is made up of a series of stages that
are independent of one another;
everywhere you see one-way arrows
with hash marks, changes are pushed
forward but do not affect the source,
from source images to the Project
to Comps to the Render Queue to
final images.

Import and Organize Source
Getting a source file from a disk or server into After Effects
is no big deal. You can use File > Import > File (or Multiple
Files), or just drag footage directly from the Finder or Win-
dows Explorer into the Project panel. You can also double-
click in an empty area of the Project panel.
Image sequences have a couple of specific extra rules.
I strongly advocate working with a sequence instead of
QuickTime in production for the following reasons:
. An image sequence is less fragile than a QuickTime
movie; if there is a bad frame in a sequence, it can be
replaced, but it will corrupt an entire movie.

8


. You can interrupt and restart an image sequence
render without then having to splice together multiple
movies.
. QuickTime in particular has its own form of color
management that isn’t entirely compatible even
with Apple’s own applications, let alone the Adobe
color management pipeline (explained in depth in
Chapter 11).
Unfortunately, none of the Adobe applications (in particu-
lar, Bridge, which should know better) has ever become
smart about recognizing sequences.
If you have a single sequence in a folder, just drag it in,
but if it’s the first one for this project, leave it selected for
a moment in the Project panel and check its fps setting at
the top; if it’s not correct, see “Project, Footage, and Com-
position Settings” later in this chapter for how to fix this,
both for that clip and for others just like it. If you really
meant to bring in that folder’s contents as individual files,
hold down the Option/Alt key as you drag it in.
Things get more complicated if you are dealing with multi-
ple image sequences in a single folder. If you’ve never run
into this or can simply keep the practice of one sequence
per folder, great, skip ahead. Otherwise, it’s better to use
the Import dialog.
With the Import dialog, it doesn’t matter which specific
image in a sequence you select; they are all imported
provided you select only one. By holding the Shift or Ctrl
(Cmd) key as you select more than one frame, however,
you can
. Specify a subset of frames to be imported instead of an
entire a sequence
. Select frames from more than one sequence in the
same folder; a Multiple Sequences checkbox appears to
make certain this is really what you want to do
. Specify sets of frames from multiple sequences (a com-
bination of the above two modes)

9


This is, in many ways, a work-around for the fact that the
After Effects importer doesn’t group a frame sequence
together the way other compositing applications do.
By default, if a sequence has missing frames (anywhere
the count doesn’t increment by 1), a color bar pattern is
inserted with the name of the ﬁle presumed missing, which
Waiting for a long 3D render?
Render the first and last 3D frames
helps you track it down (see “Missing Footage,” later in this
only, with their correct final chapter).
sequence numbers, and import
them using the Import dialog with The Force Alphabetical Order checkbox in the Import
Force Alphabetical Order unchecked. dialog is for cases where the frame does not increment by
You now have a placeholder of the 1. Say you rendered “on twos,” creating every other frame
correct length that is fully set up as from a 3D app; check this box and you avoid color bars on
soon as the file is rendered.
every other frame.
Want to be rehired repeatedly as a freelancer or be the
hero on your project? Make it easy for someone to open
your project cold and understand how it’s organized.
An ordinary project can be set up like the one shown in
Figure 1.5; I often leave only the main composition in the
root area of the project and place everything else in an
appropriate subfolder.

Figure 1.5 Consider how best to orga-
nize source into folders to keep things
as well organized in the Project as you
would want them on the source drive.
Once you have a structure you like,
you can reuse it. (Chapter 4 says more
about how you might design it.)

10


On a more ambitious project, it’s worth organizing a proj-
ect template so that items are easy to find in predictable
locations. Chapter 4 offers suggestions.

Context-Clicks (and Keyboard Shortcuts)
Stay away from the bar—the top menu bar, that is. I often
refer to context-clicking on interface items. This is “right-
clicking” unless you’re on a Mac laptop or have an ancient
one-button mouse, in which case you can hold down the
Control key. Here’s what happens:
. Context-click a layer in the Timeline for access to the
full Layer menu, minus a few less useful items, such
as the Adobe Encore submenu, and with killer addi-
tions such as Reveal Layer Source in Project and Reveal
Expression Errors.
. Context-click on a layer in a Composition viewer for
many of the same items, plus the Select option at the
bottom of the menu, which gives you all of the items Figure 1.6 One of the biggest pro-
below your pointer (Figure 1.6). ductivity boosts in After Effects comes
from using the right mouse button
. Context-click a panel tab to reveal the panel’s menu
of a two- (or more) button mouse, as
(also found at the upper right), where a bunch of context menus exist throughout After
options that even advanced After Effects users hardly Effects and are always right under
know exist can be found, such as the View Options for a your cursor. This Layer context menu
contains everything you’d want from
Composition viewer. the Layer menu, plus extra Timeline-
. Context-click an item in the Project panel to, among specific commands.
other things, reveal it in the Finder or Explorer.
Having these kinds of options right under cursor keeps you
focused.

Missing Footage
After Effects will link to any source footage file that can be
found on your system or network, source which can easily
become unlinked if anything moves or changes. To relink
When you context-click a Project
an item, find it in the Project panel and double-click it (or item and choose Replace Footage,
Ctrl+H/Cmd+H), or context-click and choose Replace you get a couple of extra options—
Footage > File. Placeholder and Solid—in case
the footage won’t be available
This is also a surreptitious way to replace a source file until later.
without any fuss or bother. If instead you need only to
reload or update a source, context-click and choose

11


Reload Footage (Ctrl+Alt+L/Cmd+Option+L). You can
even edit a file in its source application and update it auto-
matically in After Effects with Edit > Edit Original (Ctrl+E/
Cmd+E), as long as you don’t try anything tricky like saving
it as a new file.
Sometimes it’s difficult to locate a missing file or frame in
your project. You may have used the Find Missing Footage
checkbox in previous versions, and you may wonder where
it has gone. You’re not alone.
To search for particular types of footage in your project,
including missing source, use search (Ctrl+F/Cmd+F)
Figure 1.7 Missing footage is replaced
with color bars, both in the Project in the Project panel and the following commands
thumbnail and anywhere the footage (Figure 1.7):
appears in the project. New in After
Effects CS4, you can reveal all missing . missing is the replacement for the Find Missing Foot-
files in a Project by typing the word age checkbox
“missing” in the Project search field,
highlighted in yellow. . unused gets you all of the source that isn’t in any comp
. used
. Text strings that appear in the Project panel (say, tif
or Aug 26)
Make sure to check out that last one; it’s a totally new
option in After Effects CS4. The date column in the Proj-
ect panel may be hidden by default; context-click to reveal
it, then type in yesterday’s date using a three-letter month
abbreviation; the Project panel now displays only the items
that were introduced or updated yesterday.
Unrecognized file formats are grayed out in the After
Effect Import dialog. Typically, adding a missing three-
character extension solves this, although some obscure
formats simply do not work cross-platform (for example,
Mac-generated PICT on Windows); see the “Source For-
mats” section later in this chapter for the most useful and
universal file types to use.
Because every project is likely to be moved or archived at
some point (you are making backups, right?), it’s best to
keep all source material in one master folder; this helps
After Effects automatically relink all of the related files it
finds there at once, thus avoiding a lot of tedium for you.

12


Move, Combine, and Consolidate Projects
At some point you probably will want to know how to
. Move an entire After Effects project, including its
source, or archive it
. Merge or combine two projects
. Clean up a project, getting rid of what’s not used or
extra instances of a single file
To move or archive a project with only its linked source,
choose File > Collect Files. This command was designed
to enable multimachine Watch Folder rendering (see
Chapter 4) but is also useful for backup, as it allows you to
create a new folder that contains a copy of the project and
all of its source files. The source files are reorganized with
a directory structure that directly replicates the one in the
Project panel (Figure 1.8).

Figure 1.8 Collect Files resaves all source files from your project using the same
organization and hierarchy as the Project itself.

13


Let the computer do what it does best and automate a
clean up of your source. Choose Collect Source Files: For
Selected Comps; After Effects collects only the footage
needed to create that comp. If you check Reduce Project as
well, the unused source is also removed from the collected
project.
File > Reduce Project removes excess items from the proj-
ect itself. Select the master compositions in your project
and choose File > Reduce Project; After Effects eliminates
project items not used in the selected comps. You even
get a warning dialog telling you how many items were
removed—not from the disk, only from your project.
You can instead clean out only the source footage (but
keep the comps and solids) with File > Remove Unused
Footage, which deletes from the project any footage that
Gridiron Flow, which is in late beta
as of the writing of this chapter, is a
hasn’t made its way into a comp. If the same clips have
third-party application that tracks been imported more than once, File > Consolidate All
your entire image pipeline, inside Footage looks for the extra instances and combines them,
and outside After Effects, and maps choosing the ﬁrst instance, top to bottom, in the project.
what happens to each file used in
a project. Need to combine two or more projects? Import one into
the other (just drag it in), or drag several into a new proj-
ect. The imported project appears in its own folder, and
if the projects being combined are organized using the
same set of subfolders, you can merge them with the script
rd_MergeProjects.jsx, which is included on the book’s disc
(Figure 1.9).

Figure 1.9 Load the highly useful
rd_MergeProjects.jsx script from
the redefinery folder on the book’s
disc into Adobe After Effects CS4 >
Scripts > ScriptUIPanels, and you can
then reveal it at any time from the
bottom of the Window menu. This
script takes nested folders with the
same name as those closer to the root
and merges them, while consolidat-
ing duplicate footage. It’s great for
importing a project and maintaining a
tidy structure.
File > Consolidate All Footage looks for two or more
instances of a source ﬁle and combines them, choosing the

14


first instance, top to bottom, in the project. File > Remove
Unused Footage rids a project of footage not included in
any composition (but the files do remain on your drive).

Advanced Save Options
After Effects projects are saved completely separate from
the elements they contain. They tend to be small, mak-
ing it easier to save a lot of them so that you don’t lose
your work.
File > Increment and Save attaches a version number to
your saved project or increments whatever number is
already there, at the end of the file name before the .aep
Use Increment and Save when you
extension. reach a point where you’re happy
Preferences > Auto-Save fills in the spaces between incre- with a project and ready to move
on to the next step; you can then
mented versions; toggle it on and you’ll never lose more choose File > Revert to get back
than the number of minutes you specify (Save Every 20 there in one step instead of using a
Minutes is the default), and you’ll have whatever number series of undos.
of most recent versions you prefer (Figure 1.10).

Figure 1.10 Auto-Save must be enabled, then at the default settings it creates 5
versions, writing a new one after each 20 minutes that you work (but not if your
system sits idle). When number 5 is reached, 1 is deleted and 6 is written. When
you increment the project, a new set of 5 is written. Here, each project is about
8 MB, which is large for an After Effects project but no problem compared to
source file sizes.

Project, Footage, and Composition Settings
After Effects includes a bunch of settings that you need to
understand or you will waste time and effort fighting them.
These have to do with essentials like how time, color depth,
transparency, pixel aspect, and field data are handled.

15


Figure 1.11 Project Settings are
brought up here in Chapter 1 because
they are essential, but only the biggest
and most complicated section (Color
Settings, in blue) doesn’t get a full
explanation until Chapter 11.

Project Settings
As shown in Figure 1.11, the Project Settings dialog
(Ctrl+Alt+Shift+K/Cmd+Option+Shift+K) contains three
basic sections:
. Display Style determines how time is displayed—pre-
dominantly whether a comp’s frame count is kept in
integers (frames) or in timecode. Broadly, ﬁlm projects
tend to work in frames, broadcast video projects in
timecode. This won’t effect the frame rates of your foot-
age or comps, only how they appear.
. The Color Settings section includes the project-wide
color depth (8, 16, or 32 bits per channel), as well as
color management and blend settings. Chapter 11 cov-
ers this in ample depth.
. Audio Settings affects only how audio is previewed;
lowering the rate can save RAM. I never touch this.
If you’re displaying timecode, you’ll almost never want to
change the default Auto setting unless you’re working with
footage containing more than one frame rate and need to
conform everything to a particular standard.

16


If you’re working with frames, it’s typical to start number-
ing them at 1, although the default is 0. This applies to
imported image sequences, not comps. Numbering in a
comp is determined by the Start Frame number in Compo- Feet+Frames in Display Style refers
sition Settings (Ctrl+K/Cmd+K). to actual reels of physical film, for
those who still remember the 20th
Interpret Footage century.

This book generally eschews the practice of walking
through After Effects menus, but a well-designed UI helps
you think. Decisions about how footage is interpreted are
both vital and somewhat tedious. This makes the Interpret
Footage dialog (Figure 1.12), where you can specify for
any source clip, even more vital as a pre-ﬂight checklist for
source footage:
. Alpha interpretation
. Frame Rate
. Fields and Pulldown
. Pixel Aspect Ratio (under Other Options)
. Color Management (under More Options with certain
ﬁle types and the new Color Management tab)

Figure 1.12 The Interpret Foot-
age dialog is a checklist for getting
footage settings correct before you
ever assemble a comp. Alpha (red)
determines transparency settings,
Frame Rate (yellow) is particularly
essential with an image sequence,
Fields and Pulldown (green) and Pixel
Aspect Ratio (under Other Options,
blue) convert footage optimized for
playback. The Color Management tab
(purple) gets a complete treatment in
Chapter 11.

17


To bring up the Interpret Footage dialog for a given clip,
select it in the Project panel and press Ctrl+Shift+G/Cmd+
Shift+G or context-click and select Interpret Footage >
Main. The Interpret Footage icon in the Project panel is
a new shortcut to open this dialog.

Alpha
To composite effectively you must thoroughly understand
alpha channels. Figure 1.13 shows the most visible symp-
tom of a misinterpreted alpha channel: fringing.

Figure 1.13 It’s not hard to distinguish a properly interpreted (left) from an incor-
rect alpha channel (right). The giveaway is fringing, caused in this case by the
failure to remove the black background color premultiplied into edge pixels. The
left image is unmultiplied, the right is not.

You can easily avoid these types of problems:
. If the alpha channel type is unclear, click Guess in the
mini Interpretation dialog that appears when import-
ing footage with alpha. This often (not always) yields a
After Effects does not guess an
alpha unless you specifically click
correct setting.
Guess; if you merely clear the dialog . Preferences > Import contains a default alpha channel
(Esc) it uses the previous default. preference, which is ﬁne to set on a project with consis-
tent alpha handling. If you are in any doubt about that,
set it to Ask User to avoid forgetting to set it properly.
More information on alpha channels and how they operate
is in Chapter 3.

Frame Rate
I have known experienced artists to make careless errors
with frame rate. Misinterpreted frame rate is typically an

18


issue with image sequences only, because unlike Quick-
Time, the files themselves contain no embedded frame rate.
You can also override the QuickTime frame rate, which is
exactly what After Effects does with footage containing any
You can change the default Frames
sort of pulldown (see next section). Per Second setting for Sequence
The following two statements are both true: Footage under Preferences >
Import. This should be the first
. After Effects is more flexible than just about any video thing you check when you are
application in letting you mix clips with varying frame starting a new project so you don’t
have to continually change it.
rates and letting you change the frame rate of a clip
that’s already in a comp.
. However, After Effects is very precise about how those
timing settings are handled, so it is essential that your
settings themselves be precise. If your true frame rate is
23.976 fps or 29.97 fps, don’t round those to 24 and 30,
or strange things are bound to happen: motion tracks Why Sequences?
that don’t stick, steppy playback, and more. Movie formats, in particular QuickTime (.mov), are
like fast food in a regular diet – bound to cause
The current frame rate and duration as well as other inter-
problems sooner or later when compared with the
pretation information is displayed at the top of the Project healthier image sequence option:
panel when you select a source clip (Figure 1.14). . A bad frame in a rendered sequence can
typically be quickly isolated and replaced; a
bad frame will sink an entire QuickTime movie,
sometimes costing hours.
. It’s easy to swap out a section of an image
sequence precisely, overwriting frames instead
of cutting and pasting footage.
. Still image formats have more predictable color
management settings than QuickTime.

If QuickTime is a burger—convenient but poten-
Figure 1.14 Useful information about any selected item appears atop the
tially bad for the health of your project, causing
Project panel. The carat to the right of the file name reveals specific comps in
bloat and slowness—then image sequences are
which it is used.
a home cooked meal, involving more steps but
offering more control over how they are made and
Fields, Pulldown, and Pixel Aspect Ratio consumed.
One surprise for the novice is that moving images are often
not made up of whole frames containing square pixels It’s not always practical, particularly when making
like stills. A video frame, and in particular one shot for quick edits to video footage, to convert everything
to sequences, which don’t play back so easily on
broadcast, is often interlaced into two fields, and its pixels
your system or in your nonlinear editor. Nonethe-
are stored non-square, for the purpose of faster and more less, sequences are the best choice for larger or
efficient capture and delivery. longer-form projects.

19


Fields combine two frames into one by interlacing them
together, vertically alternating one horizontal line of pixels
from the first with one from the second. The result is half
the image detail but twice the motion detail. Figure 1.15
shows this principle in action.

Figure 1.15 If a perfect ellipse trav-
eled left to right at high speed, the
interlaced result would look like this
on a single frame. This contains two
fields’ worth of motion, alternating on
vertical pixels of a single frame. If you
see something like this in your comp,
it hasn’t been removed on import.

If you’re doing any compositing, transformation, paint/
masking, or distortion —pretty much anything beyond basic
color correction—it’s best to match the Separate Fields set-
ting to that of the footage, causing After Effects to recognize
the interlace frame as two separate frames of video.
Pulldown uses fields to run 24 fps film footage smoothly
at 29.97 fps by repeating one field every five frames. This
creates a pattern that After Effects can accurately guess if
there is sufficient motion in the first few frames of the foot-
age. If not, the backup option (which still works) is trial-
and-error, trying each initial pattern listed under Remove
Pulldown until the field artifacts disappear in a 23.976 fps
comp. There are two basic types of pulldown (3:2 and 24
Pa), each with five potential initial patterns.
Pixel aspect ratio (PAR) is another compromise intended
to maximize image detail while minimizing frame size. The
pixels in the image are displayed nonsquare on the broad-
cast monitor, with extra detail on one axis compensating
for its lack on the other.

20


3:2 pulldown is the traditional for-
mat designed to make footage that
originated at 24 fps play smoothly
at 29.97 fps; telecine conversions
from film to television use this. 24P
Advance Pulldown was introduced
to reduce field artifacts by grouping
24 whole frames with 6 interlaced
frames, which are essentially filler
and can be discarded on removal
(see the diagram in Figure 1.16).

Figure 1.16 Pulldown allows 24 fps footage, the film frame rate, and allows it to
play smoothly at 30 fps; without interleaving it into fields in this manner, the
motion stutters, as it does if you try to go straight from 30 fps (no pulldown) to 24.

Your computer monitor, of course, displays square pixels,
so any clip with a non-square PAR will look odd if displayed
without compensating for the difference. Therefore, After
Effects includes a toggle below the viewer panels to stretch
the footage so that its proportions look correct (Figure 1.17)
although the footage or composition itself isn’t changed.
With some digital formats such as DV, field order and
pixel aspect are standardized and set automatically in After
Effects. With other formats, it’s best to know the correct
field order and pixel aspect as specified by the camera or
software that generated the image.

Figure 1.17 Think all of your problems with Pixel Aspect Ratio are gone with the demise of Standard Definition? Think
again. DVCPRO HD footage with Pixel Aspect Ratio Correction on (left) and off (right) via the toggle (circled in green).

21


Source Formats
After Effects is capable of importing and exporting a wide
array of footage formats, yet only a small subset of these
recur regularly in production. Table 1.1 contains a run-
down of common raster image formats and some advan-
tages and disadvantages of each.
Which formats will you use most? Probably TIFF or DPX
for source, and JPEG (with a Quality setting of 7 or higher)
for temporary storage when file space is at a premium.
One oddity of the PNG format
is that it specifies that an alpha TIFF offers lossless LZW compression, giving it an advan-
channel is saved and interpreted as tage over Adobe Photoshop, especially when you consider
Straight, with no explicit option to
change the default. that TIFF can even store multiple layers, each with its own
transparency. Other formats with lossless compression,
such as TGA, don’t support multiple bit-depths and layers
like TIFF does. PNG is more limited and slower, but the file
sizes are smaller.

TABLE 1.1 Raster Image Formats and Their Advantages
LOSSLESS LOSSY ALPHA OUTPUT
FORMAT BIT DEPTH COMPRESSION COMPRESSION CHANNEL FORMAT
TIFF 8/16/32 bit Y N Y (multiple Y
via layers)
PNG 8/16 Y N Y (Straight only) Y
CIN/DPX 10 N N N Y (Cineon 4.5 or DPX,
see Cineon Settings)
CRW 12 N N N N
EXR (now 16/32 Y N Y Y
native in CS4)
JPG 8 N Y N Y

For film and computer graphics, it is normal to pass
around CIN and DPX files (essentially the same format)
and EXR, designed (and open-sourced) by ILM specifically
to handle HDR renders with multiple channels of data
(and these can be customized to contain useful informa-
After Effects CS4 now includes EXR
tools, which are highlighted in tion such as Z depth and motion data). More on these
Chapter 12. formats is found in Chapters 11 and 12, which also include
information on working with Camera Raw CRW images.

22


Photoshop Files
Although the PSD format does not offer any type of com-
pression, it offers a few unique advantages when used with
After Effects. Speciﬁcally, PSD ﬁles
. Can be imported directly as identical After Effects
compositions; all of the data is preserved and most of
it, including blending modes and even layer styles and
text, remains editable in After Effects. In the Import
File dialog, choose Composition or Composition –
Cropped Layers using the Import Kind pop-up menu
(Figure 1.18).

Figure 1.18 Holy Twisted Terminology,
Batman! Composition Cropped Layers
means “just like in Photoshop.” The
other option, Composition, reframes
everything to the image area—and
crops any pixels that fall outside
frame! Choose the Cropped Layers
option to ensure that each layer has
its own unique Anchor Point—and is
not cropped.

. Can be created from After Effects (File > New > Adobe
Photoshop File or even Layer > New > Adobe Photo-
shop File).
. Can include moving footage. More about why you
might want to work with video in Photoshop (for its
paint tools) is included in Chapter 7.
. Can include Live 3D layers: 3D models with lighting,
material, and surface characteristics created in Photo-
shop that can be manipulated in After Effects. More
about this new feature is found in Chapter 9.
Once your source footage is imported and organized
(Chapter 4), the next step is to place it in a composition.

23


Composition Settings
My advice is to begin with your plate: the main footage,
whether a background shot or a foreground yet to be
keyed. To ensure that composition settings are exactly as
The term “plate” stretches back to
the earliest days of optical compos- they should be with the least effort, try one of the following:
iting (and indeed, of photography . Use a prebuilt project template that includes composi-
itself) and refers to the source
footage, typically the background tions whose settings match the intended output; you
onto which foreground elements can even create and save your own.
are composited. . Create a new composition by dragging the plate foot-
age (often the background plate) to the Create a New
Composition icon . This automatically matches pixel
dimensions, Pixel Aspect Ratio, Frame Rate, and Dura-
tion, all of which are crucial.
Composition Settings also contains an Advanced tab. This
pertains to temporal and spatial settings (Chapter 4) and
motion blur and 3D (Chapter 9).

Previews and View Panels
How exactly does a professional work with footage in After
Effects? I’ve noticed some good habits that experienced
pros tend to share.
A 2K plate is the minimum typical
horizontal film resolution: approxi-
mately 2000 pixels, or more precisely Resolution and Quality
2048 pixels in width. HD video is First, keep in mind that you should almost never work at
1920 pixels horizontal resolution.
Full resolution, but you should almost always leave layers
at Best quality. There are several effective ways to speed
up previews and interactivity without ever setting a layer to
Draft quality, which creates inaccurate previews by round-
ing off crucial values.
In rough order of preference, you can
. Lower viewer Resolution to Half, or in extreme cases,
One very cool new CS4 feature is the Quarter (see Note)
Auto setting under the Resolution
menu in the Comp viewer. This . Set a Region of Interest (ROI) to isolate only the area
downsamples the image so that that needs to be previewed
resolution is never higher than mag-
. Use Shift+RAM Preview to skip frames (default setting
nification; this guarantees that when
you work smaller, you work faster. of 1 skips every second frame—details in “Caching and
Previewing,” later in this chapter)

24


Figure 1.19 Highlighted in yellow are a few easy-to-miss options (from the top):
loop options let you ping-pong footage, Shift+RAM Preview is a secondary pre-
viewing option with unique settings, the default difference is a Skip setting of 1
which previews every other frame but can be raised to any number you like, and
you can preview From Current Time instead of the full Work Area.

Half resolution allows four times as much data to ﬁll a
RAM preview, and Shift+RAM Preview can reduce over-
head further by skipping every nth frame (according to the
Skip setting in the Preview panel). The default setting of 1
plays every other frame (Figure 1.19).
To quickly change the display resolution in the Composi-
tion panel, use the keyboard shortcuts shown in Table 1.2.

TABLE 1.2 Display Resolution/Size Shortcuts
RESOLUTION/SIZE KEYBOARD SHORTCUT
Full Ctrl+J/Cmd+J
Half Ctrl+Shift+J/Cmd+Shift+J
Quarter Ctrl+Shift+Alt+J/
Cmd+Shift+Option+J
Fit in viewer Shift+/
Fit up to 100% Alt+/ / Option+/

Hold down the Spacebar or activate the Hand tool
(H, Spacebar, or Middle Mouse Button) to move your
view of a clip around. To zoom in and out, you can use
. Ctrl+=/Cmd+= and Ctrl+-/Cmd+-
. Zoom tool (Z); press Alt/Option to zoom out
. Comma and period keys
. A mouse with a scroll wheel

25


Figure 1.20 Region of Interest crops the active view region. Want to keep this view? Crop Comp to Region of Interest (in
the Composition menu).

Ever notice yourself focusing only on a particular section
of a huge image? Use the Region of Interest (ROI) tool
(Figure 1.20), to deﬁne a rectangular preview region. Only
the layer data needed to render that area is calculated and
With the cursor over a specific area buffered, lengthening RAM previews.
of the frame, hold the Option/Alt
key as you zoom to keep that point
centered. Responsiveness
Has your After Effects UI slowed to crawl as you work on a
big shot? Here’s a quick triage you can try:
. Deactivate Live Update (Figure 1.21a). On by default,
this toggle enables real-time update in the viewers as
you adjust controls. Deactivate it and updates occur
only when you release the mouse.
. Hold Option/Alt as you make adjustments. With Live
Update on, this prevents views from updating. Deac-
tivate Live Update and the behavior is inverted; the
modiﬁer keys instead enable real-time updates.
. Activate Caps Lock. If you don’t mind working “blind”
for periods of time, the Caps Lock key prevents updates
to any viewer (Figure 1.21b).

26


Figure 1.21 Disabling Live Update (a)
and enabling Caps Lock (b) could be
seen as desperation moves when
interactivity becomes unacceptable,
but the former is rarely necessary (you
can do it temporarily with Option/Alt)
and the latter can actually be a handy
way to do setup as quickly as possible
without worrying about previews.

. Enable OpenGL. Preferences > Previews includes the
Enable OpenGL option, off by default (and unavailable
with older graphics cards). Enable it, and OpenGL-
OpenGL in After Effects can have
Interactive mode in a View panel is accelerated in undesirable side effects; most
certain cases, for example when positioning layers in power users tend to leave it off
3D space. There are two OpenGL options, Interactive most of the time.
and Always On; the former will help you with fast scene
setup, especially in a complicated 3D scene, and the
latter will give you the look of OpenGL at all times as
you work.
In general, the more responsive you can make your user
interface, the better will be the result because you can
make more decisions in a shorter period of time. Just leave
time to double-check the result if you are in the habit of
disabling viewers.

27


Multiprocessing
Multiprocessing, which allows After Effects to use all of the
processor cores on your system, is disabled by default; this
does not mean that After Effects doesn’t use all of your
procs, just that by default it doesn’t work on more than
one frame at a time, and thus it doesn’t maximize usage of
your system.
The Preferences > Memory & Multiprocessing dialog
contains dynamically updated information related to your
particular system; it is designed to help you make maxi-
mum use of processing power without maxing out the sys-
tem for other applications. The decision of how to balance
it is yours. By default, After Effects leaves a given amount
of RAM for other applications, and you can control this
amount in this dialog.
Most users miss the main feature here, the Render Mul-
tiple Frame Simultaneously checkbox, which is disabled by
default. Check this box and After Effects no longer waits
for one frame to render before starting the next; instead
it looks for available memory and CPU to render as many
frames as your system can handle simultaneously.
How much multiprocessing can your system handle?
Once you enable multiprocessing, you can tune how it is
handled:
. Raise the minimum allocation per CPU to help back-
ground processes: 1.0 GB for HD footage below 32 bpc,
2.0 GB for larger than HD and 32 bpc projects (Chap-
ter 11 tells you more about those).
. Weight memory more toward RAM Previews (a fore-
ground process, so you wait for it) or faster background
renders (while you continue to work).
. Spare whole CPUs for other applications.
The controls and descriptive text in this dialog, along with
the features they describe have been enhanced for CS4
(Figure 1.22).

28


Figure 1.22 Whoa—Memory and
Multiprocessing settings are not
simple, although they are thoroughly
described in this dynamic Preference.

Caching and Previewing
After Effects automatically caches footage as you navigate
from frame to frame (Page Up/Page Down) or load a RAM
preview (0 on the numeric keypad). The green line atop
the Timeline shows which frames are stored for instant
playback.
Given that After Effects tops out at a bit less than 3 GB of
physical memory per session (due to limitations in system
memory allocation), you can do better. To extend the
Figure 1.23 Enable Disk Cache and
cache from physical memory (RAM) to physical media
you may see your previews extended;
(ideally a high-speed local drive), enable Disk Cache in the blue areas of the Timeline have
Preferences > Memory & Cache. This locks away a por- been cached to disc in addition to
tion of your drive for use only by After Effects. A blue line the green areas cached into physical
memory (RAM).
shows frames loaded in the Disk Cache (Figure 1.23).
When you activate Enable Disk Cache, you must also
specify a disk location; if in doubt just create a local folder
with an intuitive name such as AE Scratch. Even the default
2 GB (2000 MB) setting greatly extends available cache
without occupying permanent disk space.

29


Disk Cache saves the time required to re-render a frame
but doesn’t necessarily deliver real-time playback and often
is not invoked when you might think it should be. The
cache is not saved between After Effects sessions.
The shortcut for Shift+RAM
Preview is, naturally enough, If reﬁned motion is not critical, use Shift+RAM Preview.
Shift+0 (on the numeric keypad).
To set the Work Area to the length Preview Settings
of any highlighted layers, use
Ctrl+Alt+B/Cmd+Option+B. Here’s some cool stuff you do to customize a RAM Preview:
. Loop options (Preview panel). Hidden among the
playback icons atop Preview is a toggle controlling how
previews loop. Use this to disable looping, or amaze
your friends with the ping-pong option.
. From Current Time (Preview panel). Tired of resetting
the work area? Toggle this on and previews begin at the
current time and roll through to the end of the comp.
. Full Screen (Preview panel). Self-explanatory and rarely
used.
. Preferences > Video Preview lets you specify Output
Device and how it is used. If you have an external video
device attached with its own monitor, you can use it to
To update an external preview
device, press /.
preview. Third-party output devices, such as Kona and
Blackmagic cards, are supported as well.

Backgrounds
You need to see what you’re doing, and when you use a
contrasting background it is like shining a light behind
layer edges. You can customize the background color of the
Composition viewer (Ctrl+Shift+B/Cmd+Shift+B or Com-
position > Background Color) or toggle the Transparency
Grid icon beneath the Composition panel to evaluate edges
in sharp relief.
I even insert background or reference footage or a custom
gradient background that I create (Figure 1.24). If it’s set
as a Guide Layer (Layer > Guide Layer or context-click
To create a basic gradient back-
ground, apply the Ramp effect to a
the layer), it does not show up when rendered or nested in
solid layer. another comp.

30


Figure 1.24 If the gradient behind a matted object is made a guide layer, you can clearly see the edge details of the fore-
ground, but the gradient doesn’t show up in any subsequent comps or renders.

Several other modes and toggles are available in the viewer
panels. Some are familiar from other Adobe applications:
. Title/Action Safe overlays determine the boundaries of
the frame as well as its center point.
. View > Show Grid (Ctrl+”/Cmd+”) displays an overlay Use Preferences > Grids & Guides
grid. to customize the Safe Margins in
. View > Show Rulers (Ctrl+R/Cmd+R) displays not only the Title/Action Safe overlay or the
appearance of grids and guides.
pixel measurements of the viewer, but allows you to add
guides as you can in Photoshop.
All of these are toggled via a single menu beneath the
viewer panel (the one that looks like a crosshair). To pull
out a guide, choose Show Rulers and then drag from either

31


the horizontal or vertical ruler. To change the origin point
(0 on each ruler), drag the crosshair from the corner
between the two rulers.
Masks, keyframes, and motion paths can get in the way.
You can
. Hide them all using View > Hide Layer Controls
(Ctrl+Shift+H/Cmd+Shift+H)
. Use the Toggle Mask and Shape Path Visibility button
at the bottom of the Comp panel
. Customize what is shown and hidden with View > View
Options (Ctrl+Alt+U/Cmd+Option+U)
Beginning in Chapter 5 you’ll be encouraged to study
images one color channel at a time. The Show Chan-
nel icon exists for this purpose (keyboard shortcuts
Alt/Option+1 through Alt/Option+4 map to R, G, B,
and A, respectively). An outline in the color of the
selected channel reminds you which channel is displayed
(Figure 1.25).

Figure 1.25 The green border indicates that only the green channel is displayed. (Image courtesy of Mark
Decena, Kontent Films.)

32


Effects & Presets
After Effects contains about 200 effects plug-ins that ship
with the application, and far more than that from third
parties. Personally, I use less than 10% of these effects
around 80–90% of the time, so my opinion is that you
don’t need to understand them all in order to use the most
powerful ones.
And even cooler, once you thoroughly understand the core
effects, you can use them together to do things with After
Effects that you might have thought required third-party
plug-ins.
To apply an effect to a layer, my advice is to avoid the
Effect menu and either context-click that layer, then use
the Effect context menu, or double-click it in the Effects &
Presets panel.
The Effects & Presets panel is a versatile tool when problem-
solving. It has options to display effects alphabetically,
without their categories, as well as a search field to help
you look for a specific effect by name, or for all the effects
whose names include a specific word, such as “blur” or
“channel” (Figure 1.26).
Animation Presets allow you to save specific configurations
of layer properties and animations, including keyframes,
effects, and expressions. They’re particularly useful when Figure 1.26 Type the word “blur” in
working with others and sharing standardized practices. the Effects & Presets search field and
Save your own by selecting the effects and properties you only effects with that text string in the
name appear. You can also choose
want to save and choose Animation > Save Animation
to display only effects with higher
Preset; save it to the Presets folder (the default location) to bit depths (when working at 16 or
have it show up when After Effects is started. 32 bpc).

Output via the Render Queue
As you well know, the way to get completed footage out
of After Effects is to render it. Here are a few things you
might not already know about the process of outputting Convert raw footage by dragging
it directly to the Render Queue, no
your work.
comp required (one is made for
To place an item in the Render Queue, it’s simplest either you). This is a quick and easy way
to convert an image sequence to a
to use a shortcut (Ctrl+M/Cmd+M) or to drag items from QuickTime movie, or vice versa.
the Project panel.

33


There are two key sections for each Render Queue item:
Render Settings and Output Module.

Render Settings: Match or Override the Comp
Render Settings breaks down to three basic sections
(Figure 1.27):

Need to render several items to . Composition corresponds directly to settings in the
one location? Set up one item, and Timeline; here you choose whether to keep or override
only thereafter, add the rest. The them. The more complex options, such as Proxy Use,
location of the first becomes the
default. are described in Chapter 4.
. Time Sampling gives you control over the timing of the
render; not just frame rate and duration but the ability
to add pulldown and fields—say when rendering a 24
fps film comp for 29.97 video—as well as motion blur
and frame blending (Chapter 2).
. Options contains one super important feature: “Skip
existing files,” which checks for the existence of a
file before rendering it. This is useful for splitting
sequences between sessions (see Chapter 4 for details
on how to use it).

Figure 1.27 Maybe it’s helpful to
see the sections of the Render Set-
tings dialog highlighted; the real
point is that the first section (in red)
looks at how an individual frame is
rendered, while the second (blue)
section focuses on timing the whole
sequence.

If you find that rendered output doesn’t match your expec-
tations, Render Settings is generally the place to look (unless
it involves color management, compression, or audio). The
Output Modules handle writing that output to a file.

34


Figure 1.28 It’s easy to miss that you can add multiple Output Modules to a single render queue item via Composition >
Add Output Module or this context menu. This is an immense timesaver, as each frame is rendered once and written as
many times as you like.

Figure 1.29 Custom-number a frame
sequence here; no convoluted work-
arounds needed.

Output Modules: Making Movies
Output Modules convert the rendered frame into an actual
ﬁle. The main decisions here concern
. Format Want the best looking half-
. Size resolution render? Use a Stretch
setting in the Output Module,
. Audio instead of half resolution in
Render Settings (which typically
. Color management
renders faster).
Several elegant and easily missed problem-solving tools are
embedded in Output Modules:
. Multiple Output Modules per Render Queue item
avoid the need for multiple passes (Figure 1.28).
. Separate Output Modules can be edited together by
Shift-selecting the modules themselves (not the Render
Queue items that contain them).
. A numbered sequence can start with any number you
like (Figure 1.29).
. Scaling can be nonuniform to change the pixel aspect
ratio.

35


. Post-Render Actions automate bringing the result back
into After Effects. Chapter 4 tells all.
. A numbered image sequence must contain a string in
Naming Conventions the format [###] somewhere within its name. Each #
Part of growing a studio is devising a naming sign corresponds to a digit, for padding.
scheme that keeps projects and renders organized. . The Color Management tab is in effect with many still
It’s generally considered good form to image formats. Chapter 11 tells all.
. Use standard Unix naming conventions (replac-
ing spaces with underscores, intercaps, dashes, . Rendered files can include XMP metadata (if toggled
or dots). on, as by default); this includes information that the
. Put the version number at the end of the proj- file came from After Effects.
ect name and the output file, and have them Save Output Modules early and often using the Make
match. To add a version number to a numbered
sequence, you can name the image sequence Template option at the bottom of the pop-up menu. If you
file something like foo_bar_[####]_v01.tif intend to render with the same settings even once more,
for version 1. it saves time. Unfortunately these cannot be easily sent to
. Pad sequential numbers (adding zeros at the another user.
beginning) to keep things in order as the
overall number moves into multiple digits. Optimized Output
Following are some suggested output settings (Render Set-
And remember After Effects itself doesn’t like file tings and Output Modules) for specific situations:
names above 32 digits, so find a system that is
concise or risk having key information, toward the . Final output should match the delivery format; it’s
end of the name, truncated in the Project panel. usually an editor who decides this. Merely choosing the
default Lossless setting is not sufficient if, for example,
you’ve been working in 16 bpc to render a 10-bit final
(Lossless is only 8 bit). For sending files internally, TIFF
with lossless LZW compression is solid and can handle
Photo-JPEG is universally available, higher bit depths and color management.
even in older versions of QuickTime. . Low-loss output is mostly up to you; QuickTime with
Plus, at 100%, it provides 4:4:4
chroma sampling, and at 75%, Photo-JPEG at around 75% is an old standby, but all
4:2:2 (see Chapters 6 and 11 for QuickTime formats may display inconsistent gamma,
more on chroma). depending on version and platform.
. Online review typically should be compressed outside
of After Effects; such aggressive compression formats as
H.264 are most successful on multiple passes.
After Effects offers a number of output formats and can
Chapter 4 tells more about how to be useful for simple file conversion; you need only import
send your project to Adobe Media
source and drag it directly to the Render Queue, then add
Encoder for multipass encoding;
this requires Adobe CS4 Production settings and press Render.
Premium.

36


Study a Shot like an Effects Artist
Seasoned visual effects supervisors miss nothing. Fully
trained eyes do not even require two takes, although in the
highest end facilities, a shot loops for several minutes while The Trials and Tribulations of QuickTime
the team picks it apart. QuickTime is the most ubiquitous and universal
playback format among video professionals,
This process, though occasionally hard on the ego, makes despite that it is proprietary. There are design deci-
shots look good. sions behind QuickTime that don’t change unless
Apple decides to change them.
You can and should scrutinize your shot just as carefully in
After Effects. Speciﬁcally, throughout this book I encour-
Some of these amount to a gotcha:
age you to get in the following habits:
. Color management of QuickTime remains (at
. Keep an eye on the Info palette (Figure 1.30). this writing) a moving target, with MOV files
appearing differently when they are moved
from one platform, software, or even monitor,
to another. When I say “software” I even mean
Apple applications, which don’t always seem
to agree on how to display the format.
. High Quality toggle in QuickTime Player is
unchecked by default. The client is screaming
about why the shot looks fuzzy and you have
to direct them to Window > Show Movie
Figure 1.30 Whoops! The swatch (from the pixel under the cursor) looks Properties > Video Track > Visual Settings and
black, but a glance at the Info palette shows that it is actually 2% gray. If the little toggle to the lower right.
you’re uncertain about pixel values, have a glance at the Info panel. . Many useful formats are Mac-only, such as
Uncompressed 10-bit 4:2:2, an Apple format.
. Loop or rock and roll previews. . The notorious QuickTime 7.2 caused After
. Zoom in to the pixel level, especially around edges. Effects renders to fail with a write-error.
. Failed or corrupted QuickTime files are typically
. Examine footage channel by channel (Chapter 5).
unplayable, and the larger they are, the more
. Turn the Exposure control in the comp viewer up susceptible they seem to be to being taken
and down to make sure everything still matches down by a spurious frame or two.
(Chapter 5).
. Assume there’s a ﬂaw in your shot; it’s the only way Most larger facilities rely on QuickTime only for
around getting too attached to your intentions. client delivery and in the edit suite. It’s incon-
venient to deal with sequences when there’s no
. Approach your project like a computer programmer free standalone player for them of the quality of
and minimize the possibility of bugs (careless errors). FrameCycler, but sequences cause fewer of these
Aspire to design in modules that anticipate what might types of problems in exchange for the overall
inconvenience.
change or be tweaked.

37

The right word may be effective, but no word was ever
as effective as a rightly timed pause.
—Mark Twain

The Timeline

T he Timeline can be considered After Effects’ killer
application. More than any other feature, it extends After
Effects unique versatility to a wide range of work. With the
Timeline at the center of the compositing process, you
can time elements and animations precisely as you control
their appearance.
The Timeline panel is also a user-friendly part of the appli-
cation and full of hidden powers. By mastering its usage,
you can streamline your workﬂow a great deal, setting the
stage for more advanced work.
One major source of these hidden powers is the Timeline’s
set of keyboard shortcuts and context menus. These are
not extras to be investigated once you’re a veteran, but
small productivity enhancers that you can learn gradually
all the time.

Organization
The goal here isn’t to keep you organized but to get rid of
everything you don’t need and put what you do need right
at your ﬁngertips.

Column Views
You can context-click on any column heading to see and
toggle available columns in the Timeline, or you can start
with the minimal setup shown in Figure 2.1 and then aug-
ment or change the setup with the following tools:

40


Figure 2.1 This very most basic Timeline setup is close to optimal, especially if space is tight; it leaves everything you need
within a single click, such as Toggle Switches/Modes. No matter how big a monitor, every artist tends to want more space
for the keyframes and layers themselves.

. Lower-left icons : Most (but not quite all) of
the extra data you need is available via the three toggles
found at the lower left of the Timeline.
. Layer Switches and Transfer Controls are the
most-used; if you have plenty of horizontal space, leave
them both on, but the F4 key has toggled them since
the days when 1280 × 960 was an artist-sized display.
. Time Stretch toggles the space-hogging timing col-
umns. I’m unconvinced they’re needed; the one thing I
do with this huge set of controls is stretch time to either
double speed or half speed (50% or 200% stretch,
respectively), which I can do by context-clicking Time >
Time Stretch.
. Layer/Source (Alt/Option key toggles): What’s in a
name? Nothing until you customize it; clear labels and
color (below) boost your workﬂow.
To rename an item in After Effects,
. Parent: This one is so often on when you don’t need it highlight it and press Return
and hidden when you do (see “Spatial Offsets” later in instead of clicking and hovering (as
this chapter); use context-clicking to show/hide it. in the typical operating system).

. I can’t see why you would disable AV Features/Keys; it
takes effectively no space.
The game is to preserve horizontal space for keyframe data
by keeping only the relevant controls visible.

41

Chapter 2 The Timeline

Color Commentary
When dissecting something tricky, it can help to use
. Solo layers to see what’s what
. Locks for layers that should not be edited further
To change the visibility (rather than . Shy layers to reduce the Timeline to only what’s
the solo state) of selected layers, needed
choose Layer > Switches > Hide
Other Video. . Color-coded layers and project items
. Tags in the comments ﬁeld
Solo layers hide visible layers that are not solo’d. They are
great except that other layers can stay mysteriously hidden
in subsequent comps and renders. For this reason, I like to
set Solo Switches to All Off in my default Render Settings.
I don’t often lock layers, but it makes a heck of a lot of
To unlock a number of layers at sense particularly for layers that you don’t want “nudged”
once, choose Layer > Switches > out of position, such as adjustment layers, track mattes,
Unlock All Layers (Ctrl+Shift+L/
and background solids (but once they’re locked, you can’t
Cmd+Shift+L).
adjust anything until you unlock them). If you’re a super-
organized person, you can use layer locks to check layers in
and out; the locked ones are completed—for now.
Employing shy layers provides a fantastic workaround
for the cluttered Timeline. Layers set to shy disappear
from the Timeline (once the Timeline’s own Shy toggle is
Shy layers can greatly reduce clutter
in the Timeline, but if they ever trick enabled) but remain visible in the Composition viewer. If
you, study the Index numbers; if you’ve ever worked with an imported 3D track composition
any are missing from the sequence, from such software as SynthEyes or Boujou, you know that
that’s a shy layer. they typically come with hundreds of null layers. I tend to
make these shy immediately, leaving only the camera and
background plate ready for compositing.
Colors are automagically assigned to speciﬁc types of layers
(like Cameras, Lights, and Adjustment Layers) according
to Preferences > Label Defaults. I often give unique colors
to track matte layers so I remember not to move them. On
someone else’s system, the colors may change according to
their preferences, although they will correspond.
Comments are generally the least used column in the
Timeline, but that could change if more people start using
a script called “Zorro-The Layer Tagger,” which manages

42


Figure 2.2 Comments have become much more useful with the new Timeline search field in CS4, and the third-party script
shown here, “Zorro-The Layer Tagger” (from aescripts.com), makes them even more useful. Zorro not only holds a list of
keywords to apply to or remove from selected layers, it even does nifty stuff such as make layers with a given tag instantly
solo or shy.

the process of adding tags to layers and using them to cre-
ate selection sets (Figure 2.2).
Layer Markers (and, new in CS4, Comp Markers) can hold
visible comments. You can add a layer marker for a given
point in time with the * key on your numeric keypad,
meaning you can add them while looping up a RAM Pre-
view in real time. Comp Markers are added using Shift and
the numbers atop your keyboard. I double-click them and
make notes, both for timing and when I want to make sure
a note won’t get lost in the hidden Comment column.

Navigation and Shortcuts
Keyboard shortcuts make a huge difference in your ability
to work speedily and effortlessly in the Timeline.

Time Navigation
Many users—particularly editors—learn time navigation
shortcuts right away. Others primarily drag the current
time indicator, which can quickly become tedious. See if
there are any here you don’t already know:
. Home, End, Page Up, and Page Down correspond Laptop users in particular may
to moving to the ﬁrst or last frame of the composi- prefer Ctrl/Cmd+Left and Right
Arrow as an alternative to Page
tion, one frame backward, and one frame forward,
Up and Page Down.
respectively.

43


. Shift+Page Up and Shift+Page Down skip ten frames
backward or forward, respectively.
. Shift+Home and Shift+End navigate to the work area
In and Out points respectively, and the B and N keys set
these points at the current time.
. I and O keys navigate to the beginning and end frames
of the layer.
. Press Alt+Shift+J/Opt+Shift+J or click on the current
time status at the upper left of the Timeline to navigate
to a specific frame or timecode number. In this dialog,
Don’t bother with punctuation when
entering time values into a number enter +47 to increment 47 frames or –47 to decrement
field in After Effects. 1000 is ten the same number; if you entered –47, that would navi-
seconds (10:00) in Timecode mode. gate to a negative time position instead of offsetting by
that number.

Make Layers Behave
We were reviewing film-outs of shots in progress from The
Day After Tomorrow at The Orphanage when my shot began
to loop; it looked out a window at stragglers making their
The increment/decrement method
operates in most number fields
way across a snow-covered plaza and featured a beauti-
throughout After Effects (including ful matte painting by Mike Pangrazio. About two-thirds
Composition Settings). through the shot came a subtle pop. At some point, the
shot had been lengthened, and a layer of noise and dirt I
had included at approximately 3% transparency (for the
window itself) had remained shorter in a subcomposition.
With power comes responsibility. After Effects allows you to
time the entrance and exit of layers in a way that would be
excruciating in other compositing applications that have
no built-in sense of a layer start or end. To avoid the gotcha
where a layer or comp comes up short, it is wise to make
everything way longer than you ever expect you’ll need—
overengineer in subcomps and trim in the master comp.
To add a layer beginning at a specific time, drag the ele-
ment to the layer area; a second time indicator appears
that moves with your cursor (horizontally). This deter-
mines the layer’s start frame. If other layers are present
and visible, you can also place it in layer order by dragging
it between them.

44


Here are some other useful tips and shortcuts:
. Ctrl+/ (Cmd+/) adds a layer to the active composition.
. Ctrl+Alt+/ (Cmd+Opt+/) replaces the selected layer in
a composition (as does option dragging one element The keyboard shortcut Ctrl+/
over another—either in the Timeline or the Project (Cmd+/) adds selected items
panel—this can be hugely useful). as the top layer(s) of the active
composition.
. J and K navigate to the previous or next visible
keyframe, layer marker, or work area start/end,
respectively.
. Ctrl+Alt+B/Cmd+Option+B sets the Work Area to the
length of any selected layers. To reset the Work Area to
the length of the composition, double-click it.
To trim a composition’s Duration
. Numeric keypad numbers select layers with that to the current Work Area, choose
number. Composition > Trim Comp to Work
Area.
. Ctrl+Up Arrow (Cmd+Up Arrow) selects the next layer
up; down works the same.
. Ctrl+] (Cmd+]) and Ctrl+[ (Cmd+[) move a layer
up or down one level in the stack. Ctrl+Shift+] and
Ctrl+Shift+[ move a layer to the top or bottom of the
stack.
. Context-click > Invert Selection to invert the layers cur-
rently selected. (Locked layers are not selected, but shy
layers are selected even if invisible.)
. Ctrl+D (Cmd+D) to duplicate any layer (or virtually any
selected item).
. Ctrl+Shift+D (Cmd+Shift+D) splits a layer; the source
ends and the duplicate continues from the current
time.
For those who care, a preference
. [ and ] move the In or Out points of selected layers to controls whether split layers are cre-
the current time. Add Alt (Option) to set the current ated above or below the source layer
frame as the In or Out point, trimming the layer. (Preferences > General > Create
Split Layers Above Original Layer).
. Double-ended arrow icon over the end of a trimmed
layer lets you slide it, preserving the In and Out points
while translating the timing and layer markers (but not
keyframes).
. Alt+PgUp/PgDn (Option+PgUp/PgDn) nudges a layer
and its keyframes forward or backward in time. Alt/
Option+Home or End moves the layer’s In point to the
beginning of the comp, or the Out point to the end.

45


Layer > Transform (or context-click a layer > Transform)
includes three ways to fill a frame with the selected layer:
. Ctrl+Alt+F (Cmd+Option+F) centers a layer and fits
It can be annoying that the Work both horizontal and vertical dimensions of the layer,
Area controls both preview and whether or not this is non-uniform scaling.
render frame ranges because the
two are often used exclusively
. Ctrl+Alt+Shift+H/Cmd+Option+Shift+H centers but
of one another. By employing a fits only the width.
separate “Render Final” composi- . Ctrl+Alt+Shift+G/Cmd+Option+Shift+G centers but
tion that contains the Work Area
for the final render, you can mess
fits only the height.
with your other composition Work Those shortcuts are a handful; context-clicking the layer
Areas as much as you want without for the Transform menu is nearly as easy.
affecting the render range.
Timeline Views
After Effects has a great keyframe workflow. These short-
cuts will help you work with timing more quickly, accu-
rately, and confidently:
. The ; key toggles all the way in and out on the Time-
line: single frame to all frames. The slider at the bottom
of the Timeline zooms in and out
more selectively.
. The scroll wheel moves you up and down the layer
stack.

Hold down the Shift key as you . Shift+scroll-wheel moves left and right in a zoomed
drag the Current Time Indicator to Timeline view.
snap current time to comp or layer . Alt/Option+scroll zooms dynamically in and out of
markers or visible keyframes.
the Timeline, remaining focused around the cursor
location.
. The key toggles between a Timeline and its Composi-
tion viewer, even if previously closed.
. The Comp Marker Bin contains markers you can
draw out into the Timeline ruler. You can replace their
sequential numbers with names.

Keyframes and the Graph Editor
Transform controls live under every layer’s twirly arrow.
There are keyboard shortcuts to each Transform property.
For a standard 2D layer these are

46


. A for Anchor Point, the center pivot of the layer
. P for Position, by default the center of the comp
. S for Scale (in percent of source)
. R for Rotation (in revolutions and degrees)
. T for Opacity, or if it helps, Opaci-“T” (which is not
technically spatial transform data but is grouped here
anyhow because it’s so essential)
Once you’ve revealed one of these, hold down the Shift
key to toggle another (or to hide another one already dis-
played). This keeps only what you need in front of you. A
3D layer reveals four individual properties under Rotation
to allow full animation on all axes.
The Alt/Option key can be used with each of these one-
letter shortcuts to add the ﬁrst keyframe; once there’s one
keyframe, any adjustments to that property at any other
frame generate another keyframe automatically.
There are selection tools to correspond to Transforms:
. V activates the Selection tool, which also moves and
scales in a view panel.
. Y switches to the Pan-Behind tool, which moves the
anchor point.
. W stands for “wotate”—it adjusts Rotation.
Once you adjust with any of these tools, an Add Keyframe
option for the corresponding Property appears under the
Animation menu, so you can set the ﬁrst keyframe without
touching the Timeline at all.

Graph Editor
To demonstrate how best to use the Graph Editor, I devised
the project 02_graphEditor.aep in the accompanying
disc’s Chapter02 directory. It contains a simple animation,
“bouncing ball 2d,” which you can create from scratch;
you can also see the steps below as individual numbered
compositions.
To enable the Graph Editor, click its icon in the Time-
line or use the shortcut Shift+F3. Below the grid that
appears in place of the layer stack are the Graph Editor
controls (Figure 2.3).

47


Figure 2.3 The Graph Editor is enabled in the Timeline instead of default Layer view. There is no option to see both
together.

Show Properties
By default, if nothing is selected, nothing displays; what
you see depends on the settings in the Show Properties
menu . Three toggles in this menu control how anima-
tion curves are displayed:
. Show Selected Properties displays animation whatever
property names are highlighted.

To work in the Graph Editor without . Show Animated Properties shows everything with key-
worrying what is selected, disable frames or expressions.
Show Selected Properties and . Show Graph Editor Set displays properties with the
enable the other two.
Graph Editor Set toggle enabled.
Show Selected Properties is the easiest to use, but Show
Graph Editor Set gives you the greatest control. You decide
which curves need to appear, activate their Graph Editor
Set toggle, and after that it no longer matters whether you
keep them selected.
To begin my bouncing ball animation, I include Position in
the Graph Editor Set by toggling its icon . Alt/Option+P
sets the ﬁrst Position keyframe at frame 0; after that, any
The other recommended change if
you are following this section is to
changes to Position are automatically keyframed.
enable Default Spatial Interpolation
to Linear in Preferences > General Basic Animation and the Graph View
(Ctrl/Command+Alt/Option+;).
Figure 2.4 shows the ﬁrst step: a very basic animation
You have to restart After Effects
before this preference is enabled. blocked in using Linear keyframes, evenly spaced. It won’t
look like a bouncing ball yet, but it’s a typical way to start
when animating, for new and experienced animators alike.

48


Figure 2.4 It’s like a bouncing ball in that the layer travels across the frame, going up and down.

To get to this point
. Having set the ﬁrst keyframe at frame 0, move the ball
off left of frame.
. At frame 24, move the ball off right of frame, creating a
second keyframe.
. Now add the bounces: At frame 6 and 18 move the ball
straight downward so it touches the bottom of frame.
This leaves ﬁve Position keyframes and an extremely
unconvincing looking bouncing ball animation. Great—it

49


always helps to get something blocked in so you can clearly
see what’s wrong. Also, the default Graph Editor view at
this point is not very helpful, because it displays the Speed
Graph, and the speed of the layer is completely steady at
this point—deliberately so, in fact. That’s why keyframes
were created in that order (although in a moment you’ll
see how to avoid having to do it that way with a new
feature).
To get the view shown in Figure 2.4, make sure Show Refer-
ence Graph is enabled in the Graph Options menu .
This is a toggle even advanced users miss, although it is now
Auto Select Graph Type selects
Speed graphs for spatial properties on by default; in addition to the not-very-helpful Speed
and Value graphs for all others. Graph you now see the Value Graph in its X (red) and Y
(green) values. However, the green values appear upside-
down! This is the flipped After Effects Y axis in action; 0 is
at the top of frame so that 0,0 is in the top left corner, as it
has been since After Effects 1.0, long before 3D animation
was added.

Ease Curves
The simplest way to “fix” an animation that looks too stiff
like this is often to add eases. For this purpose After Effects
offers the automated Easy Ease functions, although you
can also create or adjust eases by hand in the Graph Editor.
Select all of the “up” keyframes—the first, third, and
fifth—and click Easy Ease (F9). When a ball bounces, it
slows at the top of each arc, and Easy Ease adds that arc to
Mac users beware: The F9 key is the pace; what was a flat line Speed Graph now is a series
used by the system for the Exposé
feature, revealing all open panels of arcing curves (Figure 2.5).
in all applications. You can change
Technically, you could have applied Easy Ease Out
or disable this feature in System
Preferences > Dashboard & Exposé. (Ctrl/Command+Shift+F9) to the first keyframe and Easy
Ease In (Shift+F9) to the final one, because the ease in
each case only goes in one direction. The “in” and “out”
versions of Easy Easy are specifically for cases where there
are other adjacent keyframes and the ease should only go
in one direction (you’ll see one in a moment). In this case
it’s not really necessary.

50


Figure 2.5 Easy Ease is applied (top) to the mid-air keyframes; Layer view (bottom) also shows the
change from linear to Bézier with a changed keyframe icon.

Meanwhile, there’s a clear problem here: The timing of
the motion arcs, but not the motion itself, is still com-
pletely linear. Fix this in the Composition viewer by pulling
Beziers out of each of the keyframes you just eased:
1. Deselect all keyframes but leave the layer selected.

2. Make sure the animation path is displayed (Ctrl/
Command+Shift+H toggles).
3. Click on the first keyframe in the Composition viewer
to select it; it should change from hollow to solid in
appearance.
4. Switches to the Pen tool with the G key; in the Compo-
sition viewer, drag from the highlighted keyframe to the
right, creating a horizontal Bezier handle. Stop before
crossing the second keyframe.
5. Do the same for the third and fifth keyframes (dragging
left for the fifth).

51


Figure 2.6 You can tell just from the graph that this is closer to how a bouncing ball would look
over time. You can use Ctrl/Command+Shift+H to show and hide the animation path, or you can
look in the Comp panel menu > View Options > Layer Controls.

The animation path now looks more like you’d expect a
ball to bounce (Figure 2.6). Preview the animation, how-
ever, and you’ll notice that the ball appears to be riding a
pogo stick across the frame, instead of bouncing naturally.
Why is that?

Separate XYX
The Graph Editor reveals the problem. The red X graph
shows an unsteady horizontal motion due to the eases. The
problem is that the eases should be applied only to the
vertical Y dimension, whereas the X animation travels at a
constant rate.
New to After Effects CS4 is the ability to animate X and Y
(or, in 3D, X, Y, and Z) animation curves separately. This
allows you to add keyframes for one dimension only at a
given point in time, or to add keyframes in one dimension
at a time.
Select Position and click Separate Dimensions . Where
there was a single Position property, there are now two
marked X Position and Y Position. Now try the following:

52


1. Disable the Graph Editor Set toggle for Y Position so
that only the red X Position graph is displayed.
2. Select the middle three X Position keyframes—you can
draw a selection box around them—and delete them.
3. Select the two remaining X keyframes and click the
Convert Selected Keyframes to Linear button .
Now take a look in the Composition viewer—the motion is
back to linear, although the temporal eases remain on the
Y axis. Not only that, but you cannot redraw them as you
did before; enabling Separate Dimensions removes this
ability.
Instead, you can create them in the Graph Editor itself.
1. Enable the Graph Editor Set toggle for Y Position, so
both dimensions are once again displayed.
2. Select the middle Y Position keyframe, and you’ll notice Show Graph Tool Tips displays val-
two small handles protruding to its left and right. Drag ues of whatever curve is under the
each of these out, holding the Shift key if necessary to mouse at that exact point in time.
keep them flat, and notice the corresponding change
in the Composition viewer (Figure 2.7).

Figure 2.7 If Separate Dimensions is activated, pull out the handles to create the motion arcs
right in the Graph Editor; the handles are no longer adjustable in Comp view.

3. Select the first and last Y Position keyframes and click
Easy Ease; the handles move outward from each key-
frame without affecting the X Position keyframes.
4. Drag the handles of the first and last Y Position key-
frames as far as they will go (right up to the succeeding
and preceding keyframes, respectively).

53


Preview the result and you’ll see that you now have the
beginnings of an actual bouncing ball animation; it’s just a
little bit too regular and even, so from here you will give it
your own organic touch.

The Transform Box
The Transform Box lets you edit keyframe values in all
kinds of wacky ways. Toggle on Show Transform Box and
select more than one keyframe, and a white box with
There is a whole menu of options
to show items that you might think vertices surrounds the selected frames. Drag the handle
are only in Layer view: layer In/ at the right side left or right to change overall timing; the
Out points, audio waveforms, layer keyframes remain proportionally arranged.
markers, and expressions.
So, does the Transform Box help in this case? Well, it
could, if you needed to
. Scale the animation timing around a particular key-
frame: drag the anchor to that frame, then Ctrl-drag/
Cmd-drag.
. Reverse the animation: Ctrl-drag/Cmd-drag from one
edge of the box to the other (or for a straight reversal,
simply context-click and choose Keyframe Assistant >
The Snap button snaps to
virtually every visible marker, but Time-Reverse Keyframes).
not—snap!—to whole frame . Add a decay to the animation so that the ball bounces
values if Allow Keyframes Between lower each time: Alt-drag (Option-drag) on the lower
Frames is on .
right corner handle (Figure 2.8).
If you Ctrl+Alt-drag (Cmd+Option-drag) on a corner that
will taper values at one end, and if you Ctrl+Alt+ Shift-drag
(Cmd+Option+Shift-drag) on a corner, it will skew that
end of the box up or down. I don’t do this stuff much, but
with a lot of keyframes to scale proportionally, it’s a good
one to keep in your back pocket.

Figure 2.8 How do you do that? Add the Alt/Option key when dragging a corner
of the Transform Box; this adjustment diminishes the height of the ball bounces
proportionally over time.

54


Holds
At this point you may have a fairly realistic looking bounc-
ing ball; maybe you added a little Rotation animation so
the ball spins forward as it bounces, or maybe you hand-
adjusted the timing or position keys to give them that extra
little organic unevenness.
Hold keyframes won’t help improve this animation, but
you could make the ball disappear each time it hits the
ground, and reappear at the top of the arc, with no eases.
A Hold keyframe (Ctrl+Alt+H/Cmd+Shift+H) prevents
any change to a value until the next keyframe.
Add keyframes alternating between 100% and 0% Opacity
at the same time as the existing Y Position keyframes. If you
leave them as linear keyframes, the ball oscillates between
opaque and transparent, but if you toggle them to Hold
keyframes, it snaps on and off (Figure 2.9).

Figure 2.9 Linear keyframes hold a value across time, until the next keyframe value appears.

Beyond Bouncing Balls
In the (possibly likely) case that the need for a bouncing
ball animation never comes up, what does this example
show you? Let’s recap:
. You can control a motion path in the Composition
viewer using Bézier tools and the Pen tool (usage
detailed in the next chapter).
. Realistic motion often requires that you shape the
motion path Béziers and add temporal eases; the two
actions are performed independently on any given
keyframe, and in two different places (in the viewer
and Timeline).

55


Animation can get a little trickier in 3D, but the same basic
rules apply (see Chapter 9 for more).
Three preset keyframe transition types are available, each
with a shortcut at the bottom of the Graph Editor: Hold ,
Linear , and Automatic Bezier . Adjust the handles
or apply Easy Ease and the preset becomes a custom Bezier
Shape.

Roving Keyframes
Sometimes an animation must follow an exact
path, hitting precise points, but progress steadily,
with no variation in the rate of travel. This is the
situation for which Roving keyframes were devised.
Figure 2.10 shows a before and after view of a
roving keyframe; the path of the animation is iden-
tical, but the keyframes proceed at a steady rate.

Figure 2.10 Compare this graph with
the one in Figure 2.5 (top); the Speed
Graph is back to a flat-line because
the animation runs at a uniform pace.
It’s not helpful to bounce a ball, but it
works with any complex animation,
and it maintains eases on the start and
end frame.

56


Copy and Paste Animations
Yes, copy and paste; everyone knows how to do it. Here are
some things that aren’t necessarily obvious about copying
and pasting keyframe data:
. Got Excel or another spreadsheet application? Copy a
set of keyframes from After Effects and paste them into
a spreadsheet, and the After Effects keyframe format is
You can use an Excel spreadsheet
laid bare. to reformat keyframe data from
. You can paste from one property to another, so long as other applications; just paste in
the format matches (the units and number of param- After Effects data to see how it’s
formatted, and then massage the
eters). Copy the source, highlight the target, and paste. other data to match that format.
. Keyframes respect the position of the current time indi- Once done, copy and paste back
cator; the first frame is always pasted at the current time into After Effects.
(useful for relocating timing, occasionally a surprise).
. There’s a lock on the Effect Controls tab to keep a panel
forward even when you select another layer to paste to it.
. Copy and paste keyframes from an effect that isn’t
applied to the target, and it is added along with the ani-
mation, natch; you don’t get any properties you didn’t
select and copy, however.
Therefore, pay attention to the current time and what is
selected when copying, but in particular, when pasting
animation data.

Layer vs. Graph
To summarize the distinction between Layer and Graph
Editor views, in Layer view you can
. Block in keyframes with respect to the overall
composition
. Establish broad timing (where Linear, Easy Ease, and
Auto-Bezier keyframes are sufficient)
The Graph Editor is essential to
. Refine an individual animation curve
. Compare spatial and temporal data
. Scale animation data, especially around a specific pivot
point

57


. Perform extremely specific timing (adding a keyframe
is between frames, hit a specific tween point with an
ease curve)
With either view you can
. Edit expressions
. Change keyframe type (Linear, Hold, Ease In, and
You must enable Allow Keyframes so on)
Between Frames in the Graph . Make editorial/compositing decisions regarding layers
Editor if you want that feature. (start/stop/duration, split layers, order—possible in
However, when you scale a set
of keyframes using the Transform both, easier in Layer view)
Box, keyframes will often fall in By no means, then, does the Graph Editor make Layer view
between frames whether or not obsolete; it is still where the majority of compositing and
this option is enabled.
simple animation is accomplished.

Über-duper
The following keyboard shortcuts have broad usage when
applied with layers selected in the Timeline:
. U toggles all properties with keyframes or expressions
applied.
. UU (U twice in quick succession) toggles all properties
set to any value besides the default; or every Property in
the Timeline that has been edited.
. E toggles all applied effects.
. EE toggles all applied expressions.
When I use the term “toggle” in the above list, I mean that
not only do these shortcuts reveal the listed properties, they
can also conceal them, or with the Shift key, they can be
The term “überkey” seems to play
on Friedrich Nietzsche’s concept of used in combination with one another and with many of
the “übermensch”—it is a shortcut the shortcuts detailed earlier (such as the Transform short-
more powerful and important than cuts A, P, R, S, and T or the Mask shortcuts M, MM, and F).
others. You want all the changes applied to Masks and Transforms,
not Effects? UU, then Shift+E. Lose the Masks? Shift+M.
The U shortcut is a quick way to find keyframes to edit or
to locate a keyframe that you suspect is hiding somewhere.
But UU—now that is a full-on problem-solving tool all to
itself. It allows you to quickly investigate what has been
edited on a given layer, is helpful when troubleshooting

58


your own layer settings, and is nearly priceless when investi-
gating an unfamiliar project.
Highlight all the layers of a Comp and press UU to reveal all
edits, enable Switches, Modes, Parent, and Stretch columns,
Nerd-Based Compositing
and you see everything in a Comp, with the exception of
Flowchart, the After Effects nodal view, reveals the
. Contents of nested compositions, which must be truth that all compositing applications are, at their
opened (Alt/Option-double-click) and analyzed core, nodal in their logic and organization. How-
ever, this particular tree/node view is diagnostic
individually and high-level only; you can delete a layer but you
. Locked layers can’t create one here.
. Shy layers (disable them atop the Timeline to show all)
. Comp settings themselves, such as motion blur and
frame rate
In other words, this is an effective method to use to under-
stand or troubleshoot a shot.

Dissect a Project
If you’ve been handed an unfamiliar project and need to
make sense of it quickly, there are a couple of other tools
that may help.
The new one of these is probably my favorite new feature
in After Effects CS4: Miniﬂow (Shift key toggles with the
Timeline, Comp, or Layer view active, Figure 2.11b). It
quickly maps any upstream or downstream comps and
allows you to open any of them simply by clicking on it.
If you’re looking for a whole visual map of the project
instead, try Flowchart view (Ctrl+F11/Cmd+F11 or the
tree/node icon in the Composition viewer). You have
to see it to believe it: a nodal interface in After Effects
(Figure 2.11a), perhaps the least nodal of any of the major
compositing applications.
This view shows how objects (layers, compositions, and Figure 2.11 The tree/node interface in Flowchart
(a, top) is a diagnostic rather than a creative tool.
effects) are used, and in what relationship to one another. Click the icon to switch the view to flow left to
The + button above a composition reveals its components; right, which fits well on a monitor, or Option-click
for the cleanest view, toggle layers and effects off at the it to clean up the view. Its usage has largely been
superceded by the new Miniflow (b, bottom),
lower left. which focuses interactively on the current comp.
You can’t make any edits here, but you can double-click
any item to reveal it where you can edit it—back in the
Timeline, of course.

59


Keyframe Navigation and Selection
Although no shortcut can hold a candle to the überkey,
there are several other useful Timeline shortcuts:
. J and K keys navigate backward and forward, respec-
tively, through all visible keyframes, layer markers, and
Work Area boundaries; hide the properties you don’t
want to navigate.
. Click Property name to select all keyframes for a property.
. Context-click keyframe > Select Previous Keyframes
If keyframes are “hiding” outside or Select Following Keyframes to avoid difﬁcult drag
the Timeline, select all of them selections.
by clicking the Property name, . Context-click keyframe > Select Equal Keyframes to hit
Shift-drag a rectangular selection
around those you can see, and all keyframes with the same setting.
delete the rest. . Alt/Option+Shift+Transform shortcut (P, A, S, R, or T)
sets a keyframe; no need to click anywhere.
. Click a property Stopwatch to set the ﬁrst keyframe at
the current frame (if no keyframe exists), or delete all
existing keyframes.
. Cntl/Command-click an Effect stopwatch to set a
keyframe.
. Ctrl+Alt+A/Cmd+Option+A selects all visible keyframes
while leaving the source layers, making it easy to delete
them when, say, duplicating a layer but changing its
animation.
. Shift+F2 deselects keyframes only.
Read on; you are not a keyframe Jedi—yet.

Keyframe Offsets
To offset the values of multiple keyframes by the same
amount in Layer view, select them all, place the current time
indicator over a selected keyframe, and drag the setting; all
Keyframe multiselection in
standard Layer view (but not Graph change by the same increment. If instead you type in a new
Editor) is inconsistent with the rest value, or enter an offset, such as +20 or +-47, with a numer-
of the application: you Shift-click ical value, all keyframes take on the (identical) new value.
to add or subtract a single frame
from a group. Ctrl/Cmd-clicking With multiple keyframes selected you can also
on a keyframes converts it to Auto-
Bézier mode. . Alt+Right/Left Arrow (Option+Right/Left Arrow) to
nudge keyframes forward or backward in time

60


. Context-click > Keyframe Assistant > Time-Reverse Key-
frames to run the animation in reverse without chang-
ing the duration and start/end point of the selected
keyframe sequence
. Option-drag the ﬁrst or last selected keyframe to scale
timing proportionally in Layer view (or use the trans-
form box in the Graph Editor)

Spatial Offsets
3D animators will be familiar with the idea that every
object (or layer) has a pivot point. In After Effects, there
are two fundamental ways to make a layer pivot around a
different location: change the layer’s own Anchor Point
setting, or parent it to another layer.
After Effects is generally designed to preserve the appear-
ance of the composition when you are merely setting up
animation, toggling 3D on, and so forth. Therefore editing
an anchor point position with the Pan Behind tool triggers
the inverse offset to the Position property. Parent a layer to
another layer and the child layer maintains its relative posi-
tion until you further animate either of them.
It’s typically best to set up your offsets and hierarchy before
animating, although this section shows how to go about
changing your mind once keyframes are in place. The
Option key is key to all options.

Anchor Point
The Pan Behind tool (Y) repositions an anchor point in
the Composition panel and offsets the Position value to
compensate. This prevents the layer from appearing in a
different location on the frame in which you’re working.
The Position offset is for that frame only, however, so if
there are Position keyframes, the layer may appear offset
on other frames if you drag the anchor point this way. To
reposition the anchor point without changing Position
. Change the Anchor Point value in the Timeline
. Use the Pan Behind tool in Layer view instead
. Hold the Option key as you drag with the Pan Behind
tool

61


Any of these options lets you reposition the anchor point
without messing up an animation by changing one of the
Position keyframes.
You can also animate the anchor point, of course; this
allows you to rotate as you pan around an image while
keeping the view centered. If you’re having trouble seeing
the anchor point path as you work, open Layer view and
choose Anchor Point Path in the View pop-up menu (Fig-
ure 2.12).

Figure 2.12 Switch default Masks to Anchor Point Path for easy viewing and
manipulation of the layer anchor point. For the bouncing ball, you could move
the anchor point to the base of the layer to add a little cartoonish squash and
stretch, scaling Y down at the impact points.

Parent Hierarchy
Layer parenting, in which all of the Transform settings
(except Opacity, which isn’t really a Transform setting) are
passed from parent to child, can be set up by revealing the
Parent column in the Timeline. There, you can choose a
layer’s parent either by selecting it from the list, or by drag-
ging the Pickwhip to the parent layer, and use the setup as
follows:
. Parenting remains valid even if the parent layer moves,
is duplicated, or changes its name.
. A parent and all of its children can be selected by
context-clicking the parent layer and choosing Select
Children.
. Parenting can be removed by choosing None from the
Parent menu.
. Null Objects (Ctrl+Alt/Command+Option+Shift+Y)
exist only to be parents; they are actually 100 × 100
pixel layers that do not render.

62


You probably knew all of that. You might not know what
happens when you add the Option key to Parent settings:
. Hold Option as you select the None option and the
layer reverts to the Transform values it had before
being parented (otherwise the offset at the time None
is selected remains).
. Hold Option as you select a Parent layer and its Trans-
form data at the current frame applied to the child
layer prior to parenting.
This last point is a very cool and easily missed method for
arraying layers automatically. You duplicate, offset, and
parent to create the ﬁrst layer in a pattern, then duplicate
that layer and Option+Parent it to the previous duplicate.
It behaves like the Duplicate and Offset option in Illustra-
tor (Figure 2.13).

Figure 2.13 Until you know the trick, setting up a series of layers as an array
seems like a big pain. The trick is to create the first layer, duplicate, and offset;
now you have two. Duplicate the offset layer and—this is the key—Alt/
Option+Parent the duplicate to the offset. Repeat this last step with as many lay-
ers as you need; each one repeats the offset. (Tip of the hat to Danny Princz!)

63


Motion Blur
Motion blur is obviously essential to a realistic shot with a
good amount of motion. It is the natural result of move-
ment that occurs while a camera shutter is open, causing
objects in the image to be recorded at every point from the
shutter opening to closing. The movement can be indi-
vidual objects or the camera itself.
Although it essentially smears layers in a composition,
motion blur is generally desirable; it adds to persistence
of vision, relaxes the eye, and is a natural phenomenon.
Aesthetically, it can be quite beautiful.
The typical idea with motion blur in a realistic visual effects
shot is to match the amount of blur in the source shot,
assuming you have reference; if you lack visual reference, a
camera report can also help you set this correctly.
A moving picture camera has a shutter speed setting that
controls the amount of motion blur. This is not the cam-
era’s frame rate, although the shutter does obviously have
to be fast enough to accommodate the frame rate. A typical
ﬁlm camera shooting 24 frames per second has a shutter
that is open half the time, or 1⁄48 of a second.
Motion blur occurs in your natural vision, although you
might not realize it—stare at a ceiling fan in motion, and
then try following an individual blade around instead and
you will notice a dramatic difference. There is a trend in
recent years to use extremely high-speed electronic shut-
ters, which drastically reduces motion blur and gives the
psychological effect of heightened awareness by making
your eye feel as if it’s tracking motion in this manner. This
seems to have started with live television sports coverage
and the use of extremely fast shutters for slow motion
replay cameras.

Decoding After Effects Motion Blur
The Advanced tab of Composition Settings (Ctrl/
Command+K) contains Motion Blur settings (Figure 2.14):
. Shutter Angle controls shutter speed, and thus the
amount of blur.

64


Figure 2.14 These are the default settings; 16 is really too low for good looking
blur at high speed but a 180 degree shutter and –90 Shutter Angle match the
look of a film camera. Any changes you make here stick and are passed along to
the next composition, or even the next project, until you change them.

. Shutter Phase determines at what point the shutter
opens.
. Samples Per Frame applies only to 3D motion blur; it
sets the number of slices in time (samples), and thus,
smoothness.
. Adaptive Sample Limit applies only to 2D motion
blur, which automatically uses as many samples as are
needed up to this limit (Figure 2.15).
Here’s a bit of a gotcha: the default settings that you see in
this panel are simply whatever was set the last time it was
adjusted (unless never, in which case there are defaults).
It’s theoretically great to reuse settings that work across
several projects, but I’ve seen artists faked out by vestigial
extreme settings like 2 Samples Per Frame or a 720 degree
blur that may have matched perfectly in some unique case.

Figure 2.15 The low default 16 Samples Per Frame setting creates steppy looking blur on a 3D layer only; the same anima-
tion and default settings in 2D uses the higher default Adaptive Sample Limit of 128. The reason for the difference is simply
performance; 3D blur is costlier, but like many settings it is conservative. Unless your machine is ancient, boost the number;
the boosted setting will stay as a preference.

65


Shutter Angle refers to an angled mechanical shutter pres-
ent in some cameras; it is a hemisphere of a given angle
that rotates on each frame. The angle is its radius (Figure
2.16). A typical ﬁlm shutter is 180 degrees—open half the
time, or 1⁄48 of a second at 24 frames per second.
Electronic shutters are variable but refer to Shutter Angle
as a benchmark; they can operate down in the single digits
or close to a full 360 degrees. After Effects motion blur
goes to 720 degrees simply because sometimes mathemati-
cal accuracy is not the name of the game, and you simply
Figure 2.16 The 180° mechanical
want more than 360 degrees.
shutter of a film camera prevents
If you don’t know this setting, you can typically nail it by
light from exposing film half the time,
for an effective exposure of 1/48th zooming in and matching background and foreground ele-
of a second. This is the look of film; ments by eye (Figure 2.17). If your camera report includes
electronic shutters can have a higher shutter speed, you can calculate the Shutter Angle setting
equivalent radius, blurring nearly con-
tinuously from one frame to the next, using the following formula:
or a very fast shutter, which permits
little or no blur.
shutter speed = 1 / frame rate * (360 / shutter angle)
This isn’t as gnarly as it looks, but if you dislike formulas,
think of it like this: If your camera takes 24 frames per
second, but Shutter Angle is set at 180 degrees, then the

Figure 2.17 The white solid tracked to the side of the streetcar has been eye matched to have an equivalent blur by adjust-
ing Shutter Angle; care is also taken to set the Shutter Phase to –50% of the Shutter Angle so that the layer stays centered
on the track.

66


frame is exposed half the time (180/360 = ½) or 1⁄48 of a
second. However, if the shutter speed is 1⁄96 per second with
this frame rate, Shutter Angle should be 90 degrees. A 1⁄1000
per second shutter would have a 9 degree Shutter Angle.
Shutter Phase determines how the shutter opens relative to
the frame, which covers a given fraction of a second begin-
ning at a given point in time. If the shutter is set to 0, it
opens at that point in time, and the blur appears to extend
forward through the frame, which makes it appear offset.
The default –90 Shutter Phase setting (with a 180° Shutter
Angle) causes half the blur to occur before the frame so
that blur extends in both directions from the current posi-
tion. This is how blur appears when taken with a camera,
so a setting that is –50% of Shutter Angle is essential when
you’re adding motion blur to a motion tracked shot. Oth-
erwise, the track itself appears offset.

Add, Do Not Subtract
Although software may one day be developed to resolve
a blurred image, it is much, much harder to sharpen a
blurred image elegantly than it is to add blur to a sharp
image. Motion Blur comes for free when you keyframe
motion in After Effects; what about when there is motion
but no blur and no keyframes, as can be the case in pre-
exising footage?
If you have imported a 3D element with insufﬁcient blur,
or footage shot with too high a shutter speed, you have the
options to add the effect of motion blur using
. Directional Blur, which can mimic the blur of layers
moving in some uniform X and Y direction
. Zoom Blur, which can mimic motion in Z depth
(or spin)
. Timewarp, which can add motion blur without retiming
footage
Yes, you read that last one correctly. There’s a full section
on Timewarp later in this chapter, but to use it to add proce-
dural motion blur, just set Speed to 100 and toggle Enable
Motion Blur, then raise Shutter Angle and Shutter Samples
(being aware that the higher you raise them, the longer the

67


render time). The methodology is similar to that of Reel
Smart Motion Blur (RE:Vision Effects); try the demo on the
book’s disc and compare quality and render time.
Don’t forget motion blur when matching foreground and
background elements (Chapter 5). An element perfectly
matched for color and lighting will look wrong until its
motion blur also matches that of the surrounding scene.

Manipulate Time
After Effects is more ﬂexible when working with time than
most video applications. You can retime footage or mix
and match speeds and timing using a variety of methods.

Absolute (Not Relative) Time
After Effects measures time in absolute seconds, rather
than frames, whose timing and number are relative to the
number per second. If frames instead of seconds were the
measure of time, changing the frame rate on the ﬂy would
pose a much greater problem than it does.
Change the frame rate of a comp and the keyframes main-
tain their position in actual time, so the actual timing of an
animation doesn’t change (Figure 2.18), only the position
of the keyframes relative to frames. Here’s a haiku:
keyframes reposition
falling between retimed frames
time remains unchanged

Figure 2.18 The bounce animation remains the same as the comp frame
rate changes; keyframes now fall in between whole frames, the vertical lines
on the grid.

68


Likewise, footage (or a nested composition) with a mis-
matched frame rate syncs up at least once a second, but
the intervening source frames may fall in between comp
frames. Think of a musician playing 3 against 4; one sec-
ond in After Effects is the down beat of each rhythm.

Time Stretch
Time Stretch lets you alter the duration (and thus the
speed) of a source clip—but it doesn’t let you animate the
retiming itself (that requires Time Remap or Timewarp).
The third of the three icons at the lower left of the Time-
line reveals the In/Out/Duration/Stretch columns.
I mostly change the Stretch value and ﬁnd the four-column
approach redundant. I also never use a Time Stretch set-
ting that is anything but an integer multiple or division by
halves: 200%, 300%, 50% or 25%. You can do without the
columns altogether using the Time Stretch dialog (context-
click > Time > Time Stretch).
Ctrl+Alt+R (Cmd+Option+R) or Layer > Time > Time-
Reverse Layer sets the Stretch value to –100%. The
layer’s appearance alters to remind you that it is reversed
(Figure 2.19).

Figure 2.19 The candy striping along the bottom of the
layer indicates that the Stretch value is negative and the
footage will run in reverse.

Layer > Time > Freeze Frame applies the Time Remap
effect with a single Hold keyframe at the current time.

Frame Blending
Suppose you don’t use a stretch value that factors evenly
into 100% (such as 50% or 200%); footage is likely to lurch
in a distracting, inelegant fashion. Enable Frame Blending
for the layer and the composition, and After Effects aver-
ages the adjacent frames together to create a new image
on frames that fall in between the source frames. This

69


also works when you’re adding footage to a comp with a
mismatched frame rate. There are two modes:
. Frame Mix mode overlays adjoining frames, essentially
blurring them together.
. Pixel Motion mode uses optical ﬂow techniques to
track the motion of actual pixels from frame to frame,
creating new frames that are something like a morph of
the adjoining frames.
Confusingly, the icons for these modes are the same as
Draft and Best layer quality, respectively (Figure 2.20), yet
there are cases where Frame Mix may be preferable instead
The optical flow in Pixel Motion
and the Timewarp effect was of merely quicker to render. Pixel Motion can often appear
licensed from The Foundry. The too blurry, too distorted, or contain too many noticeable
same underlying technology is also frame artifacts, in which case you can move back to Frame
used in Furnace plug-ins for Shake, Mix, or move up to the Timewarp effect, with greater con-
Flame, and Nuke.
trol of the same technology (later in this chapter).

Figure 2.20 The Frame Blending
switches for the Comp and Layer are
highlighted. Just because Pixel Motion
mode uses the same icon as Best in
the Quality switch, to the left, doesn’t
mean it’s guaranteed to be the best
choice.

Nested Compositions
Time Stretch (or Time Remap) applies the main composi-
tion’s frame rate to a nested composition; animations are
not frame-blended, instead the keyframe interpolation is
resliced to the new frame rate.
Most of the time, if you put a comp with a lower frame rate
into a master comp, you intended to keep the frame rate of
Effect > Time > Posterize Time can the embedded composition. In such a case, go to the nested
force any layer to take on the speci-
fied frame rate, but effects in the
comp’s Composition Settings > Advanced panel and toggle
Time category should be applied Preserve Frame Rate When Nested or in Render Queue
before all other effects in a given (Figure 2.21). This forces After Effects to use only whole
layer. Posterize Time often breaks frame increments in the underlying composition, just as if
preceding effects.
the comp were pre-rendered with that frame rate.

70


Figure 2.21 The subcomposition will run at its own frame rate when placed in
the master composition instead of resampling, if they are different from one
another.

Time Remap
Time Remap definitely trumps Time Stretch. The phi-
losophy is elusively simple: Time has a value, just like any
other property, so you can keyframe it, ease in and out of
The final Time Remap keyframe is
it, loop and ping-pong it—basically treat it like any other one greater than the total timing of
animation data. All true, but the concept of animating and the layer (in most cases a nonexis-
bending time can become complex. tent frame) to guarantee that the
final source frame is reached, even
Ctrl+Alt+T/Cmd+Option+T or Layer > Time > Enable when frame rates don’t match.
Time Remapping sets two Time Remap keyframes: at the
beginning and one frame beyond the end of the layer.
Time remapped layers have a theoretically infinite dura-
tion, so the final Time Remap frame effectively becomes a
Hold keyframe; you can then freely scale the layer length
beyond that last frame.
Beware when applying Time Remap to a layer whose
first or last frame extends beyond Comp duration; you
may not be able to see those keyframes. In such a case, I
tend to add keyframes at the comp start and end points,
click Time Remap to select all keyframes, Shift-deselect
the ones I can see in the Timeline, and delete to get rid
of the ones I can’t see.
There is also a Freeze Frame option in After Effects;
context-click a layer, or from the Layer menu choose
Time > Freeze Frame, which sets Time Remap (if not
already set) with a single hold keyframe.

71


Timewarp
Although The Foundry’s amazing Furnace plug-ins are
not, at this writing, available for After Effects, Adobe
licensed one component of Furnace, the retiming tool
known as Kronos, to provide the technology used in Pixel
Motion and Timewarp. Pixel Motion is an automated set-
ting described earlier, and Timewarp adds a set of controls
to allow you to improve on its result using the same basic
technology.
Timewarp works by calculating the motion in the footage
in order to generation motion vectors; these describe how
each pixel moves from frame to frame. With an accurate
analysis it is then possible to generate an image made up
of those same pixels, interpolated along those vectors, with
different timing.
Timewarp is just like Time Remap in the following ways:
. It can be used to speed up, slow down, or dynamically
animate the timing of a clip.
. It incorporates the same three Frame Blending modes
(Whole Frames, Frame Mix, and Pixel Motion) as are
available to Time Remap in the Timeline settings; these
To transfer Time Remap keyframes
to Source Frame mode in Timewarp, live in Timewarp’s Method menu.
enable an expression (Chapter 10) . Time Remap keyframes can even be transferred directly
for Source Frame and enter the to Timewarp, but it requires an expression (see note)
following:
d = thisComp.
because one uses frames and the other seconds to mea-
➥frameDuration sure time.
timeRemap * 1/d
However, used with the default Pixel Motion Method
setting, Timewarp goes beyond Time Remap plus Frame
Blending with the same setting in a couple of ways:
. Enable Motion Blur toggle kicks in when footage is
sped up to add motion blur.
. Tuning section lets you reﬁne the automated results of
Pixel Motion.
Timewarp is most typically used to slow footage down,
which requires one extra bit of setup or the layer ends
short. Enable Time Remapping and extend the layer across
your Comp; the Time Remap settings don’t even matter as
Timewarp will override them.

72


Figure 2.22 This sequence was taken Matrix-style, with an array of still cameras
firing in sequence around the subject. To transform this into a slow-motion visual
effect requires optical-flow retiming, and because the foreground travels in
one direction and the background the other, it helps Timewarp to add a matte.
(Sequence courtesy of fxphd.com)

Here are some tips to improve on the default Pixel Motion
result with Timewarp:
. Matte Layer and Matte Channel menus can point to
a matte channel used to isolate foreground and back-
ground motion, as in Figure 2.22. This can be the
source alpha channel, or a matte channel from a sepa-
rate layer.
. Warp Layer control analyzes a separate layer (as speci-
ﬁed in the menu) and applies the result. Sometimes
you need to adjust footage (contrast, luminance, key)
in order to get a better result, but you don’t want those
adjustments in your shot; apply them to a duplicate
layer instead and set that.
The Tuning section is where you trade render time and
accuracy, but don’t assume that greater accuracy always
yields a better result—it’s just not so. These tools make use
of Local Motion Estimation (LME) technology, which is
thoroughly documented in the Furnace User Guide.
The following settings control how Timewarp analyzes
footage:
. Raising Vector Detail certainly increases accuracy; a
setting of 100 assigns one vector per pixel. Not only
does this drastically increase render time, it will in some
cases simply increase artifacts with fast motion. This is
because it is analyzing too much detail, similar to setting
a motion track to too small (and noisy) a search region.
. Smoothing relates directly to Vector Detail; vectors
must join similar pixels of adjacent frames but not be

73


noisy. You can raise Global Smoothness (all vectors),
Local Smoothness (individual vectors) and Smoothing
Iterations in order to combat the noise, but go very far
and lots of detail will be missed. The Foundry claims
that the defaults, which are balanced, work best for
most sequences.
. It typically takes two adjacent images—before and
after—to make a vector, but you can choose Build From
One Image to use only one if that helps.
. LME depends on object brightness remaining consis-
tent throughout the shot. Toggling Correct Luminance
Changes on warns Timewarp that there will be ﬂuctua-
tions, whether sudden (image ﬂicker) or gradual (mov-
ing highlights).
. Error Threshold evaluates each vector before letting it
contribute; raise this value and more vectors are elimi-
nated for having too much perceived error.
. Block Size determines the width and height of the area
each vector tracks; like Smoothing, lower values gener-
ate more noise, higher values less detail. The Foundry
documentation indicates that this value should “rarely
need editing.”
. Weighting lets you control how much a given color
channel is factored. The defaults may seem odd, but as
you’ll learn in Chapter 5 they correspond to how the
eye weights each color and so they produce a perfectly
monochrome image. If one channel is particularly
noisy—usually blue—you might consider lowering its
setting.
. Filtering applies to the render, not the analysis; it
increases the sharpness of the result. It will cost you
render time, so if you do enable it, wait until you’re
done with your other changes and are ready to render.
Despite that Timewarp is included free, many knowledge-
able After Effects artists still depend on Twixtor (RE:Vision
Effects); like 3D trackers, keyers, and other automated
solutions, this is more a question of what works for you
than of one being better than another. A demo of Twixtor
can be found on the book’s disc.

74


Did you notice back in the Motion
Blur section that Timewarp can
be used to generate procedural
motion blur without retiming
footage (Figure 2.23)?

Figure 2.23 Footage that is shot overcranked (at high speed, a) typically lacks
sufficient motion blur when retimed. Timewarp can add motion blur to speed up
footage; it can even add motion blur to footage with no speed-up at all, in either
case using the same optical flow technology that tracks individual pixels. It looks
fantastic.

Conclusion
The elegance and logic of the After Effects Timeline is
not always evident to the new user, but this chapter taught
you that shortcuts and other workflow enhancements help
streamline what might otherwise be tedious or exacting
edits. The Timeline and Graph Editor, once mastered, give
you the control you need over the timing and placement of
elements.
If this chapter’s information seems overwhelming on first
read, keep coming back to it so that specific tips can sink
in once you’ve encountered the right context in which to
use them.

75

CHAPTER

3
Selections:
The Key to Compositing

I’m fixing a hole where the rain gets in
And stops my mind from wandering
Where it will go.
—John Lennon and Paul McCartney

Selections:
The Key to Compositing

A particle physicist works with atoms, bakers and bank-
ers each work with types of dough, and compositors work
with selections—many different types of selections, poten-
tially thousands, each derived uniquely.
If compositing were simply a question of taking pristine,
perfect foreground source A and overlaying it onto per-
fectly matching background plate B, there would be no
compositor in the effects process; an editor could accom-
plish the job before lunchtime.
Instead, compositors break sequences of images apart and
reassemble them, sometimes painstakingly, first as a still
frame and then in motion. Often, it is one element, one
frame, or one area of a shot that needs special attention.
By the clever use of selections, a compositor can save the
shot by taking control of it.
This chapter focuses on how a layer merges with those
behind it. Then Section II, “Effects Compositing Essen-
tials,” and in particular in Chapters 6 and 7, examines
particular ways to refine selections, create high-contrast
mattes, and pull keys.

78


Selection Types
You may already be familiar with all of the ways to create
layer transparency; here’s a review just in case.

Matte
Mattes do not only apply when keying out the blue or
green from a visual effects shoot (Figure 3.1); high-contrast
(hi-con) mattes are created by maximizing the contrast of
a particular channel or area of the image. You can effec-
tively matte an image with itself—with its own highlights or
color areas.
Chapter 6 goes into depth about using mattes—the process
involves using pixel values to create transparency.

Figure 3.1 This split-screen image shows a blue-screen shoot (left) and the resulting matte.

79

Chapter 3 Selections: The Key to Compositing

Alpha Channel
An alpha or transparency channel is just a matte that has
already been created outside of After Effects and embed-
ded in an imported image. After Effects itself can, of
The built-in assumption of
unmultiplied edge pixels can, course, also create alpha and transparency channels in
in some cases, make life more rendered images (Figure 3.2).
difficult should things not go as
planned. The “Alpha Channels and After Effects is unique among compositing applications in
Premultiplication” section later in that alpha channel information persists everywhere. Every
this chapter offers the lowdown image (and viewer) contains three channels of color and a
on changing edge multiplication fourth transparency channel.
midstream.
After Effects is also unique in that, internally, it assumes
that any edge premultiplication has been removed. Inter-
nally, all alphas in After Effects are straight (see Chapter 1
for a review of how this is determined on import).

Figure 3.2 A computer-generated
baseball’s color and alpha channels.

Mask
A mask is a vector shape, drawn by hand (Figure 3.3). As
a vector shape, it can be inﬁnitely scaled without losing
any deﬁnition, but because it is drawn by hand, an ani-
mated mask has to be considered costlier than procedural
options. The process of hand-animated selection is known
as rotoscoping and is detailed in Chapter 7.

80


Figure 3.3 This split-screen view
shows the garbage matte mask that
was added to remove areas of the
stage not covered by the blue screen.

There is also a procedural option to create animated vector
masks by tracking raster data (pixel values): Layer > Auto-
trace. While technically impressive, Auto-trace is problem-
atic as a selection tool because it typically creates dozens
of masks on any but the simplest live-action shot; you have
less control than with the other methods, with extra ben-
eﬁts only if you want to do something stylized with those
masks (Figure 3.4).

Figure 3.4 Apply Auto-trace to a com-
plex image like this, where you would
hope it would help with automatic
rotoscoping, and you get dozens or
even hundreds of individual colored
mask shapes; a hi-con matte would
give you more control.

81


Blending Mode
Blending modes apply a specific mathematical formula to
the operation of compositing a layer with those behind it.
These are essential because they mimic real-world optics
(Figure 3.5).

Figure 3.5 Blending modes are the
preferred way to composite elements
that are composed predominantly
of light rather than matter. Fire and
smoke are added and multiplied into
this scene detailed in Chapter 14,
respectively.

When compositing an element that is made up more of
light or shadows than reflective surfaces, for example
footage of fire shot on black, it is essential to use blending
modes instead of a luminance matte—don’t try keying out
the black (see Chapter 14 for more details). You can of
course use selections combined with blending modes to
get the best of both worlds.
The primary useful blending modes are detailed later in
this chapter.

82


Effect
Many effects generate transparency: some (such as Levels
and Curves) by offering direct adjustment of the alpha
channel, others (in the Channel folder) by creating or
Close-Up: The Compositing Formula
replacing the alpha channel, and still others that generate
Do you really want to know what happens when
images from scratch that can optionally include an alpha you combine one layer with transparency data (an
channel. alpha channel) over another layer?

Combined Techniques
The foreground pixel values are first multiplied by
Even an ordinary effects shot will typically combine more the percentage of transparency, which, if not fully
than one of the above techniques. You will typically apply a opaque, reduces their value. The background pixels
garbage matte prior to a color key, or enhance the effect of are multiplied by the percentage of opacity (the
inverse of the foreground layer’s transparency), and
a blending mode with a matte. the two values are added together to produce the
The art is in knowing which approach to apply for a given composite. Expressed as a formula, it looks like
(Fg * A) + ((1–A)*Bg) = Comp
situation, and for this, there is no substitute for a deep
understanding of how they work.
With real RGB pixel data of R: 185, G: 144, B: 207
in the foreground and R: 80, G: 94, B: 47 in the
Compositing: Science and Nature background, calculating one edge pixel only might
look like
What exactly is happening in a simple A over B composite? [(185, 144, 207) 3 .6] + [.4 3
In After Effects, it seems nearly as natural as laying one ➥(80, 94, 47)] = (143, 124, 143)

object on top of another does in the real world. In most
other compositing applications even A over B is a compos- The result is a weighted blend between the bright-
iting operation and that is closer to the truth—a truth that ness of the foreground and the darker background.
After Effects obscures in order to make the process more
natural to an artist (particularly one who understands Other effects compositing programs, such as
Photoshop). Shake, do not take this operation for granted the
way that After Effects and Photoshop do. You can’t
Not only that, but there is more to what is going on in the simply drag one image over another in a layer
real world than might be obvious, because of the phenom- stack—you must apply an Over function to create
ena of optics. The three realms of virtual images—the this interaction.
physical world, human vision, and optics—are interdepen-
dent; if you’re shaky in your understanding of one of them, This is not a disadvantage of After Effects—it actu-
it will probably affect all three. ally makes basic compositing simpler and faster—
but it can obscure important related details such
You as a compositor are not supposed to re-create reality as edge pixel premultiplication (detailed later in
on your computer screen, but rather re-create the way the this chapter).
camera (and the eye) sees the world. This affects some-
thing even so fundamental as how the edges of things
should look in order for the eye to accept them.

83


Bitmap Alpha
A bitmap selection channel is one in which each pixel is either
fully opaque or fully transparent. This is the type of selec-
tion generated by the Magic Wand tool in Photoshop. You
can feather or blur the resulting edge, but the initial selec-
tion contains no semitransparent pixels.
This type of selection may have an occasional use, but it
belongs to the world of computers, not nature (or optics).
An edge made up of pixels that are either fully opaque or
invisible cannot describe a curve or angle smoothly, and
even a straight line looks unnatural in a natural image if it
completely lacks edge thresholding (Figure 3.6).

Figure 3.6 This bitmap image con- Feathered Alpha
tains no threshold pixels. Compare
this result with that of Figure 3.7 to
Although it’s easy enough to see that a bitmap edge does
see how your monitor displays a not occur in nature, it’s hard to imagine that hard objects
curved shape. should have transparent, feathered edges. Examine the
edge of this book. It appears sharp.
But now study an image of the same thing, and you’ll find
some degree of edge softness. Adding softness, threshold,
or “feather” to an edge approximates this softness in
the hard digital world of single pixels, which are square
and fundamentally either on or off. Properly feathered
edges can
. Approximate organic curves (Figure 3.7)
Figure 3.7 Zoom in far enough on a
diagonal or curve and you see square . Mimic optics behavior
pixels, yet further away your eye
accepts the illusion.
The first point is intuitive enough once you’ve gained
some experience working with raster images, which are digi-
tal images made up of pixels.
Optics can be observed in any photo with no compositing
whatsoever (Figure 3.8). Viewed close-up, areas at the edge
of objects become a fine wash of color combining the fore-
ground and background. This is not due to inaccuracy in
the camera; it is what happens to light as it travels around
objects in the physical world and then through the lens of
the camera (or your eye).

84


Figure 3.8 This image is not composited. Natural softness is apparent along the hard edges of the sign
despite that they are in focus.

85


Opacity
Take two identical layers, no alpha/transparency informa-
tion for either layer. Set each layer to 50% Opacity, and the
result does not add up to 100%. Here’s why.
Figure 3.9 shows light ﬁltering through two overlapping
sheets of paper. Yes, no expense is spared bringing you
these real-world simulations. Let’s suppose that each
Ctrl+Alt (Command+Option)
and the + or – key raises or lowers
sheet is 75% opaque; 25% of the background light passes
layer Opacity by 1%. As everywhere through. Add a second sheet and 25% of 25%—roughly
in After Effects, add the Shift key 6%—passes through both layers. It’s not a lot of light, but
and the increment is 10× or 10%. it’s not zero; it would take a few more layers of paper to
block out the light completely.

Figure 3.9 Although a single sheet of
paper is more than 50% opaque, two
sheets of paper layered one on top of
another are not 100% opaque. This is
how overlapping Opacity is calculated
in After Effects.

86


After Effects mimics this behavior, adding fractional and
not whole Opacity values of two or more layers. This is not
how it’s handled in most other compositing applications,
such as Shake, and so it takes some veterans by surprise.

Alpha Channels and Premultiplication
One major source of confusion and even derision for After
Effects has to do with its handling of alpha channels and
premultiplication. After Effects has a persistent concept of
the alpha channel as part of every image, and this channel
is expected always to be unmultiplied within After Effects,
whether it originated that way or not.
Any color multiplied into edge pixels is to be removed
upon import (in the Alpha section of the Interpret Footage
dialog), and reintroduced only on output. Provided those
Alpha settings are correct, this works surprisingly well, but
at some point you may need to better understand how to
take control of edge multiplication within After Effects.

Premultiplication Illustrated
Premultiplication exists for one reason only: so that ren-
dered images have realistic, anti-aliased edges before they are
composited. Without it, a 3D rendering of a silver spaceship
against a plain black background would appear to have a
crude, blocky outline because the edge pixels would not
contain any of that black background. It’s as if the ship has
already been composited against a solid color and needs to
be uncomposited before it can be composited again.
When you ask After Effects to “guess” how to interpret the
footage (on import, by choosing Guess in the Interpret
Footage dialog, or pressing Ctrl+Alt+G/Cmd+Option+G),
it looks for indications of a solid color background multi-
plied into edge pixels, indicating that the setting should be
Premultiplied.
Imagine the background value to be 0,0,0 or solid black;
an edge pixel is multiplied by 0 (making it pure black)
and then added back to the source, in proportion to
the amount of transparency in the alpha channel pixel.

87


Removing edge multiplication with the Premultiplied set-
ting subtracts this extra black from those edge pixels.
The close-ups in Figure 3.10 show a section of the same
foreground image with two alpha interpretations, one
interpreted correctly, the other not. A misinterpreted
alpha either fails to remove the background color from
the edge pixels or removes color that should actually be
present.

Figure 3.10 Motion blur and a white
background clearly reveal improper
edge multiplication, especially when
compared with the correct ver-
sion (top). There is dark matting all
around the edges, including areas
of the canopy meant to be trans-
lucent, and around the blur of the
propeller (bottom).

You may ﬁnd that fringing appears in your comps despite
your careful managing of the alpha channel interpreta-
tion on import. This does not indicate some bug in After
Most computer-generated images
are premultiplied, unless specific Effects, but rather a mystery you must solve. There are two
steps are taken to counteract the basic ways it can occur:
process. The Video Output section
of the Output Module Settings for . An alpha channel is misinterpreted in Interpret
items in the Render Queue includes Footage.
a menu to specify whether you . Edge multiplication is added within After Effects, prob-
render with Straight or Premulti-
plied alpha; by default, it is set to ably unintentionally, by applying a matte to a layer that
Premultiplied. has already been matted.
Unfortunately, artists who misunderstand the underlying
problem will resort to all sorts of strange machinations to
ﬁx the black edge, ruining what may be a perfectly accu-
rate edge matte.

88


Get It Right on Import
Preferences > Import > Interpret Unlabeled Alpha As
determines what happens when footage with an unlabeled
alpha channel is imported; the default is to Ask User.
The Ask User dialog has three choices, one of which is
checked, and a Guess button (Figure 3.11). This is confus-
ing to most users, as it seems as if After Effects has already
After Effects attempts to guess not
guessed, when it has not: It is merely using whatever was the only the setting but the background
setting the previous time. color of a premultiplied image;
generally this is black or white, but
The Guess option is not accurate 100% of the time; if the watch out for situations where a
foreground and background are similar, it can be fooled. 3D artist has become creative and
rendered against canary yellow
or powder blue. For that reason,
there is an eyedropper adjacent
to the Matted With Color setting
(Figure 3.11).

Figure 3.11 Be careful here: Many experienced artists assume that After Effects
has already made a guess (here, “Straight”) when it is merely using whatever was
set the last time. If you want it to guess, you have to click that button; ideally you
should avoid the need to do so, of course, but you didn’t need me to tell you that.

Ideally you will work on a project whose images are con-
sistent (in terms of edge multiplication and background
color); in that case, you can set an Import preference.
RGB Straight (Alt+Shift+4/
Typically, however, it’s best to be able to ﬁnd out from Option+Shift+4 or use the Show
whomever created it whether your source contains edge Channel menu at the bottom of a
multiplication, and what settings to use. viewer panel) displays the image
in straight alpha mode, as After
When that’s not possible, you need to be able to examine Effects views it internally.
your images and spot the symptoms of a misinterpreted
alpha: dark or bright fringing in the semi-opaque edges of
your foreground.

89


Solve the Problem Internally
The really gnarly fact is that premultiplication errors can
be introduced within a composition, typically by applying a
matte to footage that is already somehow blended—multi-
plied—with a background.
If you see fringing in your edges, you can try the Remove
Color Matting effect (Figure 3.12). This effect has one set-
ting only, for background color, because all it does is apply
The Remove Color Matting effect
will not work properly on a layer
the unpremultiply calculation (the antidote to premulti-
with a track matte; be sure to plication) in the same manner that it would be applied in
precompose the layer and its track Interpret Footage.
matte prior to applying Channel >
Remove Color Matting. An even better option in cases where you have an element
against black and no alpha channel is UnMult, originally
created by John Knoll and included on the book’s disc in
the Red Giant Software folder. This effect uses black areas
of the image to create transparency and removes the multi-
plied black from the resulting transparent pixels.

Figure 3.12 The plane was matted
against a white background, but
transparency has been applied via
a track matte (the equivalent of a
straight alpha), so white fringing
appears against black (top). Remove
Color Matting with Color set to pure
white corrects the problem (bottom),
but only when applied to a precomp
of the image and matte.

90


Masks
Masks are the principal non-procedural method to define
transparency in a layer; they are vector shapes that you
draw by hand. There are five basic shapes (the Q key cycles
Shape layers are directly related
through them) and the Pen tool (G) for drawing free-form. to masks; they are drawn with
A mask can be drawn in either the Composition or Layer the same tools. If a layer that can
receive a mask is selected, then
viewer. In Layer view the source image persists in its default After Effects draws a mask; other-
view; there is a Render toggle next to the Masks selection wise, it creates a new Shape layer.
in the View menu to disable all mask selections. Artists may
want to see a masked layer in the context of the comp but
find it difficult to adjust the mask in that view—in such a
case, the Layer and Composition views can be arranged
side-by-side (Figure 3.13).

Figure 3.13 With the Comp and Layer panels side-by-side, you can leave the mask enabled in the Comp but uncheck
Render in the Layer.

When using a Mask Shape tool
. Double-click the tool (in the Tools palette) to set the
boundaries of the mask shape to match those of the Look for situations where starting
layer. with a Mask Shape and editing it
can help you; for example, you can
. Press Shift to constrain proportions when drawing or make a half-circle by deleting one
scaling. vertex and adjusting two others of
. Use Ctrl/Cmd to draw from the center (with the Rect- an ellipse.
angle, Rounded Rectangle, and Ellipse tools).

91


. Click Shape under Mask Path (M) in the Layer Switches
column to open the Mask Shape dialog; here you can
enter exact mask dimensions.
. Double-click the shape with the Selection tool to acti-
vate Free Transform mode, then
. Shift-drag on corner to scale the mask
proportionally.
. Shift-drag outside corner to snap rotation to
45-degree increments.
. Shift-drag anywhere else to transform on one axis
only.
. Press the M key twice, rapidly, to reveal all Mask options
for the selected layer.
Ctrl+Alt+Shift- (Cmd+Opt+ . Press the F key to solo the Mask Feather property.
Shift-) click on Mask Expansion. Feather is applied everywhere equally on the mask,
The property disappears. Click equidistant inward and outward from the mask shape.
MM twice to reveal it as “Mask
Embiggen”—an easter egg . Use Mask Expansion to expand or (given a negative
referencing Season 3 Episode 13 of value) contract the mask area. Two masks can be used
The Simpsons. together, one contracted, one expanded, to create an
edge selection.
Chapter 7 has more information about drawing precise
masks; big, soft masks are referenced throughout the
book for all kinds of lighting, smoke, and glow effects
(Figure 3.14).

Figure 3.14 You can build up a series
of layers with soft masks to create
depth in cloud cover, these clouds are
made up of a series of overlapping
masked layers, and each mask has a
Feather value of 200–500 pixels.

92


Bézier Masks
By default, the Pen tool creates Bézier shapes; learn the
keyboard shortcuts and you can fully edit a mask without
ever clicking anywhere except right on the mask.
I like to start by placing points at key transitions and cor-
ners, without worrying about ﬁne tuning the Béziers. Or, as
I draw a point, I can
. Shift-hold and drag to move the vertex
. Hold and drag out a Bézier tangent
before drawing the next point. Once I’ve completed a
basic shape, I can activate the Pen Tool (G) and
Click a point to delete it.
Click a segment between points to add a point.
Alt/Option-click on a point to enable the Convert
Vertex tool, which toggles Bézier handles; drag a point
with no handles to create them, or click a point with
handles to kill them.
Click a Bézier handle to break the handles and adjust
them independently.
Press the Shift key with the mouse still down to pull out
Bézier handles.
Ctrl/Cmd-click to toggle the Selection tool temporarily
(to move a point).
. Press the V key to activate the Selection tool. Pressing
the G key switches back to the Pen.
. Press F2 or Ctrl+Shift+A (Cmd+Shift+A) to deselect
the current mask and start a new one without switching
tools, leaving the Pen tool active.
Context-click on a mask path to change settings in the
Mask submenu. This includes all settings from the Time-
line as well as Motion Blur settings just for the mask
(optionally separate from the Layer). The Mask and Shape
Path submenu contains special options to close an open
shape, set the First Vertex (see later in this chapter) and
toggle Rotobeziers (Chapter 7).

93


Shape Layers
Shape layers add functionality from Adobe Illustrator
directly into After Effects. The same tools can be used to
draw either a mask or a Shape layer. Here’s how they differ:
. Create a Star, Polygon, or Rounded Rectangle as a
mask and its vertices can be edited as normal Béziers.
Shapes offer a different type of control in the Timeline
of properties such as number of points and inner and
outer roundness.
. Shapes can contain Shape Effects such as Pucker &
Bloat, Twist, and Zig Zag that procedurally deform the
entire shape.
. Shapes display with two optional characteristics: Fill and
Stroke. With a Shape active, Alt/Option-click on Fill
and Stroke in the toolbar to cycle through the options
(also available in the Timeline).
. Shapes can be instanced and repeated in 2D space;
Option-drag to duplicate (as in Illustrator) or use a
Repeater effect to instance and array a shape.
Deﬁnitely consider shapes when you need a repeatable
pattern of some type, as in Figure 3.15. Using the Repeater,
you only have to adjust a single shape to edit all instances
of it and how it is arrayed. For the time being, there is no
option to array shapes in 3D.

Figure 3.15 Shapes are not mere
eye candy fodder, are they? The
sprocket holes in this film were made
with a Rounded Corner shape and
a Repeater (I even added an Inner
Shadow Layer Style to give a little feel-
ing of depth and dimension).

94


Figure 3.16 Add mode combines Figure 3.17 Subtract mode is the Figure 3.18 Intersect mode adds only
the luminance values of overlapping inverse of Add mode. the overlapping areas of opacity.
masks.

Combine Selections
By default, all masks are drawn in Add mode, meaning that
the contents of the mask are added to the layer selection,
and the area outside all of the masks is excluded. You can
set the mask mode to
. Add to mask contents to the image as a whole (Figure
3.16) Figure 3.19 The inverse of Intersect,
Difference mode subtracts overlap-
. Subtract to mask contents from displayed areas of the ping areas.
image (Figure 3.17)
. Intersect to mask contents to show only areas overlap-
ping with masks higher in the stack (Figure 3.18)
. Difference to mask contents to hide areas overlapping
with masks higher in the stack (Figure 3.19)
. None to disable the mask (Figure 3.20)
The Inverted toggle next to Mask Mode menu combined
with Add causes the areas outside the mask to be added;
combined with Subtract causes the areas outside the mask Figure 3.20 With None mode, the
to be subtracted, and so on. mask is effectively deactivated.

The Mask Opacity property (M) attenuates the strength of a
mask; setting any mask other than the first one to 0% is like
disabling it. This control works differently for the first (top)
mask. A single Add mask set to 0% Mask Opacity causes the
entire layer to disappear, inside or outside the mask.
However, if you set the first mask to Subtract, and Mask Preferences > User Interface
Opacity to 50%, it does just that—instead of the area inside Color > Cycle Mask Colors assigns
the mask reappearing, the rest of the scene becomes 50% a unique color to each new mask.
Enable it.

95


transparent, just as with Add > Inverted. If you want this
to behave as it should, set the first mask full-frame to Add
mode (just double-click the rectangle mask), then add the
Subtract mask as the second (or later).
To keep multiple masks organized
. Enable Preferences > User Interface Color > Cycle
Chapter 7 demonstrates how Mask Colors to assign a unique color to each new mask
effective rotoscoping ideally
. Press the Return key with mask selected, then type in a
involves many simple masks
working together instead of one big unique name
complex mask. . Click Mask Color swatch (to the left of the name) to
make it more visible or unique
. Context-click > Mask > Locked, Mask > Lock Other
Masks, or Mask > Hide Locked Masks to keep masks
you no longer wish to edit out of your way

Overlapping Transparency
“Density” is traditionally a film term describing how dark
(opaque) the frame of film is at a given area of the image:
the higher the density, the less light is transmitted. Masks
and alpha channels are also referred to in terms of “den-
sity,” and when two masks or mattes overlap, density can
build up when it should not (with masks) or fail to build
up when it should (with mattes).
Figures 3.21 and 3.22 show the simple solution to a com-
mon problem; the Darken and Lighten mask modes
prevent any pixel from becoming more dense than it is in
the semi-transparent areas of either matte. These modes
should be applied to the masks that are below overlapping
masks in the stack in order to work.

Overlap Inverted Layers Seamlessly
Suppose you want to slice a layer into segments and adjust
each segment individually but let them appear combined
in the final result. A gap will appear along the threshold
areas of the matte for the reasons explained in the Opacity
section earlier; two overlapping 50% opaque pixels do not
make a 100% opaque combined pixel.

96


Figure 3.21 A Darken mask (left) uses only the darker
(lower) value where threshold (semi-opaque) pixels
overlap. It prevents two masks building up density as in
Intersect mode (right).

Figure 3.22 A Lighten mask (left) uses only the lighter
(higher) value where threshold (semi-opaque) pixels
overlap. It prevents two masks building up density as in
Add mode (right).

97


Figure 3.23 Comp a layer with matte A (upper left) over one with matte B (upper right) and you get a halo along the overlapping,
inverted threshold edge pixels—around the wheels (bottom left). Alpha Add does just what the title implies, adding the alpha values
together directly (bottom right).

Just as the name implies, the Alpha Add blending mode
directly adds transparent pixels, instead of scaling them
proportionally (Figure 3.23). You can cut out a piece of a
layer, feather the matte, apply the inverted feathered matte
to the rest of the layer, and recombine them with Alpha
Add applied to the top layer. The seam disappears.

Masks in Motion
Ahead of Chapter 7’s more detailed discussion of rotoscop-
ing, here are the basics to put a mask in motion.
Alt/Option+M sets a mask keyframe to all unlocked layer
masks. Mask movement can be eased temporally, but there
are no spatial curves; each mask point travels in a com-
pletely linear fashion from one keyframe to the next. An
arced motion requires many more keyframes.

98


You can only adjust a mask point on one keyframe at a
time, even if you select multiple Mask Path keyframes
before adjusting.
To arc or offset the movement of a mask animation, you
can duplicate the masked layer and use it as an alpha track
matte for an unmasked source of the same layer (see Chap-
ter 7 for more); you can then keyframe and transform its
position and anchor point like any animated layer, with
curved paths and offsets.

Move, Copy, and Paste Masks
Copy a mask path from any compatible source
. A Mask Path property from a separate Mask or Layer
. A Mask Path keyframe from the same or a separate
Mask If the target layer has unique
. A Mask Path from a separate Adobe application such as dimensions, the mask stretches
Illustrator or Photoshop proportionally.

and paste it into an existing Mask Path channel, or paste it
to the layer to create a new Mask. If there are any key-
frames, they are pasted in as well, beginning at the current
time; make sure they don’t conﬂict with existing keyframes
in the Mask Shape.
To draw an entirely new shape for an existing, keyframed
Mask Path, use the Target menu along the bottom of
the Layer panel to choose the existing mask as a tar-
get, and start drawing. This replaces the existing shape
(Figure 3.24).

Figure 3.24 This pop-up menu along
the bottom of the Layer panel makes
it easy to create a new mask path that
replaces the shape in the target mask.
If the target mask has keyframes, After
Effects creates a new keyframe wher-
ever the new shape is drawn.

99


First Vertex
When pasting in shapes or radically changing the exist-
ing mask by adding and deleting points, you may run into
Window > Mask Interpolation is difficulty lining up the points. Hidden away in the Layer >
designed to smooth transitions Mask (or Mask context) menu, and available only with a sin-
between radically different shapes. gle vertex of the mask selected, is the Set First Vertex command.
If your mask points twist around to the wrong point during
an interpolation, setting the First Vertex to two points that
definitely correspond should help straighten things out.
This also can be imperative for effects that rely on mask
Traditional optical compositing— shapes, such as the Reshape tool (described in Chapter 7).
covering all movies made prior to
the 1990s—was capable of
bi-packing (multiplying) and Blending Modes: Compositing Beyond
double-exposing (adding) two
source frames (layers). Many Selections
sophisticated effects films were After Effects includes 36 blending modes, each created
completed using only these two
“blending modes.” with a specific purpose—except maybe Dancing Dissolve,
was that one ever useful? For effects work, moreover, the
majority of them are not particularly recommended.
So how do you tell which are the useful ones? Once
you understand how your options work, you can make
informed compositing decisions instead of relying on trial
Color range in these descriptions is
normalized to 1; the black to white and error.
range of pixel values is described as
0 to 1 instead of the normal 8 bpc
range of 0 to 255. The calculations
make more sense this way (Figure
3.25).

Figure 3.25 The panel menu for Info has more than one mode, and
you can choose whichever you like. Whichever mode you select also
carries over to the Adobe Color Picker.

100


To help you understand what the various blending modes
are doing, Figures 3.26 through 3.33 blend a grayscale
gradient over a fully saturated background. Contextual
examples using these blending modes follow in the next
section.

Figure 3.26 Blending in this figure is
set to Normal for purposes of com-
parison with those that follow.

Add and Screen
Add and Screen modes both brighten the foreground image
while making darker pixels transparent. Screen yields a sub-
tler blend than Add in normal video color space but does
not work correctly with linear color (details in Chapter 11).
Add mode is every bit as simple as it sounds; the formula is
newPixel = A + B

where A is a pixel from the foreground layer and B is a
background pixel. The result is clipped at 1 for 8- and
16-bit pixels.
Add is incredibly useful with what After Effects calls a
Linearized Working Space, where it perfectly re-creates the
optical effect of combining light values from two images.
Linear Dodge and Add are identical
It is useful for laying ﬁre and explosion elements shot in blending modes; the former is
negative space (against black) into a scene, adding noise or merely Photoshop’s term for the
grain to an element, or any other element that is made up latter.
of light and texture (Figure 3.27).

101


Figure 3.27 Add mode (top left) takes the source foreground element, the fire shot against a black background (top right),
and adds its pixel values channel by channel to the background (bottom left), causing the pure black pixels to disappear
completely (bottom right).

Screen mode has an inﬂuence similar to Add mode’s,
but via a slightly different formula. The pixel values are
inverted, multiplied together, and the result is inverted:
In Screen mode, fully white pixels
stay white, fully black pixels stay newPixel = 1–((1–A) * (1–B))
black, but a midrange pixel (0.5)
takes on a brighter value (0.75), Once you discover the truth about working linearized with
just not as bright as it would be a 1.0 gamma, you understand that Screen is a workaround,
with Add (1). a compromise for how colors blend in normal video space.
Screen is most useful in situations where Add would blow
out the highlights too much—glints, ﬂares, glow passes,
and so on (Figure 3.28).

102


Figure 3.28 The difference between Screen and Add (Figure 3.27) may be subtle in printed figures until you look closely;
notice there’s less brightness in the “hottest” areas of the fire.

Multiply
Multiply is another mode whose math is as elementary as it
sounds; it uses the formula
newPixel = A * B

Keep in mind that this formula uses color values between
0 and 1 to correspond to the colors on your monitor.
Multiplying two images together, therefore, actually has
the effect of reducing midrange pixels and darkening an
To fully comprehend the difference
image overall, although pixels that are full white in both between Add and Screen requires
images remain full white. an understanding of a linearized
working space, which is offered in
Multiply or Add has the inverse effect of Screen mode, Chapter 11.
darkening the midrange values of one image with another.
It emphasizes dark tones in the foreground without replac-
ing the lighter tones in the background, useful for creating
texture, shadow, or dark fog (Figure 3.29).

Figure 3.29 Dark smoke (actually a grayscale fractal noise pattern) is multiplied over the background, darkening the areas
that are dark in either the foreground or background further.

103


Figure 3.30 Overlay and its inverse, Hard Light, are useful for combining color and texture. Here, an instant lava lamp texture
was created using the components shown at the right: a solid with Fractal Noise applied set to Overlay mode on top of a
red-to-yellow gradient.

Overlay and the Light Modes
Overlay uses Screen or Multiply, depending on the back-
ground pixel value. Above a threshold of 50% gray (or .5
in normalized terms), Screen occurs, and below the thresh-
Overlay and the various Light
modes do not work properly with old, Multiply. Hard Light operates similarly, instead using
values above 1.0, as can occur in 32 the top layer to determine whether to screen or multiply,
bpc linearized working spaces (see so the two are inverse effects.
Chapter 11).
These modes, along with Linear and Vivid Light, can be
most useful for combining a layer that is predominantly
color with another layer that is predominantly luminance,
or contrast detail (Figure 3.30). Much of the lava texturing
in the Level 4 sequence of Spy Kids 3-D was created by using
Hard Light to combine a hand-painted color heat map
with moving fractal noise patterns.

Difference
Difference inverts a background pixel in proportion to the
Reversing layer order and swapping
foreground pixel. It can help you line up two identical lay-
Overlay for Hard Light yields an
identical result. ers, which is handy while working even if you rarely use it
for ﬁnal output (Figure 3.31).

Figure 3.31 The selection area in the foreground is identical
to the background; when they are perfectly aligned, all pixels
cancel out to black.

104


HSB and Color Modes
The Hue, Saturation, and Brightness modes each combine
one of these values (H, S, or B) from the foreground layer
with the other two from the background layer. Color takes
both the hue and saturation from the top layer, using only
the luminance (or brightness) from the underlying back-
ground (Figure 3.32).

Figure 3.32 Setting a deep-blue-colored solid to Color mode and overlaying it
on the plate footage has the effect of tinting the colors in the image blue. Artistic
uses of this mode are explored in Chapter 12.

These modes are often useful at an Opacity setting below
100%, to combine source HSB values with ones that you
choose.

Stencil, Silhouette, and Preserve Transparency
Commonly overlooked, Stencil and Silhouette blending
modes operate only on the alpha channel of the compo-
sition. The layer’s alpha or luminance values become a
Stencil Alpha and Silhouette Alpha
matte for all layers below it in the stack. Stencil makes the are useful to create custom edge
brightest pixels opaque, and Silhouette the darkest. mattes as well as a light wrap
effect, demonstrated in Chapter 12.
Suppose instead you have a foreground layer that is meant
to be opaque only where the underlying layers are opaque,
as in Figure 3.33. The small highlighted checkbox, labeled
Preserve Underlying Transparency, makes this happen,
much to the amazement of many who’ve wished for this
feature and not realized it was already there.

105


Figure 3.33 Among the hardest-to-find and most-easily-forgotten features in the Timeline is the Preserve Underlying
Transparency toggle, circled. This re-creates behavior familiar to Photoshop users, where a layer’s own transparency only
applies where it intersects with that of the underlying layer, one more way to avoid track mattes or precomping.

Luminescent Premultiply
Luminescent Premultiply is one method you can use to
remove premultiplication on the ﬂy from source footage,
retaining bright values in edge pixels that are otherwise
clipped. Premultiplication over black causes all semitrans-
parent pixels to become darker; removing it can cause
them to appear dimmer than they should.
Luminescent Premultiply can be useful in cases where an
element with transparency has been created against a black
background within After Effects, bypassing the opportunity
to remove premultiplication on import.

Track Mattes
Track mattes allow you to use the alpha or luminance
information of one layer as the transparency of another
layer (Figure 3.34). It’s a simple enough concept, yet one
that is absolutely fundamental as a problem-solving tool for
complex composites.
The perceptual difference between an alpha channel and
a track matte isn’t, for the most part, too difﬁcult to grasp.
In both cases, you have pixels with a value (in 8-bit color

106


Figure 3.34 The alpha of layer 1 is set
as the alpha of layer 2 via the circled
pop-up menu. The small icons to the
left indicate which is the image and
which is the matte.

space) between 0 and 255, whether a grayscale alpha
channel or three channels of color. With color, the three
channels are simply averaged together to make up a single
grayscale alpha. With 16 and even 32 bpc, it’s ﬁner incre-
ments in the same range.
To set a track matte, place the layer that contains the trans-
parency data directly above its target layer in the Timeline
and choose one of the four options from the Track Matte
Share a Matte
pop-up menu:
Node-based compositing programs make it easy
. Alpha Matte: The alpha channel of the track matte for a single node to provide transparency to as
layer is the alpha many others as is needed. The one-to-many capac-
ity in After Effects involves precomping the track
. Alpha Inverted Matte: Same but the black areas are matte and making all changes in the nested comp.
opaque This can be a pain, but check Chapter 4 for more
tips to smooth the process, such as the ability to
. Luma Matte: Uses the average brightness of red, green, display more than one Timeline simultaneously.
and blue as the alpha
. Luma Inverted Matte: Same but the black areas are
opaque
By default, visibility of the track matte layer is disabled
when you activate it from the layer below by choosing one
of these four modes, which is generally desirable. Some
clever uses of track mattes leave them on; for example,
by matting out the bright areas of the image and turning
on the matte, setting it to Add mode, you could naturally
brighten those areas even more.
Track mattes solve a lot of compositing problems. They
also help overcome limitations of After Effects. Chapter 7
describes how you can keep an entire sequence buffered
in RAM for fast access while rotoscoping on a track matte
layer, and how you can add tracking capabilities to masks
in this manner.

Gotchas
Even an advanced user has to pay attention when work-
ing in a composition with track mattes. Unlike parented

107


layers, track mattes do not stay connected with their target
if moved around; they must occupy the layer directly above
in order to work.
After Effects does at least help you in certain ways. Dupli-
cate a layer (Ctrl+D/Cmd+D) with a track matte acti-
vated and it moves up two layers, above the track matte
layer. Include the track matte when you duplicate and
it also moves up two layers, so layer order is preserved
(Figure 3.35).

Figure 3.35 Select and duplicate the
layers from Figure 3.34, and the two
new layers leapfrog above to maintain
the proper image/matte relationship.

There is a work-around that allows a matte layer to be
anywhere in the Timeline, but it contains its own perils.
Effect > Channel > Set Matte not only lets you choose any
Combine a track matte and an
image with an alpha channel, and layer in the comp as a matte, it keeps track if that layer
the selection uses an intersection moves to a different position. It also offers a few custom
of the two. matte handling options regarding how the matte is scaled
and combined. However, nothing you add to the other
layer, including Transform keyframes, is passed through.

Render Order
This brings us to render order with track mattes. In most
cases, adjustments and effects that you apply to the matte
layer are calculated prior to creating the target matte. To
If you’re not certain whether your
edits to the matte are being passed see how this can break, however, try applying a track matte
through, save the project and try to another track matte? It works… sometimes, but not
cranking them up so it’s obvious. often enough that it should become something you try
Then undo/revert. If it’s not work- unless you’re willing to troubleshoot it.
ing, precomp the matte layer.

Conclusion
Now that we’ve examined selections in detail, the next
chapter looks in depth at solving issues such as those
described above for render order; you’ll begin to see how
to use the Timeline as a visual problem-solving tool for
such situations.

108

CHAPTER

4
Optimize the Pipeline

Build a system that even a fool can use and only a fool
will want to use it.
—George Bernard Shaw

Optimize the Pipeline

T his chapter examines how image data flows through an
After Effects project in detail. That may not sound grip-
ping until you realize how much your creativity is ham-
pered by remaining ignorant of this stuff.
Sometimes you take the attitude of a master chef—you
know what can be prepped and considered “done” before
the guests are in the restaurant and it’s time to assemble
the piece de resistance. At other times, you’re more like a
programmer, isolating and debugging elements of a proj-
ect, even creating controlled tests to figure out how things
are working. This chapter helps you artistically and techni-
cally, as if the two can be separated.
Once you
. Understand how to use multiple compositions
. Know when to precompose
. Know how to optimize rendering time
you may find the After Effects experience closer to what
you might call “real-time.” Efficient rendering, however,
depends on well-organized compositions and the ability to
plan for bottlenecks and other complications.

Multiple Compositions, Multiple Projects
It’s easy to lose track of stuff when projects get compli-
cated. This section demonstrates
. How and why to create a project template
. How to keep a complex, multiple-composition pipeline
organized
. Shortcuts to help orient you quickly

110


These tips are especially useful if you’re someone who
knows compositing well but find After Effects disorienting
at times.

Project Templates
I’m always happiest with a project that is well organized,
even if I’m the only one likely ever to work on it. Keeping
the Project panel well organized and tidy can clarify your
ability to think about the project itself.
Figure 4.1 shows a couple of typical project templates
containing multiple compositions to create one final shot,
although these could certainly be adapted for a group of
similar shots or a sequence. When you need to return to a
project over the course of days or weeks, this level of orga-
nization can be a lifesaver.

Figure 4.1 The template on the left
is based on how a film shot might
be generically organized, including
numbering that reflects pipeline order
and Output comps in two formats.
The template on the right is the mini-
mum I would use on any project; you
need a folder for Source and one for
Precomps to keep things tidy even if
the main comp stays at the root level.

Here are some criteria to use when creating your own:
. Create folders to group specific types of elements, such
as Source, Precomps, and Reference.
. Use numbering to reflect comp and sequence order so
that it’s easy to see that order in the Project panel.

111

Chapter 4 Optimize the Pipeline

. Create a unique Final Output comp that has the format
and length of the ﬁnal shot, particularly if the format is
at all different from what you’re using to work (because
it’s scaled, cropped, or uses a different frame rate or
If nothing else, a locked, untouch-
color proﬁle).
able Final Output comp prevents
screwing up a render by having the . Use guide layers and comments as needed to help art-
Work Area set wrong because you ists set up the comp (Figure 4.2).
were using it for RAM Previews.

Figure 4.2 Here are a series of nonrendering guide layers to define action areas
and color. Notice the new feature in CS4 that shows where this comp fits in a
series of precomps; each of those buttons opens the corresponding comp.

It’s a smart idea never to work . Organize Source folders for all footage, broken down
with source footage directly, but to
as is most logical for your project.
place each source clip in a precomp.
Why? If the source changes for any The basic elements of a Master comp, source comps, and
reason, using precomps makes a render comp seem useful on a shot of just about any
it much easier to get meticulous
complexity, but the template can include a lot more than
adjustments like masks and paint
strokes to line up without needing that: custom expressions, camera rigs, Color Management
to redo them. settings, and recurring effects setups.

112


Mini-Flowchart View
The Timeline is great for timing, but the bane of the Time-
line is precomping. There are many times where there
is no way around sending a set of layers into their own
Timeline, yet most artists will do anything to avoid this step
because it locks those layers and controls out of the main
Timeline. In the past, precomping also made it harder sim-
ply to find where everything was; it is standard for an After
Effects artist working on a complex comp to be left with
the “where did I do that” feeling at one time or another.
That’s now less the case, thanks to CS4’s Mini-Flowchart, or
Miniflow. You can access it via a button in the Timeline ,
but you may have already discovered that simply tapping
the Shift key with the Timeline forward enables it.
By default, the comp order is shown flowing right to left.
The reason for this is probably that if you open subcomps
from a master comp, the tabs open to the right; however,
a convincing argument can and has been made to choose
Flow Left to Right in the panel menu of the Mini Flow-
chart. Try it and see if you breathe a little sigh of relief.
Either direction, Miniflow is the best workflow addition
to After Effects CS4. It shows only the nearest neighbor
comps (Figure 4.3) but click on the flow arrows at either
end and you navigate up or down one level in the hierar-
chy. Click on any arrows or items in between the ends and
that level is brought forward.

Figure 4.3 Miniflow is a quietly revo-
lutionary small new feature in After
Effects CS4. It makes the process of
navigating subcomps nearly effortless.

Miniflow does not display the entire compositing tree like
Flowchart view does, but it’s much more useful. The most
complicated it gets is in multiple one-to-many relation-
Miniflow makes Comp naming
ships, because you can only trace your way down one flow more important than ever. On a
at a time, but that turns out not to be such a big deal. It’s big project, pick a system and stick
much simpler to open a comp and see where it’s used than with it.
the previously existing alternatives.

113


More Ways to Manage Many Comps
Miniﬂow makes it even more possible to close all the Time-
lines when you’re done with a set of changes (Ctrl+Alt+W/
Command+Option+W) and reopen only the ones you
need, starting with the master comp.
The Lock icon at the upper left of the Composition
viewer lets you keep that Composition viewer forward while
you open another Timeline. Typically, you would lock the
master comp and double-click a nested comp to open its
Timeline; as you make adjustments in that Timeline, you
see their effect on the overall comp.
The Always Preview this View toggle lets you instead
work entirely in a precomp, but preview the result auto-
matically in the master comp (if this is toggled in that
comp) without having to switch to it. Use it if you’re only
interested in how changes look in your ﬁnal.
Ctrl+Alt+Shift+N (Cmd+Option+Shift+N) creates two
Composition viewers side-by-side, and locks one of them,
for any artist with ample screen real estate who wants the
best of both worlds.
To locate comps in the Project panel, you can
. Select an item in the Project panel; adjacent to its name
by the thumbnail at the top of the panel is a small pull-
down caret, along with the number of times, if any, the
item is used in a comp (Figure 4.4).

Figure 4.4 Click the caret next to the
total number of times an item is used
to see a list of where it is used.

. Context-click an item in the Project panel and choose
Reveal in Composition; choose a composition and that
comp is opened with the item selected (Figure 4.5).

114


Figure 4.5 Context-click any item, and under Reveal in Composition, choose
from a list, if applicable; that Timeline opens with the item selected.

. Context-click a layer in the Timeline and choose Reveal
Layer Source in Project to highlight the item in the
Project panel (Figure 4.6).

Figure 4.6 Context-click any footage item in the Timeline and you
can choose to reveal it either in the Project panel or in Flowchart view.

. Context-click in the empty area of a Timeline and
choose Reveal Composition in Project to highlight the
comp in the Project panel (Figure 4.7).
You’ve no doubt already discovered
the reversed functionality to open
nested comps; double-click opens
a nested comp, and Alt/Option-
double-click reveals it in the Layer
viewer. It’s friendlier for new users
but requires the rest of us to adjust.
Figure 4.7 Find the empty area below
the layers in the Timeline and context-
click; you can reveal the current comp
in the Project panel.

115


Precomping and Composition Nesting
Precomping is often regarded as the major downside of
compositing in After Effects, because it obscures vital infor-
mation from view. Artists will let a composition become
completely unwieldy, with dozens of layers, rather than bite
the bullet and precomp. Yet precomping is an effective way
to solve problems and optimize a project, provided you
plan things out a little.
Just to get our terms straight, precomping is the action of
selecting a set of layers in a composition and assigning
them to a new subcomp. Closely related to this is composi-
tion nesting, the action of placing one already created com-
position inside of another.
Typically, you precomp by selecting the layers of a composi-
tion that can and should be grouped together and choos-
ing Precompose from the Layer menu (keyboard shortcut
Ctrl+Shift+C/Cmd+Shift+C).
Two options are presented, the second one unavailable
if multiple layers are selected: to leave attributes (effects,
transforms, masks, paint, blending modes) in place or
transfer them into the new composition.

Why Precomp?
Precomping prevents a composition from containing too
many layers to be seen and managed in one Timeline, but
it also allows you to do the following:
. Reuse a set of elements and manage them from one
place.
. Fix render order problems, for example apply an effect
ﬁrst, then a mask.
. Organize your project; grouping elements that are
interrelated is a good idea in a nodal application such
as Nuke or Shake too, because it inﬂuences how you
think about the composite.
. Consider an element or set of layers “done” (and even
prerender them as proxies, as discussed later in this
chapter).

116


If you’re already comfortable with the idea of precomping,
focus on the last point. This chapter describes the advan-
tages to finishing an element, if only for the time being,
instead of always keeping options open. It’s a secret of
most “real-time” compositing systems that they rely heav-
ily on the equivalent of prerendered subcomps (they just
don’t call them that).

Gotchas
Precomping several layers together can solve problems, but
it can also create new ones. Common gotchas include
. You want some but not all properties to be precomped. The script preCompToLayerDur.jsx
. It can be a pain to undo in order to restore precomped from Dan Ebberts is included
on the disk; even if a layer to be
layers to the master composition. precomped doesn’t start at frame 0,
. You must deal with the all-or-nothing behavior of blend- the precomp does.
ing modes and 3D layers, depending on the Collapse
Transformations setting.
. It can be unclear how motion blur, frame blending, and
collapsed transformation switches in the master compo-
sition affect nested comps.
. Layer timing (duration, In and Out points) and dimen-
sions can become constraints.
If you recognize any of these problems and haven’t solved
them, check out the following useful strategies.

Boundaries of Time and Space
Each composition in After Effects contains its own fixed
timing and pixel dimensions, which a couple of strategies
help navigate:
. Make source compositions longer than the shot is ever
anticipated to be to avoid truncating timing.
Figure 4.8 The nested comp has a
. Enable Collapse Transformations for the nested com- blue background and the leg of the
position to ignore its boundaries (Figure 4.8). letter “p” extends outside its boundar-
ies (top); a simple quick-fix is to enable
These two simple rules will cover many situations, but Collapse Transformations, and the
Collapse Transformations has extra benefits and limita- boundaries of the nested comp are
tions of its own. ignored (bottom).

117


Collapse Transformations
Switches Affect Nested Comps is enabled by default in Gen-
eral Preferences. Enable frame blending or motion blur in a
master composition and those switches are passed through
to any nested composition. 3D position data and blending
modes, on the other hand, are ignored if Collapse Transfor-
mations is not enabled (Figure 4.9). Enable it and it is as if
the precomposed layers resided in the Master comp.

Figure 4.9 The highlighted switch has two roles (and two names): Collapse
Transformations passes through blending modes and 3D positions from nested
comps as if not precomped; Continuous Rasterization maintains vector source
without rasterizing until final render and applies to source in formats such as
Adobe Illustrator.

Collapse Transformations prevents the application of a
blending mode on the collapsed layer. Apply any effect to
the layer (even disable it) and After Effects must render
Grow Bounds
the collapsed layer (making it what the Adobe developers
Sometimes enabling Collapse Transformations is
not desirable—for example, if you set up 3D layers call a parenthesized comp, one that has qualities of being
in a subcomp and don’t want their position to be both rendered and live). You can then apply a Blending
changed by a camera in the Master comp. The Grow mode, but 3D data is not passed through.
Bounds effect overcomes one specific (and fairly
rare) problem, where layers in a nested comp are Thus if you want to collapse transformations but not 3D data,
cut off by the comp boundary. applying any effect—even one of the Expression Controls
effects that don’t by themselves do anything—will parenthe-
size the comp. It’s a good trick to keep in your pocket.

Nested Time
After Effects is less rigid than most digital video applica-
tions when working with time. All compositions in a given
project need not use the same frame rate, and as has been
shown, you can change the frame rate of an existing com-
position on the ﬂy, and keyframes retain placement relative
to overall time.

118


With power comes responsibility, of course, so pay particu-
lar attention when you
. Import an image sequence.
. Create a new composition from scratch.
. Embed a composition with a given frame rate into
another with a different frame rate.
In the first two cases you’re just watching out for careless Annoyed to find sequences import-
errors, but in the third, you might have reason to maintain ing at the wrong frame rate for
different frame rates in subcomps, in which case you must your project? Change the default
set it deliberately on the Advanced tab of the Composition Sequence Footage Frames per Sec-
ond under Preferences > Import.
Settings dialog, or apply the Posterize Time effect set to
the desired rate.

Advanced Composition Settings
In addition to the Motion Blur settings introduced in
Chapter 2 and detailed in Chapter 8, Composition Set-
tings: Advanced contains two Preserve toggles that
influence how time and space are handled when the com-
position is nested into another.
Preserve Frame Rate maintains the frame rate of the compo-
sition wherever it goes—into another composition with a dif-
ferent frame rate or into the Render Queue with different
frame rate settings. So if a simple animation cycle looks right
at 4 frames per second (fps), it won’t be expanded across
the higher frame rate but will preserve the look of 4 fps.
Preserve Resolution When Nested controls what is called
concatenation. Typically, if an element is scaled down in a
precomp and the entire composition is nested into another
comp and scaled up, the two operations are treated as one,
so that no data loss occurs via quantization. This is con-
catenation, and it’s usually a good thing. If the data in the
subcomp is to appear pixilated, as if it were scaled up from
a lower-resolution element, this toggle preserves the big
pixel look.

Adjustment and Guide Layers
Two special types of layers that don’t render, adjustment
and guide layers, offer extra benefits that might not be
immediately apparent.

119


Adjustment Layers
An adjustment layer is itself invisible, but its effects are
applied to all layers below it. It is a fundamentally simple
feature with many uses. To create one, context-click in an
empty area of the Timeline, and choose New > Adjustment
Layer (Ctrl+Alt+Y/Command+Option+Y) (Figure 4.10).

Figure 4.10 Any pixel layer becomes
an adjustment layer by toggling this
switch. An adjustment layer created
by After Effects is a white, comp-sized
solid with this switch toggled on.

Adjustment layers allow you to apply effects to an entire
composition without precomping it. That by itself is pretty
cool, but there’s more:
. Move the adjustment layer down the stack and layers
above it are unaffected.
. Change its timing and the effects appear only on
frames within the adjustment layer’s In/Out points.
. Time any effect in an adjustment layer by setting layer
In/Out points.
. Use Opacity to attenuate the amount effects are applied.
Many effects do not themselves include so direct a con-
trol, even when it makes perfect sense to “dial it back
50%,” which you can do by setting Opacity to 50%.
. Apply a matte to an adjustment layer to apply an effect
to one area of the image.
. Add a blending mode and the adjustment layer is
ﬁrst applied and then blended back into the result
(Figure 4.11).
It’s a good idea 99% of the time to make sure that an
adjustment layer remains 2D and at the size and length of
the comp, as when applied; it’s rare that you would ever
Alpha channel effects change the
alphas of the layers below, not of want to transform an adjustment layer in 2D or 3D, but it
the adjustment layer itself. is possible, so don’t let it happen by accident. If you
enlarge the composition, you must resize the adjustment
layers as well.

120


Figure 4.11 The ball on the top left was created with
a Shape layer. To the right is the ball with a Glow effect
applied (default settings), and with the Blending mode set
to Normal. The bottom image shows what occurs when the
Blending mode is changed to Add: The result of the second
ball has been added to the first.

Guide Layers
Like adjustment layers, guide layers are normal layers with
special status. A guide layer appears in the current compo-
sition but not in any subsequent compositions or the final
render (unless you specifically override this functionality in
Render Settings, which kind of defeats the purpose). Com-
mon uses include
. Foreground reference clips
. Temporary backgrounds to check edge transparency
. Text reminders (Specific render instructions? Add a
text layer with a bullet-pointed list in the render comp
and set it as a guide layer.)
. Adjustment layers that are used only to check images
(described further in the next chapter); a layer can be
both an adjustment and a guide layer
Any image layer can be made a guide layer either by
context-clicking it or by choosing Guide Layer from the
Layer menu. Within the current comp, you’ll notice no

121


difference (Figure 4.12). You can still apply effects to this
layer or have other layers refer to it, and it is fully visible.
Nest this composition in another composition, however,
and the guide layer disappears.

Figure 4.12 A gradient can clarify
matte problems; to ensure that it
doesn’t render, make the background
a guide layer.

Render Pipeline
To become an expert compositor is to precisely understand
the order in which actions are performed on an image,
also known as the render pipeline. For the most part render
order is plainly displayed in the Timeline and follows con-
sistent rules:
. 2D layers are calculated from bottom to top of the
layer stack.
. Layer properties (masks, effects, transforms, paint, and
type) are calculated from top to bottom (twirl down to
3D calculations are precise well see the order).
below the decimal level, but do
. 3D layers are of course calculated based on distance
round at some point. Coplanar 3D
layers can thus introduce rendering from the camera; coplanar 3D layers respect stack-
errors and should generally be ing order and behave—relative to one another—like
avoided by precomping them in 2D. 2D layers.

122


In a 2D composition, After Effects starts at the bottom
layer and calculates any adjustments to it in the order that
properties are shown, top to bottom. Then, it calculates
adjustments to the layer above it, composites the two of
them together, and continues up to the top layer of the
stack, while the properties of an individual layer render top
to bottom (Figure 4.13).

Figure 4.13 2D layers render starting The Transform effect allows you to
with the bottom layer, rendering transform before other effects are
and compositing each layer above
applied to avoid precomping for
in order. Layer properties render in
this purpose.
the order shown when twirled down;
there is no direct way to change the
order of these categories.

So, although effects within layers always calculate prior to
transforms, by applying an effect to an adjustment layer
above, you guarantee that it is rendered after the trans-
forms of all layers below it.
Although the UI doesn’t prohibit
Track mattes (and blending modes) are applied last, after all you from doing so, don’t apply a
other layer properties (masks, effects, and transforms) have track matte to another track matte
been calculated, and after their own mask, effect, and trans- and expect consistent results.
form data are applied. Therefore, you don’t generally need Sometimes it works, sometimes it
doesn’t.
to pre-render a track matte just because you’ve edited it.

Optimize Previews and Renders
As I work, I organize portions of my master comp that I
consider ﬁnished into their own subcomps, and if they
require any render cycles at all, I prerender them. Failure
Preferences > Display > Show
to commit to decisions—keeping options open—costs time Rendering in Process in Info Panel
and efﬁciency. It’s as true in After Effects as it is in life as a and Flowchart shows what is
whole. happening on your system. It is
disabled by default because it
Pre-rendering a subcomp does, however, lead to a decision requires extra processing power;
about what happens after you render it. I leave it on because it costs more
time not to know.
Post-Render Options
Tucked away in the Render Queue panel, but easily visible
if you twirl down the arrow next to Output Module, is a

123


menu of three post-render actions. After the render is com-
plete, you can choose
. Import: Simply imports the result
. Import & Replace Usage: Keeps the source comp but
replaces its use—or that of any other element you
choose instead—in the project
. Set Proxy: Adds a proxy to the source comp (or any
other item you specify)
With either of the latter two, the Pickwhip icon adjacent
to the menu can be clicked and dragged to whatever item
in the Project panel needs replacement so that if you’ve
already created a pre-render or proxy, you can replace it
(Figure 4.14).

Figure 4.14 Virtually any Project item can be the target for replacement or a proxy; click and drag the Pickwhip icon to
choose the item to be replaced by the render.

Proxies have the best potential to speed up your work on
a heavy project, but not everyone likes them because their
implementation complicates the rendering process.
If you choose Import & Replace
Usage and then need to change Proxies and Pre-Renders
back, Alt/Option-drag the source Any visual item in your Project panel can be set with a
comp over the replacement clip
proxy, which is an imported image or sequence that stands
in the Project to globally replace
its usage. in for that item. Its pixel dimensions, color space, com-
pression, and even its length can differ from the item it
replaces; for example, you can use a low-resolution, JPEG-
compressed still image to stand in for a full-resolution,
moving image background.

124


To create a proxy, context-click an item in the Project panel
and choose Create Proxy > Movie (or Still). A render queue
item is created and automatically renders at Draft quality
and half-resolution; the Output Module settings create a
video file with alpha, so that transparency is preserved, and
the Post-Render Action uses the Set Proxy setting.
Figure 4.15 shows how a proxy appears in the Project
panel. Although the scale of the proxy differs from that of
the source item, transform settings within the comps that
use this item remain consistent with those of the source
item so that it can be swapped in for final at any time. This
is what proxies were designed to do, to allow a low-resolu-
tion file to stand in, temporarily and nondestructively, for
the high-resolution final.

Figure 4.15 The black square icon to
the left of an item in the Project panel
indicates that a proxy is enabled; a
hollow square indicates that a proxy is
assigned but not currently active. Both
items are listed atop the Project panel,
the active one in bold.

By default, the source file or composition is used to render
unless specifically set otherwise in Render Settings > Proxy
Use. Choosing Use Comp Proxies Only, Use All Proxies, or
Current Settings options (Figure 4.16) allows proxies to be
used in the final render. To remove them from a project,
select items with proxies, context-click (or go to the File
menu), and choose Set Proxy > None.

Figure 4.16 I typically set Proxy Use to
Current Settings, but Use Comp Prox-
ies Only lets you set low-res stand-ins
for footage and full-res pre-renders for
comps, saving gobs of time.

125


Similar to proxies, but with a different intended use, are
pre-rendered elements. With a composition selected,
choose Composition > Pre-render and a moving image ﬁle
is set to render at Best quality, full resolution; the Import
and Replace Usage is set for the Output Module.
You don’t have to follow this usage; I have made proxies
behave as pre-rendered elements and set my Render Set-
tings to Current so I can choose whether to use the proxy
See Chapter 1 for specifics on
multiprocessing, caching, and or re-render. Managing that process, however, adds an
previewing. extra set of steps.

Background Renders
One way to get a lot more power out of After Effects is to
improve upon the standard method of rendering, which
ties up the application itself, and most of the machine’s pro-
cessing power, for as long as is needed to output footage.

aerender
Almost any production system these days has multiple
processors and cores, and this has opened the possibility of
background rendering, which allows a render to occur in the
background while you continue to work in the foreground.
The last edition of this book mentioned using the com-
mand line (in Terminal Unix shell on Mac or the DOS
shell in Windows) to run aerender; to do this, you can
locate it in your After Effects CS4 folder and drag it into
the shell window; then press Enter to get the manual.
Arguments in quotes can be added to the command
aerender and the location string of the project ﬁle.

However, that’s a little complicated and involves typing
in commands, not the favorite activity of most visual art-
ist types. For that reason I highly recommend loading a
free script from aescripts.com called BG Renderer into
the Script UI Panels folder, also in the After Effects CS4
folder. This script automatically sets up the command line
not only to render in the background, freeing the open
application to be used for more work, but using extra
commands such as nice to set the priority and number of
processors that you specify.

126


Network Rendering
The aerender command is also key to third-party render-
ing solutions such as Rush Render Queue (http://seriss.
com/rush/). These programs run scripts that manage
the process of running aerender on multiple machines
(Figure 4.17). The better applications among them, Rush
included, are capable of far more than just straight-ahead
renders; you can, for example, have one render wait until
a certain time or for another one to complete before com-
mencing, and you can automatically requeue renders that
fail for any reason.

Figure 4.17 One look at Rush Render
Queue’s UI will tell you you’re not in
After Effects land anymore, but it’s
plenty sophisticated at managing
network renders from After Effects and
the many other applications, 2D and
3D, that enable command line render
management.

This type of software is not generally even implemented via
a standard installer; required instead are the implementa-
tion skills of a system administrator or equivalent technical
expert. Most larger facilities have just such a “geek” on
staff, and many advanced After Effects users are themselves
capable of setting this up in smaller studios.

Multiple After Effects Versions
Alternatively, you can open multiple versions of After
Effects including the UI. Although memory intensive, this
allows you to actually investigate, compare, and edit two
projects at once, which is occasionally handy.
On a Mac, all you need to do is locate Adobe After Effects
CS4.app in the Finder (most likely in Applications/Adobe

127


After Effects CS4) and duplicate it (Cmd+D). You now
have two versions of the application that will open sepa-
rately, and you are free to render a project in one version
while continuing to work in the other. Give one of them a
unique name so you can tell them apart.
On Windows, you can open a second version of After
Effects from the command line. From the Start menu,
choose Run, type cmd, and click OK. In the DOS shell that
opens, drag in AfterFX.exe from your Programs folder and
then add “ –m” (that’s a space, a dash, and m as in multi-
ple). Voilà, a second version initializes. Write a .bat ﬁle and
you can do all of this with a double-click. The same basic
trick even works in the Mac Terminal.

Watch Folder
The slightly senile grandaddy of network rendering on
After Effects is Watch Folder. File > Watch Folder looks in a
given folder for projects ready to be rendered; these are set
up using the Collect Files option. The help topic “Render-
ing on the Network: Using a Watch Folder” page includes
everything you need to know, so there’s no reason to reiter-
ate setup here.
Watch Folder is okay on small, intimate networks, but it
has to be set up manually on each machine, and if any-
thing goes wrong, it has to be re-queued. A single machine
has so much rendering power these days, it is years since
I’ve bothered using it even though my studio is about the
right size for it: big enough to have several systems, small
enough to have them all in the same physical space.

Adobe Media Encoder
Adobe Media Encoder is a dedicated render application
for certain video formats, including Flash Video, H.264,
MPEG-2, and Adobe Clip Notes. Choose one of these in
Output Module settings and the Encoder reveals many spe-
cialized options having to do with that particular format.
Owners of Adobe Production Premium or Master Collec-
tion can also access Media Encoder from Premiere Pro,
where it can perform multipass encoding (essential for

128


the best quality compression) and Speech Search, this new
ability to convert spoken words on the soundtrack to timed
text. Each word shows up as a layer marker, theoretically
making it easy to time footage that includes speech, a very
cool concept.

Project Optimization
Finally, to finish Section I of this book, here’s a clean sweep
of preferences, memory management settings, and what
do to if After Effects crashes.

Setting Preferences and Project Settings
The preference defaults have changed in version CS4 and
you may be happy with most of them. Here, however, are
a few you might want to adjust that haven’t been men-
tioned yet:
. Preferences > General > Levels of Undo: The default is
32, which may be geared toward a system with less RAM
than yours. Setting it to the maximum value of 99 won’t
bring the application to a grinding halt, but it may
shorten the amount of time available in RAM Previews.
. Preferences General: Check the options Allow Scripts
to Write Files and Access Network to use some of the
scripts included with this book. Most advanced users
also prefer Default Spatial Interpolation to Linear
(Chapter 2).
. Preferences > Display: I tend to check all three boxes; I
don’t need to wait for thumbnails to update from some
network location each time I select a source file, I like
Alt+Ctrl+Shift/Option+Cmd+
to see rendering progress even though it costs process- Shift immediately after launching
ing time (as I said earlier), and I have a good OpenGL After Effects resets Preferences.
card so I hardware accelerate the UI. Hold Alt/Option while clicking OK
to delete the shortcuts file as well.
. Preferences > User Interface Colors: The default UI
is darker than it used to be, so I am often fine with its
shade. If you like it darker, consider enabling Affects
Label Colors so they darken too. Either way, Cycle
Mask Colors so that multiple masks applied to a layer
automatically have different colors (I have no idea why
that’s disabled by default).

129


Hack Shortcuts, Text Preferences, or Projects
After Effects Shortcuts and Preferences are saved as text
files that are fully editable and relatively easy to understand,
although if you’re not comfortable with basic hacking
(learning how code works by looking at other bits of code)
I don’t recommend it. The files are located as follows:
Windows: Documents and Settings[user profile]Applica-
tion DataAdobeAfter Effects8.0
Mac OS: Users/[user profile]/Library/Preferences/
Adobe/Adobe After Effects/8.0/
The names of the files are
Adobe After Effects 8.0 Prefs
Adobe After Effects 8.0 Shortcuts
These can be opened with any text editor that doesn’t add
its own formatting and works with Unicode. The default
applications, TextEdit on the Mac and Notepad on Win-
dows, are acceptable, although there are more full-featured
alternatives. Make a safety copy before editing by simply
duplicating the file (any variation in the file name causes it
not to be recognized by After Effects). Revert to the safety
by giving it the original file name should anything start to
go haywire after the edit.
The Shortcuts file includes a bunch of comments at the
top (each line begins with a # sign). The Shortcuts them-
selves are arranged in a very specific order that must be
preserved, and if you add anything, it must be added in the
right place. You can add the line
”NewEffectsLayer” = “(Cmd+Option+Y)”

or on Windows
”NewEffectsLayer” = “(Ctrl+Alt+Y)”

between NewDebugComp and NewLight—this gives you a
shortcut to create a new adjustment layer. If you under-
stand this basic format, you can change other shortcuts
to be what you like. For example, if you don’t like the fact
that Go To Time was changed in CS3 (apparently to align
it with other Adobe applications), search for GoToTime and

130


make your changes to the shortcut in quotes after the =
sign; “(Alt+Shift+J)” becomes “(Ctrl+G)” on Windows,
“(Opt+Shift+J)” becomes “(Cmd+G)” on Mac (but make
sure to change the Group shortcut to something else).
Be extra careful when editing Preferences—a stray charac-
ter in this file can make After Effects unstable. Most of the
contents should not be touched, but here’s one example
of a simple and useful edit (for studios where a dot is
preferred before the number prefix instead of the under-
score): change
“Sequence number prefix” = “_”

to
“Sequence number prefix” = “.”

In other cases, a simple and easily comprehensible numeri-
cal value can be changed:
”Eye Dropper Sample Size No Modifier” = “1”
”Eye Dropper Sample Size With Modifier” = “5”

In many cases the value after the = is a binary yes/no value,
expressed as 0 for no or 1 for yes, so if you’re nostalgic
for how the After Effects render chime sounded in its first
A fantastic script for specifying your
several versions, find own modifier keys called KeyEd
”Play classic render chime” = “0” Up was developed specifically for
After Effects by Jeff Almasol, author
and change the 0 to a 1. Save the file, restart After Effects, of other scripts included with this
and invoke nostalgic memories of past renders. book.

XML
After Effects CS4 projects can be saved as .aepx files. These
are identical to use, but are written in plain Unicode text;
you can edit them with an ordinary text editor. Most of
what is in these files is untouchable, but a few strings, such
as source file locations, are easily recognizable.
This feature was added for one reason only: scriptability.
Anyone capable of writing scripts to, say, swap source files
procedurally (and you know who you are) now has a way
to get at this data without having to open After Effects in
order to do it.

131


This feature may open the door to more easily editable
properties appearing in an .aepx file in future versions of
After Effects, but these files are far from the equivalent
of, say, a Shake script, in which every virtually line is easily
On the Mac: Force a Crash
understood and potentially editable.
One benefit of After Effects has historically been
that it is among the most stable applications in its
category, and when it does crash, it attempts to do Memory Management
so gracefully, offering the option to save before it In OS X, After Effects can see and use up to 3.5 GB of
exits. The new auto-save options, if used properly,
physical memory. Your machine may have more total RAM
further diminish the likelihood that you are ever
likely in danger of losing project data. than this, but most applications on a Mac are still limited
to 32-bit 4 GB address spaces.
For OS X users, there is an extra feature that may On Windows XP, the maximum amount of memory sup-
come in handy if the application becomes unre- ported for a single application is 4 GB (again, using 32-bit
sponsive but does not actually crash. 4 GB address spaces). According to Microsoft, however,
“The virtual address space of processes and applications is
Open Terminal, and enter ps –x (then press still limited to 2 GB unless the /3GB switch is used in the
Return) to list all processes. Scan the resulting list Boot.ini file.” Editing this file is out of the scope for this
for After Effects and note its PID (Process ID) value.
book, so check out microsoft.com for specific information.
Extra memory is most helpful for multiprocessing, because
Now enter kill –SEGV ### where “###” is
replaced by the After Effects PID value. This causes each process has the just described memory potential. With
the application to crash with a save opportunity. multiprocessing, you could use the maximum memory on
an 8 core (or more—only rumored at this writing) system.

Conclusion
You’ve reached the end of Section I (assuming you’re read-
ing this book linearly, that is) and should now have a firm
grasp on getting the most out of the After Effects workflow.
Now it’s time to focus more specifically on the art of visual
effects. Section II, “Effects Compositing Essentials,” will
teach you the techniques, and Section III, “Creative Explo-
rations,” will show you how they work in specific effects
situations.
Now comes the fun part.

132

SEC
SEC

II
Effects Compositing
Essentials
Chapter 5 Color Correction 135
Chapter 6 Color Keying 175
Chapter 7 Rotoscoping and Paint 211
Chapter 8 Effective Motion Tracking 237
Chapter 9 The Camera and Optics 269
Chapter 10 Expressions 309
Chapter 11 32-Bit HDR Compositing
and Color Management 345

CHAPTER

5
Color Correction

Color is my obsession, joy and torment… one day, at
the deathbed of a dear friend, I caught myself in the act
of focusing on her temples, analyzing the succession
of appropriately graded colors which death imposed on
her motionless face.
—Claude Monet

Color Correction

N o skill is as essential for a compositor as the ability to
authoritatively and conclusively take control of color, such
that foreground and background elements seem to inhabit
the same world, shots from a sequence are consistent with
one another, and their overall look matches the artistic
direction of the project.
The compositor, after all, is typically the last one to touch
a shot before it goes into the edit. Inspired, artistic color
work injects life, clarity, and drama into standard (or even
substandard) 3D output, adequately (or even poorly) shot
footage, and flat, monochromatic stills. It draws the audi-
ence’s attention where it belongs, never causing them to
think about the compositing at all.
Good compositors are credited with possessing a “good
eye,” but color matching is a skill that you can practice
and refine even if you have no feel for adjusting images—
indeed, even if you consider yourself color blind.
And despite the new color tools that appear each year to
refine your ability to dial in color, for color matching in
After Effects, three color correction tools allow you to do
most of the heavy lifting: Levels, Curves, and Hue/Satu-
ration (and because Levels and Curves overlap in their
functionality, in many cases you’re just choosing one or
two of the three). These endure (from the earliest days of
Photoshop) because they are stable and fast, and they will
get the job done every time—just learn how to use them,
and keep practicing.

136

II: Effects Compositing Essentials

A skeptic might ask
. Why these old tools with so many cool newer ones?
. Why not use Brightness & Contrast to adjust, you know,
brightness and contrast, or Shadow and Highlight if
that’s what needs adjustment?
. What do you mean I can adjust Levels even if I’m color
blind?
This chapter holds the answers to these questions and
many more. First, we’ll look at optimizing a given image
using these tools, and then we’ll move into matching a
foreground layer to the optimized background, balanc-
ing the colors. The goal is to eliminate the need to hack
at color work and to build skills that eliminate a lot of the
guesswork.
This chapter introduces topics that resound throughout
the rest of the book. Chapter 11 deals specifically with
HDR color, and then Chapter 12 focuses on specific light
and color scenarios, while the rest of Section III describes
how to create specific types of effects shots.

Optimized Levels
What constitutes an “optimized” clip? What makes a color
corrected image correct? Let’s look at what is typically
“wrong” with source footage levels and the usual meth-
The term “plate” stretches back to
ods for correcting them, laying the groundwork for color the earliest days of optical compos-
matching. As an example, we’ll balance brightness and con- iting (and indeed, of photography
trast of a plate image, with no foreground layers to match. itself) and refers to the source foot-
age, typically the background onto
which foreground elements will be
Levels
composited. A related term, “clean
Levels may be the most-used tool in After Effects, and yet plate,” refers to the background
it’s rare to find detailed descriptions of how best to use it. with any moving foreground ele-
ments removed; its usage is covered
It consists of five basic controls—Input Black, Input White,
in the following chapter.
Output Black, Output White, and Gamma—each of which
can be adjusted in five separate contexts (the four individ-
ual image channels R, G, B, and A, as well as all three color
channels, RGB, at once). There are two different ways to
adjust these controls: via their numerical sliders or by drag-
ging their respective carat sliders on the histogram. The
latter is the more typical method for experienced users.

137

Chapter 5 Color Correction

Contrast: Input and Output Levels
Four of the ﬁve controls—Input Black, Input White,
Output Black, and Output White (Figure 5.1)—determine
brightness and contrast, and combined with the ﬁfth,
Gamma, they offer more precision than is possible with the
effect called Brightness & Contrast.
Figure 5.2 shows a Ramp effect applied to a solid using the
default settings, followed by the Levels effect. Move the
black caret at the lower left of the histogram—the Input
Black level—to the right, and values below its threshold
Figure 5.1 Possibly the most used (the numerical Input Black setting, which changes as you
“effect” in After Effects, Levels consists move the caret) are pushed to black. The further you move
of a histogram and five basic controls
the caret, the more values are “crushed” to pure black.
per channel; the controls are typically
adjusted using the triangles on the
histogram, although the correspond-
ing numerical/slider controls appear
below.

Two check boxes at the bottom of
the Levels effect controls specify
whether black and white levels
“clip” on output. These are checked
on by default, and until you work in
HDR (Chapter 11), you might as well
ignore their very existence; they
handle values beyond the range
that your monitor can display.

Figure 5.2 Levels is applied to a layer
containing a Ramp effect at the default
settings, which creates a smooth gradi-
ent from black to white. The spikes
occur simply because the gradi-
ent height does not have an exact
multiple of 256 pixels. The histogram
in Figure 5.1 corresponds to the image
shown here, unadjusted. (Image from
the film “Dopamine,” courtesy of Mark
Decena, Kontent Films.)

138


Move the Input White carat at the right end of the histo-
gram to the left, toward the Input Black caret. The effect is
similar to Input Black’s but inverted: more and more white
values are “blown out” to pure white (Figure 5.3).
Compositing is science as well as
Either adjustment effectively increases contrast, but note art, and so this section employs a
that the midpoint of the gradient also changes if one is useful scientific tool: the control,
which is a study subject that elimi-
adjusted further than the other. In Figure 5.3 Input Black
nates random or hidden variables.
has been adjusted more heavily than Input White, causing This control for the Levels effect is a
the horizon of the gradient to move closer to white and grayscale gradient I generated with
more of the image to turn black. You can re-create this the Ramp effect. You can often set
up similar experiments to clarify
adjustment with Brightness & Contrast (Figure 5.4 on the
your understanding of computer
next page), but to do so you must adjust both contrast graphics applications.

Figure 5.3 Raising Input Black and lowering Input White has the effect of increas-
ing contrast at either end of the scale; at an extreme adjustment like this, many
pixels in an 8 bpc or 16 bpc project are pushed to full white or black or “crushed.”

139


Figure 5.4 You can use Brightness &
Contrast to match the look of Figure
5.3’s gradient, but if you need to shift
the gamma (midpoint) you need a
different tool.

Figure 5.5 The source (left) was balanced for the sky, leaving foreground detail too dark to make out. Raising Brightness to
bring detail out of the shadows makes the entire image washed out (middle); raising Contrast to compensate completely
blows out the sky (right). Madness.

and brightness, with no direct control of the midpoint
(gamma) of the image (Figure 5.5).
Reset Levels (click Reset at the top of the Levels effect
controls) and try the same experiment with Output Black
You can reset any individual effect
control by context-clicking it and and Output White, whose controls sit below the little
choosing Reset. You know it’s indi- gradient. Output Black speciﬁes the darkest black that can
vidual if it has its own stopwatch. appear in the image; adjust it upwards and the minimum
value is raised.
Similarly, lowering Input White is something like dimming
the image, cutting off the maximum possible white value at
the given threshold. Adjust both and you effectively reduce
contrast in the image; with them close together, the gradi-
Note that you can even cross the ent becomes a solid gray (Figure 5.6).
two carets; if you drag Output
White all the way down and Output Evidently the Input and Output controls have the opposite
Black all the way up, you have effect on their respective black and white values, when
inverted your image (although the examined in this straightforward fashion. However, there
more straightforward way to do are even situations where you would use them together.
this typically is to use the Channel:
Invert effect).

140


Figure 5.6 Raising Output Black and lowering Output White reduces contrast in
the dark and light areas of the image, respectively; this doesn’t produce such a
beautiful image but comes into play in the Matching section.

As is the case throughout After Effects, the controls are
operating in the order listed in the interface. In other
words, raising the Input Black level ﬁrst crushes the blacks,
and then a higher Output Black level raises all of those
pure black levels as one (Figure 5.7 on the next page). It
does not restore the black detail in the original pixels; the
blacks remain crushed, they all just become lighter.
If you’re thinking, “So what?” at this point, just stay with
this—the controls are being broken down to build up an
understanding.

141


Figure 5.7 Black and white levels
crushed by adjusting the Input con-
trols aren’t then brought back by the
Output controls, which instead simply
limit the overall dynamic range of the
image, raising the darkest possible
black level and lowering the brightest
possible white.

Brightness: Gamma
As you adjust the Input Black and White values, you
may have noticed the third caret that maintains its place
between them. This is the Gamma control, affecting
midtones (the middle gray point in the gradient) at the
highest proportion and black and white, not at all. Adjust it
over the gradient and notice that you can push the grays in
the image brighter (by moving it to the left) or darker (to
the right) without changing the black and white levels.
Many images have healthy contrast, but a gamma boost
gives them extra punch. Similarly, an image that looks a
bit too “hot” may be instantly adjusted simply by lowering
gamma. As you progress through the book, you will see
that it plays a crucial role not only in color adjustment
but also in the inner workings of the image pipeline itself
(more on that in Chapter 11).

142


In most cases, the histogram won’t itself offer much of a
clue as to whether the gamma needs adjusting, or by how
much (see “Problem Solving using the Histogram,” later in
this chapter for more on the topic). The image itself pro-
Geek Alert: What Is Gamma, Anyway?
vides a better guide for how to adjust gamma (Figure 5.8).
It would be so nice simply to say, “gamma is the
So what is your guideline for how much you should adjust midpoint of your color range” and leave it at that.
gamma, if at all? I first learned to adjust too far before The more accurate the discussion of gamma
becomes, the more obscure and mathematical
dialing back, which is especially helpful when learning. An it gets. There are plenty of artists out there who
even more powerful gamma adjustment tool that scares understand gamma intuitively and are able to
most novice artists away is Curves (more on this later). work with it without knowing the math behind
it—but here it is anyhow, just in case.
By mixing these five controls together, have we covered
everything there is to know about using Levels? No—
Gamma adjustment shifts the midpoint of a color
because there are not, in fact, five basic controls in Levels
range without affecting the black or white points.
(Input and Output White and Black plus Gamma), but This is done by taking a pixel value and raising it to
instead, five times five (RGB, Red, Green, Blue, and Alpha). the inverse power of the gamma value:
newPixel = pixel (1/gamma)

You’re probably used to thinking of pixel values
as fitting into the range 0 to 255, but this formula
works with values normalized to 1. In other words,
all 255 8-bit values occur between 0 and 1, so 0 is
0, 255 is 1, and 128 is .5—which is how the math
“normally” operates behind the scenes.

Why does it work this way? Because of the magic of
logarithms: Any number to the power of 0 is 1, any
number to the power of 1 is itself, and any frac-
tional value (less than 1) raised to a higher power
approaches 0 without ever reaching it. Lower the
power closer to 0 and the value approaches 1,
again without ever reaching it. Not only that, but
the values distribute proportionally, along a curve,
so the closer an initial value is to pure black (0) or
pure white (1) the less it is affected by a gamma
adjustment.

Figure 5.8 The top image is improved with a boost to gamma using Curves
(explained later on); the indication that this is a good idea comes not from
the histogram, which looks fine, but from the image itself, which visibly lacks
foreground detail.

143


Individual Channels for Color Matching
Many After Effects artists completely ignore that pop-up
menu at the top of the Levels control that isolates red,
green, blue, and alpha adjustments, and even those who
do use it once in a while may do so with trepidation; how
can you predictably understand what will happen when you
adjust the ﬁve basic Levels controls on an individual chan-
nel? The gradient again serves as an effective learning tool
to reveal what exactly is going on.
Reset the Levels effect applied to the Ramp gradient once
more. Pick Red, Green, or Blue in the Channel pop-up
menu under Levels and adjust the Input and Output
Same Difference: Levels (Individual Controls)
carets. Color is introduced into what was a purely gray-
Both Levels and Levels (Individual Controls)
accomplish the exact same task. The sole difference scale image. With the Red channel selected, by moving
is that Levels lumps all adjustments into a single Red Output Black inward, you tint the darker areas of the
keyframe property, which expressions cannot use. image red. If you adjust Input White inward, the midtones
Levels (Individual Controls) is particularly useful to and highlights turn pink (light red). If, instead, you adjust
. Animate and time Levels settings individually Input Black or Output White inward, the tinting goes in
. Link an expression to a Levels setting the opposite direction—toward cyan—in the correspond-
. Reset a single Levels property (instead of the ing shadows and highlights. As you probably know, on the
entire effect) digital wheel of color, cyan is the opposite of red, just as
magenta is the opposite of green and yellow is the opposite
Levels is more commonly used, but Levels (Indi- of blue (a sample digital color wheel and a visual guide to
vidual Controls) is sometimes more useful. how levels adjustments operate on individual channels are
included on the book’s disc).
Gradients are one thing, but the best way to make sense
of this with a real image is to develop the habit of studying
footage on individual color channels as you work. This is
the key to effective color matching, detailed ahead.
Along the bottom of the Composition panel, all of the
icons are monochrome by default save one: the Show
Channel menu. It contains ﬁve selections: the three color
channels as well as two alpha modes. Each one has a short-
cut that, unfortunately, is not shown in the menu: Alt+1
through Alt+4 (Option+1 through Option+4) reveal each
color channel in order. These shortcuts are toggles, so rese-
lecting the active channel toggles RGB. A colored outline
around the edge of the composition palette reminds you
which channel is displayed (Figure 5.9).

144


Figure 5.9 Four Views mode is generally intended for 3D use, but it can also be used to show RGB and individual red, green,
and blue channels. This becomes extremely useful for color matching; for now, observe how radically different the three
channels appear, and the colored outline showing which is which.

Try adjusting a single channel of the gradient in Levels
while displaying only that channel. You are back on famil-
iar territory, adjusting brightness and contrast of a gray-
scale image. This is the way to work with individual channel
adjustments, especially when you’re just beginning or if
you are at all color-blind. As you work with actual images
instead of gradients, the histogram can show you what is
happening in your image.

The Levels Histogram
You might have noticed the odd appearance of the histo-
gram for an unadjusted gradient. If you were to try this
setup on your own, depending on the size of the layer to
which you applied Ramp, you might see a histogram that is
ﬂat along the top with spikes protruding at regular inter-
vals (Figure 5.10).

145


Figure 5.10 Strange-looking histograms. A colored solid (top) shows three
spikes, one each for the red, green, and blue values, and nothing else. With Ramp
(bottom) the distribution is even, but the spikes at the top are the result of the
ramp not being an exact multiple of 255 pixels, causing certain pixels to recur
more often than others.

The histogram is exactly 256 pixels wide; it is effectively
a bar chart made up of 256 single pixel bars, each corre-
sponding to one of the 256 possible levels of luminance in
An often overlooked feature of
Levels is that it allows direct adjust-
an 8 bpc image (these levels are displayed below the his-
ment of brightness, contrast, and togram, above the Output controls). In the case of a pure
gamma of the grayscale transpar- gradient, the histogram is ﬂat because luminance is evenly
ency channel (Alpha Channel). distributed from black to white; if spikes occur in that case,
More on this in Chapter 6.
it's merely an indication that the image height in pixels is
not an exact multiple of 256.
In any case, it’s more useful to look at real-world examples,
because the histogram is useful for mapping image data
that isn’t plainly evident on its own. Its basic function is
to help you assess whether any color changes are liable to
help or harm the image. There is in fact no one typical

146


or ideal histogram—they can vary as much as the images
themselves, as seen back in Figure 5.8.
Despite that fact, there’s a simple rule of thumb for a basic
contrast adjustment. Find the top and bottom end of the Moving Input White and Input
RGB histogram—the highest and lowest points where there Black to the edges of the histogram
is any data whatsoever—and bracket them with the Input is similar to what the Auto Levels
Black and Input White carets. To “bracket” them means to effect does. If that by itself isn’t
enough to convince you to avoid
adjust these controls inward so each sits just outside its cor- using Auto Levels, or the “Auto”
responding end of the histogram (Figure 5.11). The result correctors, consider also that they
stretches values closer to the top or bottom of the dynamic are processor intensive (slow) and
range, as you can easily see by applying a second Levels resample on every frame (so the
result is not consistent from frame
effect and studying its histogram. to frame, which isn’t generally
what you want).

Figure 5.11 Here is a perfect case for bringing the triangle controls correspond-
ing to Input Black and Input White in to bracket the edges of the histogram,
increasing contrast and bringing out vibrant colors without losing detail.

Try applying Levels to any image or footage from the disc
and see for yourself how this works in practice. First crush
the blacks (by moving Input Black well above the lowest

147


black level in the histogram) and then blow out the whites
(moving Input White below the highest white value). Don’t
go too far, or subsequent adjustments will not bring back
that detail—unless you work in 32 bpc HDR mode (Chap-
Footage is by its very nature ter 11). Occasionally a stylized look will call for crushed
dynamic, so it is a good idea to
contrast, but generally speaking, this is bad form.
leave headroom for the whites
and foot room for the blacks until Black and white are not at all equivalent in terms of how
you start working in 32 bits per
channel. Headroom is particularly
your eye sees them. Blown-out whites are ugly and can be
important when anything excep- a dead giveaway of an overexposed digital scene, but your
tionally bright—such as a sun eye is much more sensitive to subtle gradations of low black
glint, flare, or fire—enters frame. levels. These low, rich blacks account for much of what
makes ﬁlm look like ﬁlm, and they can contain a surprising
amount of detail, none of which, unfortunately, would be
apparent on the printed page.
The occasions on which you would optimize an image by
raising Output Black or lowering Output White controls
are rare, as this lowers dynamic range and the overall
Many current displays, and in
particular flat-panels and projec- contrast. However, there are many uses in compositing
tors, lack the black detail that can for lowered contrast, to soften overlay effects (say, fog and
be produced on a good CRT monitor clouds), high-contrast mattes, and so on. More on that
or captured on film. The next time later in this chapter and throughout the rest of the book.
you see a projected film, notice
how much detail you can see in the
shadows and compare. Problem Solving Using the Histogram
As you’ve no doubt noticed, the Levels histogram does not
update as you make adjustments. After Effects lacks a panel
equivalent to Photoshop’s Histogram palette, but you can,
of course, apply a Levels effect just for the histogram, if only
for the purposes of learning (as was done in Figure 5.11).
The histogram reveals a couple of new wrinkles in the
backlit shot from Figure 5.5, now adjusted with Levels to
bring out foreground highlights (Figure 5.12). At the top
end of the histogram the levels peak into a spike. This may
indicate clipping and a loss of image detail.
At the other end of the scale is the common result of a
Gamma adjustment: a series of spikes rising out of the
lower values like protruding hash marks, even though
a 16 bpc project prevents quantization. Raising Gamma
stretches the levels below the midpoint, causing them to
clump up at regular intervals. As with crushing blacks and

148


Figure 5.12 In the first instance of Levels, Gamma is raised and the Input White
is brought in to enhance detail in the dark areas of the foreground. The second
instance is applied only to show its histogram.

blowing out highlights—the net effect is a loss of detail,
although in this case, the spikes are not a worry because
they occur among a healthy amount of surrounding
data. In more extreme cases, in which there is no data in
between the spikes whatsoever, you may see a prime symp-
tom of overadjustment, banding (Figure 5.13).

Figure 5.13 Push an adjustment far enough and you may see quanti-
zation, otherwise known as banding in the image. Those big gaps in
the histogram are expressed as visible bands on a gradient. Switching
to 16 bpc from 8 bpc is an instant fix for this problem in most cases.

149


Banding is typically the result of limitations of 8-bit color,
and 16-bit color mode was added to After Effects 5.0 spe-
ciﬁcally to address that problem. You can switch to 16 bpc
by Alt-clicking (Option-clicking) on the bit-depth identi-
ﬁer along the bottom of the Project panel (Figure 5.14)
or by changing it in File > Project Settings. Chapter 11
explains more.

Figure 5.14 An entire project can be toggled from the default 8-bit color mode
to 16-bit mode by Alt-clicking (Option-clicking) the project color depth toggle in
the Project panel; this prevents the banding seen in Figure 5.13.

Perfecting Brightness with Curves
Curves rocks. I heart Curves. The Curves control is particu-
larly useful for gamma correction, because
. Curves lets you fully (and visually) control how adjust-
ments are weighted and roll off.
. You can introduce multiple gamma adjustments to a
single image or restrict the gamma adjustment to just
one part of the image’s dynamic range.
. Some adjustments can be nailed with a single well-
placed point in Curves, in cases where the equivalent
adjustment with Levels might require coordination of
three separate controls.
It’s also worth understanding Curves controls because they
are a common shorthand for how digital color adjustments
are depicted; the Curves interface recurs in many—most—
color correction toolsets.
Curves does, however, have drawbacks, compared with
Levels:
. It’s not immediately intuitive, it can easily yield hideous
results if you don’t know what you’re doing, and there
are plenty of artists who aren’t comfortable with it.

150


. Unlike Photoshop, After Effects doesn’t offer numeri-
cal values corresponding to curve points, making it a
purely visual control that can be hard to standardize.
. Without a histogram, you may miss obvious clues about
the image (making Levels more suitable for learners).
The most daunting thing about Curves is clearly its inter-
face, a simple grid with a diagonal line extending from
lower left to upper right. There is a Channel selector at the
top, set by default to RGB as in Levels, and there are some
optional extra controls on the right to help you draw, save,
and retrieve custom curves. To the novice, the arbitrary
map is an unintuitive abstraction that you can easily use
to make a complete mess of your image. Once you under-
stand it, however, you can see it as an elegantly simple
description of how image adjustment works. You’ll ﬁnd the
equivalent Curves graph to the Levels corrections on the
book’s disc.
Figure 5.15 shows the more fully featured Photoshop
Curves, which better illustrates how the controls work.

Figure 5.15 Photoshop’s more deluxe
Curves includes a histogram, built-in
presets, displays of all channels
together, and fields for input and
output values for a given point on
the curve.

151


Figures 5.16a through d show some basic Curves adjust-
ments and their effect on an image. Figures 5.17a through f
use linear gradients to illustrate what some common Curves
settings do. I encourage you to try these on your own.

Figure 5.16a The source image. Figure 5.16b An increase in gamma above the shadows.

Figure 5.16c A decrease in gamma.

Figure 5.16d Both corrections combined.

Figures 5.16a through d What you see in an image can be heavily influenced by gamma and contrast.

152


Figure 5.17a The default gradient and Curves setting. Figure 5.17b An increase in gamma.

Figure 5.17c A decrease in gamma. Figure 5.17d An increase in brightness and contrast.

Figure 5.17e Raised gamma in the highlights only. Figure 5.17f Raised gamma with clamped black values.

Figures 5.17a through f This array of Curves adjustments applied to a gradient shows the results of some typical settings.

153


More interesting are the types of adjustments that only
Curves allows you to do—or at least do easily. I came to
realize that most of the adjustments I make with Curves
fall into a few distinct types that I use over and over, and so
those are summarized here.
The most common adjustment is to simply raise or lower
the gamma with Curves, by adding a point at or near the
middle of the RGB curve and then moving it upward or
downward. Figure 5.18 shows the result of each. This pro-
duces a subtly different result from raising or lowering the
Gamma control in Levels because of how you control the
roll-off (Figure 5.19).

Figure 5.18 Two equally valid gamma adjustments employ a single point adjust-
ment in the Curves control. Dramatically lit footage particularly benefits from the
roll-off possible in the highlights and shadows.

154


Figure 5.19 Both the gradient itself
and the histogram demonstrate that
you can push the gamma harder, still
preserving the full range of contrast,
with Curves rather than with Levels,
where you face a choice between
losing highlights and shadows some-
what or crushing them.

155


The classic S-curve adjustment, which enhances brightness
and contrast and introduces roll-offs into the highlights
and shadows (Figure 5.20), is an alternative method to get
the result of the double curves in Figure 5.16d.

Figure 5.20 The classic S-curve
adjustment: The midpoint gamma in
this case remains the same, directly
crossing the midpoint, but contrast is
boosted.

Some images need a gamma adjustment only to one end
of the range—for example, a boost to the darker pixels,
below the midpoint, that doesn’t alter the black point and
doesn’t brighten the white values. Such an adjustment
requires three points (Figure 5.21):
. One to hold the midpoint
. One to boost the low values
. One to ﬂatten the curve above the midpoint

Figure 5.21 The ultimate solution to
the backlighting problem presented
back in Figure 5.5: Adding a mini-
boost to the darker levels while leav-
ing the lighter levels flat preserves the
detail in the sky and brings out detail
in the foreground that was previously
missing.

A typical method for working in Curves is to begin with
a single point adjustment to adjust gamma or contrast,

156


then to modulate it with one or two added points. More
points quickly become unmanageable, as each adjust-
ment changes the weighting of the surrounding points.
Typically, I will add a single point, then a second one to
restrict its range, and a third as needed to bring the shape
of one section back where I want it. The images used
for the figures in this section are included as single stills
on the book’s disc for your own experimentation; open
05_colorCorrection.aep to find them.

Just for Color: Hue/Saturation
The third of three essential color correction tools in After
Effects is Hue/Saturation. This one has many individual-
ized uses:
. Desaturating an image or adding saturation (the tool’s
most common use)
. Colorizing images that were created as grayscale or
monochrome
. Shifting the overall hue of an image
. De-emphasizing, or knocking out completely, an indi-
vidual color channel
All of these uses recur in Chapters 12 through 14 to create
monochrome elements, such as smoke, from scratch.
The Hue/Saturation control allows you to do something
you can’t do with Levels or Curves, which is to directly
control the hue, saturation, and brightness of an image.
Chapter 12 details why Tint, not
The HSB color model is merely an alternate slice of RGB Hue/Saturation, is the right tool
color data. All real color pickers, including the Apple and to convert an entire image to
Adobe pickers, handle RGB and HSB as two separate but grayscale.
interrelated modes that use three values to describe any
given color.
In other words, you could arrive at the same color adjust-
ments using Levels and Curves, but Hue/Saturation gives
you direct access to a couple of key color attributes that
are otherwise difficult to get at. To desaturate an image is
essentially to bring the red, green, and blue values closer
together, reducing the relative intensity of the strongest
of them; a saturation control lets you do this in one step,
without guessing.

157


Often is the case where colors are balanced but merely too
“juicy” (not a strictly technical term, but one that may make
sense in this context), and lowering the Saturation value
somewhere between 5 and 20 can be a direct and effective
way to pull an image adjustment together (Figure 5.22).
It’s essential to understand the delivery medium as well,
because ﬁlm is more tolerant and friendly to saturated
images than television.

Figure 5.22 Boosting a saturated image’s contrast can make its saturation a bit
too juiced up with color; if you recognize this, a simple and modest pull-back in
overall Saturation is a quick solution.

158


The other quick ﬁx that Hue/Saturation affords you is a
shift to the hue of the overall image or of one or more of
its individual channels. The Channel Control menu for
Hue/Saturation includes not only the red, green, and
blue channels but also their chromatic opposites of cyan, When in doubt about the amount
of color in a given channel, try
magenta, and yellow. When you’re working in RGB color,
boosting its Saturation to 100%,
these secondary colors are in direct opposition, so that, for blowing it out—this makes the
example, lowering blue gamma effectively raises the yellow presence of that tone in pixels very
gamma, and vice versa. easy to spot.

The HSB model includes all six individual channels,
which means that if a given channel is too bright or over-
saturated, you can dial back its Brightness & Saturation
levels, or you can shift Hue toward a different part of the
spectrum without unduly affecting the other primary and
secondary colors. This can even be an effective way to
reduce blue or green spill (Chapter 6).

More Color Tools and Techniques
This section has laid the foundation for color correction in
After Effects using its most fundamental tools. The truth,
of course, is that there are lots of ways to adjust the color
One alternative usage of these
levels of an image, with new ones emerging all the time. basic color correction tools is to
Alternatives used to create a speciﬁc look—layering in a apply them via an adjustment
color solid, creating selections from an image using the layer, because you can then dial
image itself along with blending modes, and more—are them back simply by adjusting
the layer’s opacity or hold them
explored in Section III of this book.
out from specific areas of the
image using masks or track matte
Color Finesse and Magic Bullet Looks selections.
Color Finesse is a sophisticated color correction system
included with After Effects Professional; because it runs as
a separate application and does not allow you to adjust cor-
rections in the context of a composite, it is more suitable
for overall color look adjustments. It is interesting to notice
that it is in many ways a collection of the tools in this sec-
tion, assembled into one toolset.
An alternative approach can be found in Magic Bullet
Looks from Red Giant Software. This package also operates
via a separate application and is thus more suited to overall
color looks than compositing and matching. It uses a novel
metaphor—that of the actual production pipeline—and

159


includes filters designed to emulate lighting effects, matte
box filters, lens effects, and film stocks.
Much of what these tools do can be recreated with the sim-
pler tools previously described, plus a few extras. However,
most artists, particularly those just learning, find that they
prefer the more intuitive approach of these higher-level
tools for complex color corrections.

Three-Way Color
One thing sorely missing from After Effects is a three-
way color corrector, typically featuring color wheels to
adjust three distinct color ranges: shadows, midtones,
and highlights. Look at any contemporary feature film
or major television show and you’re likely to find strong
color choices that strongly deviate from how the original
scene must have looked: In an ordinary day-lit scene the
shadows might be bluish, the midtones green, and the
highlights orange.
Chapter 12 includes a “roll your A third-party tool, Magic Bullet Colorista from Red Giant
own” recipe and Animation Preset
Software makes this type of correction quick and intuitive
for cheap and cheerful three-way
color correction. in After Effects (Figure 5.23); a demo version is included
on the book’s disc.

Figure 5.23 Red Giant Software’s Colorista adds color wheels to control shadows, midtones, and highlights (or lift, gamma,
gain); this makes it possible to radically transform the color look of an image and achieve the kind of look a colorist might
aim for.

160


Color Matching
Having examined the color correction tools in depth, it’s
now time for the bread and butter work of compositing: to
match separate foreground and background elements such
that the scene appears to have been shot at once.
Although it certainly requires artistry to do well, this is
a learnable skill with measurable objective results. The
process obeys such strict rules that you can do it without
an experienced eye for color. Assuming the background
(or whatever source element you’re matching) has already
been color-graded, you even can satisfactorily complete
a shot on a monitor that is nowhere near correctly cali-
brated, and the result will not even suffer from your own
color blindness, if that’s an issue.
How is that possible?
As with so much visual effects work, the answer is derived
by correctly breaking down the problem. In this case, the
job of matching one image to another obeys rules that can
be observed channel-by-channel, independent of the ﬁnal,
full-color result.
Of course, effective compositing is not simply a question
of making colors match; in many cases that is only the ﬁrst
step. You must also obey rules you will understand from
the careful observation of nature described in the previous
chapter. And even if your colors are correctly matched, if
you haven’t interpreted your edges properly (Chapter 3)
or pulled a good matte (Chapter 6), or if such essential ele-
ments as lighting (Chapter 12), the camera view (Chapter
9), or motion (Chapter 8) are mismatched, the composite
will not succeed.
These same basic techniques will work for other situa-
tions in which your job is to match footage precisely—for
example, color correcting a sequence to match a hero shot
(the one determined to have the right color juju), a pro-
cess also sometimes known as color timing.

161


The Fundamental Technique
Integration of a foreground element into a background
scene often follows the same basic steps:
1. First match overall contrast without regard to color,
using Levels (and in most cases, working only on the
Green channel). When matching the black and white
points, pay attention to atmospheric conditions.
2. Next, study individual color channels and use Levels to
match the contrast of each channel (as needed—not all
images contain so fundamental a color imbalance).
3. Match the color of the midtones (gamma), channel
by channel, using Levels or Curves. This is sometimes
known as gray matching and is easiest when an object in
the background scene is known to be colorless gray (or
something close).
4. Evaluate the overall result for other factors inﬂuencing
the integration of image elements—lighting direction,
atmospheric conditions, perspective, grain or other
ambient movement, and so on (all of which and more
are covered in this book).
The overall approach, although not complicated or even
particularly sexy, can take you to places your naked eye
doesn’t readily understand when it is looking at color. Yet,
when you see the results, you realize that nature beats logic
every time.
The sad truth is that even an experienced artist can be
completely fooled by the apparent subjectivity of color.
Figure 5.24 shows an example in which seeing is most deﬁ-
Figure 5.24 There are no yellow dots
in the top image, and no blue dots
nitely not believing. Far from some sort of crutch or nerdy
in the middle image; the four dots detail, channel-by-channel analysis of an image provides
shown in the bottom image are iden- fundamental information as to whether a color match is
tical to their counterparts in the other
within objective range of what the eye can accept.
two images.

Ordinary Lighting
We begin with a simple example: inserting a 3D element
lit with ordinary white lights into a daylight scene. As you
can see in Figure 5.25, the two elements are close enough
in color range that a lazy or hurried compositor might be
tempted to leave it as is.

162


Figure 5.25 An unadjusted
foreground layer (the plane)
over a day-lit background.

With only a few minutes of effort, you can make the plane
look as though it truly belongs there. Make sure the Info
palette is somewhere that you can see it, and for now,
choose Percent (0–100) in that palette’s wing menu to
have your values line up with the ones discussed here (you
can, of course, use whatever you want, but this is what I’ll
use for discussion in this section).
This particular scene is a good beginner-level example
of the technique because it is full of elements that would
appear monochromatic under white light; next we’ll move
on to scenes that aren’t so straightforward. The back-
ground is dominated by colorless gray concrete, and the
foreground element is a silver aircraft.
Begin by looking for suitable black and white points to use
as references in the background and foreground. In this
case, the shadow areas under the archways in the back-
ground and underneath the wing of the foreground plane
are just what’s needed for black points—they are not the
very darkest elements in the scene, but they contain a simi-
lar mixture of reﬂected light and shadow cast onto similar
surfaces, and you can expect them to fairly nearly match.
For highlights, you happily have the top of the bus shelter
to use for a background white point, and the top silver
areas of the plane’s tail in the foreground are lit brightly
enough to contain pure white pixels at this point.

163


Figure 5.26 The target highlight and shadow areas for the foreground and back-
ground are outlined in yellow; levels corresponding to each highlight (in Percent
values, as set in the panel menu) are displayed in the adjacent Info palette.

Figure 5.26 shows the targeted shadow and highlight
regions and their corresponding readings in the Info
palette. The shadow levels in the foreground are lower
(darker) than those in the background, while the back-
ground shadows have slightly more red in them, giving the
background warmth absent from the unadjusted fore-
ground. The top of the plane and bus shelter each contain
levels at 100%, or pure white, but the bus shelter has lower
blue highlights, giving it a more yellow appearance.
To correct for these mismatches, apply Levels to the
foreground and move the Output Black slider up to about
7.5%. This raises the level of the blackest black in the
image, lowering the contrast.

164


Having aligned contrast, it’s time to balance color. Because
the red levels in the background shadows are higher than
blue or green, switch the Composition panel to the red
channel (click on the red marker at the bottom of the
panel or use the Alt+1/Option+1 shortcut), causing a thin
red line to appear around the viewer. You can now zoom in
on an area that shows foreground and background shad-
ows (Figure 5.27).

Figure 5.27 Evaluate and match black and
white levels on individual channels. In this
case the image is “green matched,” with the
RGB adjustment first made for the green
channel, then the red and blue channels are
adjusted individually. These subtle adjust-
ments are all that’s needed in this case to
match the element.

Black levels in the red channel are clearly still too low in
the foreground, so raise them to match. Switch the Chan-
nel pop-up in Levels to Red, and raise Red Output Black
slightly to about 3.5%. You can move your cursor from fore-
ground to background and look at the Info palette to check
whether you have it right, but the great thing about this
method is that your naked eye usually evaluates variations
in luminance correctly without the numerical reference.
Now for the whites. Because the background highlights
have slightly less blue in them, switch to the blue channel
(clicking the blue marker at the bottom of the Composi-
tion panel or using Alt+3/Option+3). Pull back slightly to

165


where you can see the top of the bus shelter and the back
of the plane. Switching Levels to the blue channel, lower
the Blue Output White setting a few percentage points to
match the lower blue reading in the background. Back in
RGB mode (Alt+3/Option+3 toggles back from blue to
RGB), the highlights on the plane take on a more sunlit,
yellow quality. It’s subtle, but it seems right.
What about the midtones? In this case, they’re taking care
of themselves because both the foreground and back-
ground are reasonably well balanced and these corrections
The human eye is more sensitive
to green than red and blue. Often,
are mild.
when you look at a shot channel by Figure 5.28 displays the result, with the same regions tar-
channel, you will see the strongest
brightness and contrast in the geted previously, but with the levels corrected. To add an
green channel. For that reason, extra bit of realism, I also turned on motion blur, without
a sensible approach to matching yet bothering to precisely match it (something you will
color may be to get the overall learn more about in Chapter 8). You see that the plane is
match in the ballpark so that the
green channels match perfectly,
now more acceptably integrated into the scene.
and then adjust the other two Work on this composite isn’t done either; besides match-
channels to make green work. That
way, you run less risk of misadjust- ing the blur, you can add some sun glints on the plane as
ing the overall brightness and it passes, similar to those on the taxi. On the other hand,
contrast of your footage. you can tell that the blur on the plane is too heavy for the
pilot’s absence from the cockpit to be noticeable, a good
example of how an initial pass at a composite can save a lot
of extra work.

Figure 5.28 This is a better match,
particularly in the shadow areas;
motion blur helps sell the color adjust-
ment as well.

166


Dramatic Lighting
Watch a contemporary feature film objectively for color
and you may be shocked at how rare ordinary day-lit scenes
such as the plane example are. Dramatic media—not just
films but television and theater—use color and light to
create mood, to signify key characters and plot points, and
more. If you’re working with a daring cinematographer, or
even heed the advice in the Foreword, you’ll be happy to
know that this matching technique is even more impressive
with strong lighting.
The composite in Figure 5.29 clearly does not work; the
foreground element does not even contain the scene’s
dominant color, and is white-lit. That’s fine; it will better
demonstrate the effectiveness of this technique.
This section discusses colors
It helps that both the foreground and the background expressed as percentages; to see
elements have some areas that you can logically assume to the same values in your Levels
effect, use the wing menu of the
be flat gray. The bridge has concrete footings for the steel Info palette to choose Percent for
girders along the edges of the road, while the can has areas the Color Display.
of bare exposed aluminum.

Figure 5.29 Not only is it clear that the can does not belong in the color environment of the background, the mismatch is
equally apparent on each color channel. (Image courtesy of Shuets Udono via Creative Commons license.)

167


The steps to color match a scene like this are as follows:
1. Apply Levels to the foreground layer.

2. Switch the Comp view to Green (Alt+2/Option+2). Not
only is this the dominant color in this particular scene,
but it is dominant in human vision, so green-matching
is the ﬁrst step in most scenes, not just this one.
3. Begin as if you are looking at a black-and-white photo-
graph, and match the element to this dark contrasty scene
using Levels in the RGB channel. If the element needs
more contrast in the shadows and highlights, as this one
does, raise Input Black and lower Input White; if it needs
less, adjust the Output controls instead. Finally, adjust the
gamma; in this scene, should it come down to match the
darkness of the scene or up so the element stands out
more? The end result should look like a monochrome
photo whose elements match believably (Figure 5.30a).
4. Switch the view (Alt+1/Option+1) and the Levels control
to the Red channel and repeat the grayscale match-
ing process. Clearly, the foreground element is far too
bright for the scene. Speciﬁcally, the darkest silver areas
of the can are much brighter than the brightest areas of
the concrete in the background. Therefore, adjust the
gamma down (to the right) until it feels more like they
inhabit the same world. Now have a look at the high-
lights and shadows; the highlights look a little hot, so
lower Red Output White (Figure 5.30b).
5. Now move over to Blue in the view (Alt+3/Option+3)
and in Levels. In this case, there is almost no match
whatsoever. The can is much brighter and more washed
out than the background. Raise Input Blue and bring
gamma way down. Now the can looks believably like it
belongs there (Figure 5.30c).
It’s strange to make all of these changes without ever look-
ing at the result in full color. So now, go ahead and do that.
Astoundingly, that can is now within range of looking like
it belongs in that scene; the remaining adjustments are
subjective. If you want the can to pick up a little less green
from the surroundings as I did, lower Green Input White.
Back in the RGB channel, adjust Gamma according to how

168


Figures 5.30a, b, and c (top to bottom) It’s fun and satisfying to pull off an extreme match like this
channel by channel. The Levels settings come from looking for equivalent black/white/midpoints in
the image and just analyzing whether the result looks like a convincing black-and-white image on
each channel.

169


much you want this element to pop. And of course ﬁnish
compositing it: Defocus slightly with a little fast blur, add a
shadow, and you start to believe it (Figure 5.31).

Figure 5.31 The result includes a
subtle shadow that has been color
matched as well as a final adjustment
to the white contrast.

When There’s No Clear Reference
The previous examples have contained fairly clear black,
white, and gray values in the foreground and background
elements. Life, of course, is not always so simple.
Figure 5.32 is a scene that lacks any obvious gray values to
match; the lighting is so strong, it’s hard to tell what color
anything in the scene was originally, or whether there were
any neutral black, white, or gray items.

Figure 5.32 Sometimes a source
scene will have completely crazy light-
ing. Once you are confident about
how to match it, you may say to an
image that is blown out and overbal-
anced in one direction, “bring it on.”
(Image courtesy of Jorge L. Peschiera
via Creative Commons license.)

170


Figure 5.33 This one requires as much intuition as logic, but the channel-by-channel approach completely works.

The technique still works in this case, but it may require
more in the way of trial and error, or artist’s intuition.
Looking at each individual color channel, only green is
even close to a plausible match right off the bat; the red
channel contains blown-out whites, and the blue channel is
so dark (and grainy) it hardly exists.
Once again, just try to get the brightness and contrast
adjusted, working channel by channel, and you get an
initial result something like Figure 5.33. Considering how
It’s often a good idea to take a break
subjective the adjustments are by necessity in this case, this when trying to finalize fine color
isn’t half bad; and ﬁne adjustments to the RGB channel adjustment. When you come back,
can bring it where it needs to go. even to a labored shot, you regain
an immediate impression that can
The ability to match color without seeing an image in full save you a lot of noodling.
color is so powerful that it can seem almost magical the
ﬁrst few times you try it. Why, then, do so few artists work
this way? I would have to say that laziness and ignorance
are the main culprits here. Switching channels seems like
a pain, and few untrained artists clearly realize that color
works like this.

171


Direction and Position
These examples use simple 8-bit 3D elements rendered on
a single pass. The skill of dealing with these is a basic skill
of compositing—color matching—but it’s not as if, having
mastered this skill, you are prepared to match anything
anywhere. An ideal element generated in 3D software
would of course have multiple passes, giving you more
control (Chapter 12); even short of that, if the lighting and
perspective of an element are wrong, there’s no practical
way to make it match (Figure 5.34).

Figure 5.34 All of the 2D composit-
ing trickery in the world can’t change
the fact that this element is angled
wrong. It is also lit from the wrong
side. (Source clip from “Jake Forgotten,”
courtesy of John Flowers.)

On the other hand, compositing has consistently freed art-
ists from spending too long hanging around in 3D software
trying to solve everything. Figure 5.35 shows the simplest
solution to the previous problem: match the camera angle
and basic lighting by observing what’s in the scene. From
looking at the pool balls and shadows, it seems apparent
that there are a couple of overhead lights nearby and that
the one off camera right is particularly strong.

Figure 5.35 The angle and lighting
have been roughly matched in 3D;
rather than tweaking it further there,
work on getting a quicker and more
accurate result in 2D.

172


I match the angle by placing the background shot into the
background of the 3D software’s camera view, and I make
sure that there are a couple of lights roughly matched to
that of the scene so that they produce the correct shading
and specular highlights. From there, I have an element
that does not match perfectly, but I’m done with what I
need to do in 3D if I want to be done.
More complex and dynamic perspective, interactive light-
ing, animation, global illumination, and so on certainly
make more things that can be done in 3D, yet at the end
of the day, the smart computer graphics artist kicks a
scene over to 2D as soon as the elements are within range
(Figure 5.36).

Figure 5.36 The color-matched final
includes a shadow.

Gamma Slamming
Maybe you’ve seen an old movie on television—the exam-
ple I think of first is Return of the Jedi (before the digital
re-release)—in which you see black rectangular garbage
mattes dancing around the Emperor’s head, inside the
cloak, that you obviously shouldn’t be seeing. Jedi was made
prior to the digital age, and some of the optical composites
worked fine on film, but when they went to video, subtle-
ties in the black levels that weren’t previously evident sud-
denly became glaringly obvious.
Don’t let this happen to you! Now that you know how to
match levels, put them to the test by slamming the gamma
of the image. The simplest way to do this is just to adjust
the Exposure control at the lower right of the Viewer up
or down. You can also make an adjustment layer, set it as a

173


guide layer so that it cannot render, and add a Curves or
Levels effect to push gamma up or down (different than
exposure—you may prefer it).
Slamming (Figure 5.37) exposes areas of the image that
might have been too dark to distinguish on your monitor; if
the blacks still match with the gamma slammed up, you’re
in good shape. Similarly, slamming up will give you a clear
glimpse of grain matching (and slamming the alpha, even
more so). Slamming down can reveal whether highlight lev-
els match; it’s particularly useful if you’re working in linear
HDR with over-range values (Chapter 11).

Figure 5.37 Slamming gamma is like
shining a bright light on your scene. In
this case it reveals the need for grain
and a mismatch in the shadow color.

This is a useful habit anywhere that there is a danger of
subtle discrepancies of contrast; you can use it to examine
a color key, as you’ll learn in the next chapter, or a more
extreme change of scene lighting. Every big effects studio
I’ve worked at examines footage this way before it’s sent
for ﬁnal.

Conclusion
This chapter has covered some of the basics for adjusting
and matching footage. Obviously there are exceptional sit-
uations, some of which occur all of the time: depth cueing,
changes in lighting during the shot, backlighting, interac-
tive light, and shadow. There are even cases in which you
can, to some degree, relight a shot in After Effects, intro-
ducing light direction, exchanging day for night, and so
on. You’ll discover more in Chapter 12.

174

Slow down, I’m in a hurry.
—Franz Mairinger (Austrian Equestrian)

Color Keying

C olor keying was devised in the 1950s as a clever means
to combine live-action foreground footage and back-
grounds that could come from virtually anywhere. What
was once a fragile and expensive proposition is now fully
mainstream; whole films such as 300, now rely on this
technique, while the Colbert Report invites anyone with a
computer to try the “Green Screen Challenge” (and runs
entries from none less than John Knoll).
The process goes by many names: color keying, blue
screening, green screening, pulling a matte, color dif-
ferencing, and even chroma keying—a term from analog
color television, the medium defined by chroma and heav-
ily populated with weather forecasters.
The purpose of this chapter is to help you not only with
color keying of blue- and green-screen footage but with all
cases in which pixel values (hue, saturation, and/or bright-
ness) stand in for transparency, allowing compositors to
effectively separate the foreground from the background
based on color data.
All of these methods extract luminance information that
is then applied to the alpha channel of a layer (or lay-
ers). The black areas become transparent, the white areas
For those reading nonlinearly, this
chapter extends logically from fun- opaque, and the gray areas gradations of semi-opacity; it’s
damental concepts about mattes the gray areas that matter.
and selections in Chapter 3.

Good Habits and Best Practices
Before we get into detail about specific keying methods
and when to use them, I’ll share some top-level advice to
remember when creating any kind of matte.

176


Figure 6.1 The background influences what you see. Against black, almost no detail is visible (left). Checkerboard reveals
shadows (middle), but flaws in the matte are clearest with a bright, solid, contrasting background (right). (Source footage
courtesy of Pixel Corps.)

. Introduce contrast. Use a bright, saturated, contrasting
background (Ctrl+Shift+B/Cmd+Shift+B) such as yel-
low, red, orange, or purple (Figure 6.1). If the fore-
ground is to be added to a dark scene, a dark shade is
okay, but in most cases bright colors better reveal matte
problems. Solo the foreground over the background
you choose.
. Protect edge detail. This is the name of the game (and
the focus of much of this chapter); the key to winning
is to isolate edges as much as possible and focus just on
them so as to avoid crunchy, chewy mattes (Figure 6.2).

Figure 6.2 A matte like this could
be called “chewy,” and is typically the
result of clamping the foreground or
background (or both) too far.

. Keep adjustments simple, and be willing to start over.
Artists spend hours on keys that could be done more
effectively in minutes, simply by beginning in the right
place. There are many complex and interdependent

177

Chapter 6 Color Keying

steps involved with creating a key; if you’re hung up on
one, it’s time to try a different approach.
. Constantly scan frames and zoom into detail. When
possible, start with a tricky area of a difficult frame;
look for motion blur, fine detail, excessive color spill,
and so on, and keep checking various areas in various
modes (Figure 6.3).

Figure 6.3 A glimpse of the alpha
channel can reveal even more
problems, such as faint holes in the
foreground, which should be solid
white.

. Break it down into multiple passes. This is the single
most important concept novices miss. Most mattes
will benefit from separate garbage, core, and edge
passes—a process detailed later in this chapter—and
in many cases it helps to create a separate pass just for
delicate edges: hair, a translucent costume, motion
blur, and so on.
I encourage you to review this list again once you’ve
explored the rest of this chapter.

178


Linear Keyers and Hi-Con Mattes
There are cases in which edge detail is less of a factor
because the matte is used to adjust, not layer, an element;
for example, you could hold out the highlight areas of an
image for adjustment using a high-contrast (hi-con) matte.
You might create this matte with a linear keyer.
Linear keyers are relatively simple and deﬁne a selection
range based on a single channel only. This could be it red,
green, or blue, or just overall luminance. They’re useful
in a wide variety of cases outside the scope of blue- and
green-screen shots, although similar principles apply with
Keylight. (Keylight is covered later in this chapter.)
The most useful linear keyers are
. Extract
. Linear Color Key
The keyers to avoid are
. Luma Key
. Color Key
because each is limited to a bitmap (black and white) selec-
tion. Only by choking and blurring the result with Edge
Thin and Edge Feather controls can you add threshold
adjustment.

The Extract and Linear Color Key
Extract is useful for luminance (luma) keying, because it uses
the black and white points of an image or any of its individ-
ual channels. Linear Color Key is a more appropriate tool
When, Exactly, Is Linear Keying Useful?
to use to isolate a particular color (or color range).
Keying using a single channel (or the average of
multiple channels) is useful to matte an element
Extract using its own luminance data, in order to hold out
The Extract key includes a histogram to help you isolate specific portions of the element for enhancement.
For example, you duplicate a layer and matte its
thresholds of black and white; these are then graded with
highlights to bloom them (see Chapter 12).
black and white softness settings. You can work with aver-
aged RGB luminance, or you can access histogram controls
for each of the color channels.

179


Figure 6.4 When extracting an image using luminance, it’s better to examine all three channels and choose the one that is
closest to the matte you want than to simply rely on averaged luminance.

One of the three color channels nearly always has better
deﬁned contrast than overall luminance, which is merely
an average of the three. Either green or red is typically the
brightest and most contrasty channel, while blue generally
All Channels Are Not Created Equal has higher noise (Figure 6.4).
If you set an RGB image as a luma matte, the red,
Extract is interactive and easy to use and is a cousin to Lev-
green, and blue channels are averaged together
to determine the luminance of the overall image. els. Its histogram shows the likely white or black thresholds
However, they are not weighted evenly, because on each channel. You bring in the White Point or Black
that’s not how the eye sees them. Details about how Point (the upper of the two small square controls below
to work with this fact can be found in Chapter 12.
the histogram) and then threshold (soften) that adjust-
ment with the White Softness or Black Softness controls
If you find yourself wanting to use a particular (the lower of the two small squares).
channel as a luma matte, use Effect > Channel >
Shift Channels; set Take Alpha From as Full On and
Linear Color Key
the other three channels to whichever channel—
red, green, or blue—is most effective. The Linear Color Key offers direct selection of a key color
using an eyedropper tool. The default 10% Matching

180


Softness setting is arbitrary and deﬁnes a rather loose
range. I often end up with settings closer to 1%.
Note that there are, in fact, three eyedropper tools in the
Linear Color Key effect. The top one deﬁnes Key Color,
and the other two add and subtract Matching Tolerance. I
tend not to use these because they don’t work in the Comp
viewer; the main Key Color eyedropper and the Matching
sliders work for me (Figure 6.5).

Figure 6.5 Suppose you wish to make a change to the distant, out-of-focus area of this shot. By selecting a prominent
object in the foreground, the sweater, and adjusting the selection, you can create a matte that separates the foreground
and background. (Image courtesy of Eric Escobar.)

There’s a hidden trick to getting better results with Linear
Color Key. Because it is linear, it will pick up hues that
seem unrelated. To reduce the effect of these, you can
add a second instance of Linear Color Key. Under Key

181


Operation, changing the default Key Colors setting to
Keep Colors does nothing if it’s the first instance except
annul the effect. On the second instance, Keep Colors is
unaffected by the first instance and can bring back hues
that were already keyed. The one-two punch will often
deliver the best result (Figure 6.6).

Figure 6.6 By adding a second instance
of Linear Color Key set to Keep Colors, I’m
able to get rid of the extra selection areas
in the background, apply the matte, and
add a glow effect that influences only the
background.

Difference Mattes
A difference matte is simple in principle: Frame two shots
identically, the first containing the foreground subject, the
other without it (commonly called a clean plate). Compare
the two images and remove everything that matches identi-
cally, leaving only the foreground subject. It sounds like
the type of thing a computer was built to do.
In practice, of course, there are all sorts of criteria that
preclude this from actually working very well, specifically
. Both shots must be locked off or motion stabilized to
match, and even then, any offset—even by a fraction of
a pixel—can kill a clean key.
. The foreground element may be rarely entirely
unique from the background; low luminance areas, in

182


particular, tend to be hard for the Difference Matte
effect to discern.
. Grain, slight changes of lighting, and other real-world
variables can cause a mismatch between two otherwise
identical shots. Raising the Blur Before Difference
setting helps correct for this, but only by introducing
inaccuracy.
To try this for yourself, begin with a locked-off shot con-
taining foreground action, ideally one in which a char-
acter enters the frame. Duplicate the layer, and lock off
an empty frame of the background using Layer > Time >
Freeze Frame. Apply Difference Matte to the top layer.
Adjust Tolerance and Softness; if the result is noisy, try rais-
ing the Blur Before Difference value.
Figure 6.7 shows the likely result of attempting to key this
footage using only Difference Matte. It’s not a terrible way
to isolate something when clean edges are not critical, but
it cannot compare with more sophisticated methods for
removing a solid color background.

Figure 6.7 I was hoping to grab just the shadows on the floor with a difference matte applied to the image on the left
using the middle image as a Difference Layer; unfortunately, subtle stuff like that tends to be indistinguishable from noise,
even with clean, low-grain source.

Luma Mattes
The honest truth is that I don’t use the above effects very
often; there are less immediately obvious but more power-
ful ways to generate transparency from color data:
. Blending modes
. Blending modes with color adjustments
. Shift channels and levels
In the interest of full disclosure, I often use track mattes
(as detailed in Chapter 3) with a duplicate of the layer,

183


instead of luminance keys applied directly to the layer.
With a track matte, I can use Levels to work with color
ranges on all channels in order to reﬁne transparency.
Other artists may consider this approach more cumber-
some, preferring to work with a single layer and an effect,
so the choice as always is yours.

Blue and Green Screen Keys
Keylight is useful in many keying situations, not just studio-
created blue- or green-screen shots. For example, you can
use Keylight for removal of a murky blue sky (Figure 6.8).
You wouldn’t use Keylight to pull a luminance key, however,
or when you’re simply trying to isolate a certain color range
within the shot; its effectiveness decreases the further you
get from the three primary colors.
Keylight is most typically used on footage shot against a
uniform, saturated, primary color background, where pres-
ervation of edge detail is of utmost importance.

Figure 6.8 Keylight knocks out the background in one pass, with no adjustments, despite low saturation in the sky.

Figure 6.9 outlines the basic keying process detailed in
the following pages. No two complex shots are the same,
but something like this approach should help you pull a
good matte, regardless of the tools used. The next section
reviews these steps, speciﬁcally using Keylight.

184


Figure 6.9 This chart summarizes
steps that can yield a successful
color key.

The basic steps are
1. Garbage matte any areas of the background that can
easily be masked out. “Easily” means you do not have an
articulated matte (you don’t animate individual mask
points). As a rule of thumb, limit this to what you can
accomplish in about 20 minutes or less (Figure 6.10).

Figure 6.10 The quick-and-dirty garbage matte isolates the action; you don’t waste time on areas of the frame that aren’t
overlapped by action.

185


2. Attempt a first pass quickly, keeping this matte on the
loose side (preserving as much edge detail as possible)
to be refined later.
3. Preview this at full resolution, in full motion, against
a bright primary color. In rare cases, you’re done, but
before you throw up your arms in victory, carefully
examine the alpha channel of the toughest frames
(usually the ones with the most motion, reflection, or
semi-opaque detail). Note any obvious holes in the fore-
ground or areas of the background that have failed to
disappear, as well as any noise in the solid areas of the
matte (Figure 6.11).

Figure 6.11 The shadows need to be
eliminated with a hard matte; they go
up the wall and off set. A hard matte
negatively influences fine detail such
as motion blur and hair, any of which
may require its own holdout matte.

If things look noisy and chaotic in the alpha channel or
the edges are clamped and chewy, you can
. Start over and try a new pass
. Apply noise reduction to the plate, then start over
(see “Noise Suppression” later in this chapter)
. Articulate or track garbage/holdout mattes to iso-
late problem portions of the footage (see “Beyond
Keylights: Better Mattes,” later in this chapter)
4. If necessary, separate the plate for multiple passes. At
the very least, it’s often useful to create a solid core and
a completely transparent background so that you can
focus only on the edge (detailed in the next section).
You may also need to separate individual parts of a layer
for a separate keying pass, such as hair or a fast-moving,
motion blurred limb.

186


5. Refine the edge. Zoom in on a challenging area of the
foreground edge (200% to 400%), and refine the key
to try to accommodate it, using strategies outlined in
the steps that follow. Challenging areas may include
. Fine detail such as hair
. Motion blurred foreground elements
. Cast shadows
You must also watch out for, and consider rotoscoping,
foreground features that can threaten an effective key,
such as
Holdout Mattes
. Areas of the foreground that reflect the background A holdout matte isolates an area of an image for
color separate treatment. I recommend that you think
of a color key as an edge matte surrounded by
. Edge areas whose color nearly matches the back- two holdout mattes: one for the core, one for the
ground background.

. Areas of poor contrast (typically underlit regions of
the shot)
6. Preview the shot in full motion. Again, note holes and
noise that crop up on individual frames, and use the
strategies outlined in the section “Beyond Keylights:
You can follow along with the steps
Better Mattes” to overcome these problems. Approach- using the blueScrn_mcu_HD.mov
es you may add at this stage are footage included in this chapter’s
Pixel Corps folder on the book’s disc.
. More holdout mattes (typically masks), either for Completed versions of several of
the purpose of keying elements individually or ro- the examples from this chapter are
toscoping them out of the shot (Figure 6.12) contained in the 06_colorKey.aep
project.
. Isolation of the matte edge, for the purpose of refin-
ing or blurring it (see “Matte Problems” later in this
chapter)

Figure 6.12 Holdout mattes seem
tedious to create and animate, but for
true tedium try spending hours keying
a shot in one pass that could have
been broken down into a series of 10
minute roto masks.

187


Figure 6.13 Even with a well-shot, high-definition source (left), it is imperative that you get the Screen Colour setting right
to preserve all of the transparent detail in this shot. Choosing a darker background color (center) creates a more solid initial
background, but a lighter color selection (right) preserves far more detail. (Source footage courtesy of Pixel Corps.)

Keylight
The first decision in Keylight is most important: sampling
a color for the Screen Colour setting (Keylight reveals
its UK origins at Framestore CFC with that u). Specific
tips to do so are up ahead in the section, “Generate the
Screen Matte.”
In the best-case scenario, you create any necessary garbage
mattes and then follow these steps:
1. Use the Screen Colour eyedropper to sample a typical
background pixel. View defaults to Final Result so you
get a matte instantly; set the background to a bright
The Eyedropper and the Info Panel
color and solo the plate layer, or examine the alpha
With the Info panel active, sample a pixel in After
Effects, either using an eyedropper tool or simply channel (Alt+4/Option+4).
by moving your cursor around a viewer. The Info 2. If in doubt about the Screen Colour setting, turn the
panel then displays the value that belongs to the
exact pixel underneath the selection point of the effect off and set another instance, repeating as neces-
cursor (the lower-left corner of the eyedropper, the sary until you have one that eliminates the maximum
upper-left corner of the pointer). unwanted background (Figure 6.13); you can then
delete the rest.
As in Photoshop, the After Effects eyedropper Now, as needed, look for areas to refine.
samples only the color state of a pixel; transparency
of any given pixel is always evaluated as if 100%. 3. Switch View to Status. Opaque pixels are displayed as
white, transparent pixels are black, and those contain-
ing any amount of transparency are gray (Figure 6.14).
It’s an exaggerated view of the alpha channel matte.

188


Figure 6.14 What looks like a good initial Combined Matte (left) turns out to have a lot of semi-transparent pixels where
you ideally want full opacity or full transparency in Status view (right).

4. Still in Status view, you have the option to try Screen
Balance at settings of 5.0, the default 50.0, and 95.0,
although Keylight will preselect it based on your back-
ground color selection. This setting controls the weight-
ing between the primary matte color (blue, green, or
red) and each of the secondaries.
5. If the background is not solid black, you can boost
Screen Gain until the gray mostly disappears in Status
view, although the ideal is not to raise this value at all
Keylight's built-in spill suppression
(the next section explains how). Use this setting as can and will enhance the appear-
sparingly as possible. ance of grainess, particularly with
4:2:2 or another compressed source.
6. Optionally, set the Despill Bias using the eyedropper. The workaround (detailed ahead) is
Sample an area of the foreground that has no spill and to apply the keyed layer as a track
should remain looking as is (typically a bright and satu- matte instead.
rated skin tone area). I very rarely do this (see Notes).
The rare perfect footage is now completely keyed. If it isn’t
perfect (Figure 6.15), this is a decision point; how can you
best divide this matte into multiple passes?

Figure 6.15 Not bad for a single pass,
although just between you and me,
this matte has some light foreground
holes that need closing. A multi-pass
approach will improve even a simple
matte like this.

189


The Apple Shake-based implementation of Keylight adds
inputs for a garbage matte and a holdout matte. This
method of isolating and focusing on the edge inspired
me to devise the following workflow to achieve the same
result in an After Effects precomp, with as few extra steps
as possible.

The Three-Pass Method
This breakdown will work with any software keyer, but it’s
especially helpful if you’re trying to get the most out of
Keylight.
1. Rename the Keylight effect by selecting it and pressing
the Return key. Rename it Keylight gMatte.
The first goal is to create a background matte that is
100% transparent and outside of all edge detail by sev-
eral pixels. To do so, follow these steps:
a. In Keylight Status or Combined Matte view (de-
pending on your preference), raise Clip Black just
enough to remove gray pixels from the background
in Status view, then lower the Clip White setting just
above the same number (Figure 6.16).

Figure 6.16 The garbage matte, or
gMatte, has been crushed; Clip Black
has been raised just enough to elimi-
nate the noise in the lower-left corner
(from 6.14, right), Clip White has been
lowered near the same number for a
hard matte. This is then spread with a
negative Simple Choker setting and
the layer blending mode has been set
to Stencil Alpha to isolate the exterior
of the edge.

b. Important step: set View to Intermediate Result.

c. If necessary, choke this matte further to eliminate
tracking markers and other small bits of noise. You
can lower the Screen Shrink/Grow setting (under
Screen Matte in Keylight) or apply a Simple Choker
effect and raise the Choke Matte number. Quick
and dirty is fine here.

190


d. Whether or not you’ve choked the matte, the most
essential step is to spread it until it’s well outside
the foreground edge. You can apply Simple Choker
and lower the Choke Matte setting to a negative
value, quite possibly in the double digits, even using
a second instance if the maximum setting of –100 is
reached without completely exceeding the fore-
ground boundary.
e. Garbage matte any items in the background that
have not disappeared and hand-animate the matte
as needed (refer back to Figure 6.12).
You should see a generous blue (or green) buffer
around the foreground edge (Figure 6.17).

Figure 6.17 Eliminating non-
overlapping areas of the background
with a garbage matte is step one.

2. Duplicate the layer and rename it Core. Reset the
Simple Choker effects and solo the layer.
This matte ﬁlls in all holes inside the matte edge with-
out coming close to overlapping that edge. To create
this layer
a. In Keylight Status or Combined Matte view (your
preference), lower Clip White until the foreground
is completely opaque. Raise Clip Black to a value
just below the same number.
b. If necessary, spread this matte to eliminate fore-
ground holes by raising the Screen Shrink/Grow
setting (under Screen Matte in Keylight). You
can also lower the Choke Matte setting in Simple
Choker to a negative value.

191


c. Whether or not you’ve spread the matte, choke it
until it’s well inside the foreground edge. If there
was no spread operation, use Screen Shrink/Grow
in Keylight; otherwise, use Simple Choker.
d. Add a layer and garbage matte; any remaining holes
resulting from extra-stubborn shiny, reﬂective cos-
tumes or props.
e. Soften this layer as needed with a Fast Blur; with
all three layers on, toggle the cMatte on and off
to make sure it has little or no effect on the edge
(Figure 6.18).

Figure 6.18 The core matte isn’t
pretty; it’s job is to sit within the edge
and fill holes. Here it is blurred to
assure that any threshold with the
actual edge is not noticeable.

3. Apply a second instance of Keylight on the layer con-
taining the gmatte. If you like, rename this effect Key-
light Edge. You have isolated the edge to the maximum
extent possible (Figure 6.19).

Figure 6.19 The final result allows an
Edge matte that has no adjustments
to Keylight whatsoever: a maximally
subtle edge.

Now just work on that edge layer—starting over with its
key, if you want, because it’s now a much simpler problem
that demands the lightest possible touch. Your goal should
be to use three controls only to reﬁne it: Screen Colour,

192


Clip Black, and Clip White. If possible, do not adjust
Screen Gain.
This is not the furthest you can go to break down a color
keyed matte; problematic footage will require further steps,
described in the following section.

Get the Best Out of Keylight
You’re no doubt familiar with the areas where you want
to take a good look when you’re evaluating the quality of
a matte:
. Hair detail: Are all of the “wispies” coming through?
(Figure 6.20a)
. Motion blur or, in unusual cases with a defocus, lens
blur: Do blurred objects appear chunky or noisy, or do
they thin out and partially disappear? (Figure 6.20b)
. Screen contamination areas: Do holes remain in the
foreground? (Figure 6.20c)
. Shadows: Are they keying as desired (usually all or
nothing)? (Figure 6.20d)

Figures 6.20a through d Fun chal-
lenges you may encounter when
pulling a color key include wispy hair
(a), motion blur (b), contamination
of foreground elements by the back-
ground color (c), and shadows (d).

A B

C D

193


There may also be fundamental problems, including
. Ill-defined foreground/background separation or
semitransparency throughout the foreground or
Switch a RAM Preview to Alpha background
Channel view (Alt+4/Option+4); . Crunchy, chewy, or sizzling edges (other terms include
the cache is preserved (a note for
those who remember when this
boiling, which is more often applied to roto, and chunky
was not the case). or clumpy fine detail)
. Noise
. Excessive fringing or choking
. Errors in the spill suppression
There are specific tools to deal with these; before getting
to those, it’s a good idea to make sure you’re not fighting
the fundamentals.

Generate the Screen Matte
A few decisions are absolutely essential in Keylight and the
rest are essentially compensatory. I have seen artists spend
literally days on keys that sadly could be improved in under
an hour with a different approach.
The core of Keylight is screen matte generation, and the
most essential step is choosing the exact color to key. From
that, Keylight makes weighted comparisons between its
saturation and hue and that of each pixel, as detailed in
Table 6.1. From this, you see that the ideal background is
distinct and saturated.

TABLE 6.1 How Keylight Makes Its Key Decisions
COMPARED TO SCREEN COLOR, PIXEL IS KEYLIGHT WILL
Of a different hue Consider it foreground, making it
opaque
Of a similar hue and Key it out completely, making it
more saturated transparent
Of a similar hue, but Subtract a mathematically weighted
less saturated amount of the screen color and make
it semitransparent

194


This section details the other tools that contribute to the
screen matte:
. Screen Gain emphasizes the saturation of the back-
ground pixels The controls atop Keylight (from
Screen Colour down to Alpha Bias)
. Screen Balance delineates the background hue from differ fundamentally from the rest;
the other primary colors. they work together to generate the
actual matte, while the rest adjust
. Bias is for color-correction.
the result of that operation.
Screen Gain
Before I describe Screen Gain, a disclaimer: The ideal
Screen Gain setting is 100, no change to the default. This
adjustment is compensation for a poorly lit matte, and
while raising it may make the matte channel look better,
you are also likely to see increased color grain and lost
edge detail with values above the default.
A clear symptom that the background lacks sufficient
intensity is that areas of the background are of a consistent
hue yet fail to key easily. Similarly, a clear symptom that
your foreground is contaminated with reflected color from
the background is that it appears semi-opaque. Screen
Gain is designed to help in these cases.
Screen Gain boosts (or reduces, although I’ve never used
a setting lower than the default) the saturation of each
pixel before comparing it to the screen color. This effec-
tively brings more desaturated background pixels into the
A Rosco Ultimatte Blue screen con-
keying range. tains quite a bit of green—much
A major point of holding out the edge using the three-pass more than red, unless improperly
lit. Ultimatte Green screens,
method described above is to avoid having to use Screen meanwhile, are nearly pure green
Gain whatsoever; it can be useful to quickly refine a crude (Figure 6.21).
one-pass key.

Figure 6.21 The Rosco colors:
Ultimatte Blue, Ultimatte Green, and
Ultimatte Super Blue. Blue is not pure
blue, but double the amount of green,
which in turn is double the amount
of red. Ultimatte Green is more pure,
with only a quarter the amount of red
and no blue whatsoever. Lighting can
change their hue (as does converting
them for print in this book).

195


Screen Balance
Keylight is designed to expect one of the three RGB color
values to be the dominant background color. It is even
more effective, however, if it knows whether one of the
two remaining colors is more prevalent, and if so, which.
Screen Balance compensates for a dominant secondary
background color.
On this theory, you would employ a balance of 95% with
blue screens and leave it at 50% for green screens, and in
version 1.2 of Keylight, that is exactly how the plug-in sets
Screen Balance depending on whether you choose a blue
or green Screen Colour setting. I hardly touch this control
because Keylight added this automatic adjustment in ver-
sion 1.1.
The generalized recommendation from The Foundry is to
set it “near 0, near 100, and compare” these to the default
setting of 50 to evaluate which one works best. In other
words, imagine there are three settings (instead of 100)
and try 5%, 50%, and 95%.

Bias
The Bias settings are Despill Bias and Alpha Bias. These
color correct the image in the process of keying by scaling
the primary color component up or down (enhancing or
reducing its difference from the other two components).
The Foundry recommends that in most cases you leave
Alpha Bias at the default and that you click the Despill Bias
eyedropper on a well-lit skin tone that you wish to pre-
serve; despill pivots around this value.
Like Screen Gain, Bias has an unpleasant side effect: it
can signiﬁcantly increase the graininess of keyed footage.
There are other despill methods available (later in this
chapter) that don’t add this complication.

Refinement
When you spot an area that looks like a candidate for
reﬁnement, save (to hold an undo point should you need

196


to use File > Revert), zoom in, and create a region of inter-
est around the area in question.
Now take a look at the tools provided by Keylight to
address some common problems. Hold down Alt/Option to center a
zoom around your cursor.
Holes and Edges
The double-matte method (core and edge) shortcuts a lot
of the tug of war that otherwise exists between a solid fore-
ground and subtle edges. Even with this advantage, both
mattes may require adjustments to the Clip White or Clip
Black controls.
Keep the largest possible difference (or delta, if you prefer)
between these two settings, as this is where all of your gray,
semitransparent alpha pixels live. The closer the two num-
bers get, the closer you are to a bitmap alpha channel, in
which each pixel is pure black or white—a very bad thing
indeed (Figure 6.22).

Figure 6.22 Here’s how the hair looks
without a separated edge matte.
Not nice.

If you push too far, you can try restoring back toward the
initial matte with Clip Rollback. Its value is the number
of pixels from the edge that are rolled back relative to
the original, unclipped screen matte. So if your edges
were subtle but sizzling on the ﬁrst pass, but removing
noise from the matte hardened them, then this tool may
restore subtlety.

197


Noise Suppression
For seriously sizzling mattes, Keylight includes a Screen
Pre-blur option that I would reserve for footage with a
clearly evident noise problem, such as heavy compression.
Blurring source footage before keying adds inaccuracy and
is something of a desperation move. The footage itself does
not appear blurred, but the matte does.
A better alternative for a fundamentally sound matte is
Screen Softness, under the Screen Matte controls. This
Chroma Subsampling: The 411 on control blurs the screen matte itself, so it has a much better
4:1:1, 4:2:2, and 4:2:0
chance of retaining detail than a pre-blur approach. As
Video images are RGB on your computer, but
video devices themselves use Y'CrCb, the digital
was shown in Chapter 3, edges in nature are slightly soft,
equivalent of YUV. Y' is the luminance or brightness and a modest amount of softness is appropriate even with a
signal (or “luma”); Cr and Cb are color-difference perfectly healthy matte.
signals (roughly corresponding to red-cyan and
blue-yellow)—you could call them chrominance
or “chroma.”

It turns out that the human eye is much more
particular about gradations in luma than chroma,
as is amply demonstrated in Figure 6.23.

The standard types of digital video compression
take advantage of this fact. Figure 6.24 shows
the difference between straight RGB and 4:2:2
compression, which is common to popular formats
including DVCPRO HD and DVCPRO50, ProRes
422 and cameras such as the Sony F900, as well
as 4:1:1, which is used by DVCPRO and NTSC DV.
Almost as bad for keying purposes is 4:2:0, the
MPEG-2 (DVD), HDV, and PAL DV format.

As you might imagine, chromatic compression is Figure 6.23 The source (top-left image) is then converted to YUV and blurred
far less than ideal for color keying (Figure 6.25), 100 pixels, first just the UV or chroma (lower-left image) and then just the Y
hence the workarounds in this section. or luma (right). Even very heavy Chroma Blur can be accepted by the eye—it
even adds a softness to the highlights—whereas heavy luma blur is completely
unacceptable.

Figure 6.24 4:4:4 is just pixels, no
chroma subsampling, where 4:2:2 and
4:1:1 groups the nearest neighbor-
ing pixels, giving them identical
luminance according to the patterns
shown here.

198


Figure 6.25 Key a 4:1:1 image (left) and Keylight’s Status view (right) clearly shows the horizontal blocks associated with
that type of chroma subsampling (images are zoomed 4x).

The Despot cleanup tools are meant to ﬁll matte holes, but
at sufﬁciently high levels they add blobbiness, so they are
rarely useful. An alternative approach, particularly with DV
formats (which, by the way, are guaranteed to add com-
pression noise and are not recommended for blue-screen
and green-screen work), is to
1. Convert the footage to YUV using Channel Combiner
(the From pop-up menu). This will make the clip look
very strange, because your monitor displays images as
RGB. Do not be alarmed (Figure 6.26).

Figure 6.26 This is how an image
converted to YUV should look on an
RGB monitor—weird. The point is
not how YUV looks, but what you can
do to adjust it before using Channel
Combiner to round-trip it back to RGB.

199


2. Apply Channel Blur to the green and blue channels
only, at modest amounts (to gauge this, examine each
channel as you work—press Alt+2/Option+2 or Alt+3/
YUV is the digital version of the Option+3 while zoomed in on a noisy area). Make sure
broadcast video color space. It is Repeat Edge Pixels is checked.
used in component PAL television
and is functionally similar to YIQ, 3. Round-trip back from YUV to RGB, using a second
the NTSC variant. In After Effects instance of Channel Combiner.
YUV, the red channel displays the
luminance value (Y) of the shot, 4. Apply Keylight.
while the green and blue channels
display blue and red weighted Fringing and Choking
against green (U and V).
Fringing (excess edge opacity) and choking (lost edge
detail) are the ultimate tests of an otherwise problem-free
matte.
Screen Grow/Shrink deals with this issue directly; ideally
this control would be used for Garbage and Core passes
(described earlier) but not the Edge itself; otherwise it may
be a symptom of a more fundamental problem, an indica-
tion that it’s time to start over.
It’s also not the last resort for choking and spreading a
matte; alternatives follow in “Beyond Keylights: Better
Mattes.”

Spill Suppression
Keylight suppresses color spill (foreground pixels contami-
nated by reﬂected color from the background) as part
of the keying operation. Thus spill-kill can be practically
automatic if you pull a good initial key.
There are a surprising number of cases in which Keylight’s
spill suppression is not what you want, for the following
reasons:
. Dramatic hue shifts occur to items whose colors are
anywhere near green (for example, cyan) or opposite
green (for example, magenta). It’s challenging enough
to keep green off of a green set, let alone its neighbor-
ing and opposite hues.
. These hue shifts can also add graininess, even to foot-
age that was shot uncompressed and has little or no
source grain.

200


Figure 6.27 Her face doesn’t even look the same without the highlights reflected with the green. Even worse, at this magni-
fication, it’s easy to see that the amount of grain noise has increased significantly. It’s a definite case for pulling the matte on
one pass and applying spill suppression separately.

In Figure 6.27, notice how the whole shape of the girl’s
face seems to change due to the removal of highlights via
spill suppression.
Should Keylight’s spill suppression become unwieldy or
otherwise useless for the preceding reasons, there is an
easy out: ordinarily View is set to Final Result, but set it to
Intermediate Result for the matte applied to the Alpha
without any change to RGB. The CC Composite effect does
the same thing, eliminating all RGB changes from preced-
ing effects but keeping the alpha.
Keylight itself also includes spill suppression tools, under
Edge Colour Correction, that influence only the edge
pixels. Enable its checkbox and adjust the controls below,
softening or growing the edge as needed to increase the
area of influence. Sometimes adjusting Luminance or Satu-
ration of edges is a quick fix.
There are better ways than Keylight to kill spill, detailed in
the next section, which moves beyond this tool.

Beyond Keylight: Better Mattes
So, to recap, the number one thing you can do to improve
a matte is to break it down into component parts:
. Garbage, Core, and Edge mattes (isolate the threshold
areas)
. Holdout masks for problematic areas of the frame (big
hair, fast motion, varied lighting, and shadow)

201


. Temporal split (if light conditions change as the shot
progresses)
All other tricks fall short if you’re not willing to take the
trouble to do this. Here are other steps to take to ensure a
good matte.

On Set
Your time is often considered cheaper than that of a full
crew on set—in many cases, justifiably so. That means—
shocking, I know—you’ll be fixing things in post that
shoulda coulda woulda been handled differently on set. If,
on the other hand, you have the opportunity to supervise
on set, knowledge is power.
A hard cyclorama, or cyc (pronounced like “psych”) is far
preferable to soft materials, especially if the floor is in shot.
If you can’t rent a stage that has one, the next best thing
might be to invest in a roll of floor covering and paint it,
to get the smooth transition from floor to wall, as in Figure
6.28 (assuming the floor is in shot).
Regarding the floor, don’t let anyone walk across it in
street shoes, which will quickly contaminate it with very
visible dust. There are white shoe-cover booties often
used specifically to avoid this, and you can also lay down
big pieces of cardboard for the crew to use setting up. Be
pedantic about this if you’re planning to key shadows.

Figure 6.28 On a set with no hard
cyclorama, you can create the effect of
one—the curve where the wall meets
the floor—using a soft blue-screen
instead. It doesn’t behave as well (note
the hotspot on the curve), but it will
certainly do in a pinch and is much
preferable to removing the seam
caused by the corner between the
wall and floor.

202


Lighting is, of course, best left to an experienced Director
of Photography (D.P.) or gaffer (one who has shot effects
before is hugely helpful), and any kind of recommenda-
tions for a physical lighting setup are beyond the scope of
this book. Because you’ll spend more time examining this
footage than anyone else, here are a few things to watch
for on set:
. Light levels on the foreground and background should
have matching intensity. A spot light meter tells you if
they do.
. Diffuse lights are great for the background (often a set
of large 1 K, 2 K, or 5 K lights with a silk sock covering
them, Figure 6.29), but fluorescent lights will do in a
pinch, on a smaller set. With fluorescents you just need
more instruments to light the same space. Kino Flo
lights are a popular option for smaller sets as well.

Figure 6.29 The larger the set, the
more diffuse white lights you’ll see in
the grid, to eliminate hotspots in the
background.

The Right Color
The digital age lets shooters play fast and loose
with what they consider a keyable background.
You will likely be asked (or attempt) to pull mattes
. Maintain space between the foreground and back- from a blue sky, from a blue swimming pool (like
I did for Pirates of the Caribbean), or from other
ground. Ten feet is ideal. monochrome backgrounds.
. Avoid unintentional shadows, but by all means light
for shadows if you can get them and the floor is clean.
How different must the background color be from
Note that this works only when the final shot also has a the foreground? The answer is “not as much as you
flat floor. probably think.” I have had little trouble keying a
girl in a light blue dress or a soldier in a dress blue
. Where possible, avoid having talent sit, kneel, or lie
uniform. This is where it can be hugely helpful to
down directly on the floor or any other keyable surface; have any type of capture device on set—even a
point and shoot camera—and pull a test matte.

203


not only does an astonishing wash of shadow and reﬂec-
tion result, but there is no realistic interaction with the
surface, which is especially noticeable if they are to end
up on carpet, or grass, or the beach. If possible, use real
surfaces and furniture in these cases.
. Here’s a novel idea: Shoot exteriors outside where pos-
sible, forgoing the cyc and controlled lighting environ-
ment for chromatic tarps and the sun, which is a hard
lighting source to fake.
. Record as close to uncompressed as possible. Even
prosumer HD cameras these days almost always have
an HDMI port that outputs live, uncompressed signal;
Shoot a lot of reference of the set,
anything and everything you can pair this with a workstation or laptop containing a video
think of. If you plan to recreate the capture card and high-speed storage and you can get
lighting, it’s also a great idea to take 4:2:2 or better practically for free.
HDR images using bracketed expo-
sures—the same image shot at . Shoot clean plate: a few frames of the set only, particu-
various f-stops. Photoshop includes larly on a locked-off shot and each time a new set-up
the File > Automate > Merge to occurs.
HDR function to combine these into
a 32 bpc linear light image. In this day and age of quick camera to laptop transfer, it’s
great to have the means on-set to pull test comps; they not
only help ensure that the result will key properly, they give
the D.P. and talent a better idea of where they are, and
where they can lead to more motivated light from the D.P.
and more motivated action from the talent, who otherwise
must work in a void.

Matte Problems
There are speciﬁc tools to help you manipulate transpar-
ency once a matte exists.
Minimax is powerful but imprecise, operating as it does
in whole pixel increments. It provides a quick way to
spread or choke pixel data, even without alpha channel
information (it can also operate on individual channels of
luminance).
Simple Choker allows you to choke or spread alpha chan-
nel data (via a positive or negative number, respectively)
at the sub-pixel level (in other words, use decimal values).
That’s all it does. You can push it hard and even use more
than one instance if the 100 pixel limit gets in your way, as
it can with garbage mattes.

204


Matte Choker looks and even sounds more deluxe than
Simple Choker (something about that that “Simple” label)
but it merely adds softness controls that can just compli-
cate a bad key. I use Simple Choker most of the time, and I
don’t feel like a simpleton for doing so.

Edge Selection
Earlier in this chapter the three-pass method for deriving
a matte was introduced; using this method is one way to
derive an edge matte. However, the simplest means to this
end is probably as follows:
1. Apply Shift Channels. Set Take Alpha From to Full On
and all three color channels to Alpha.
2. Apply Find Edges (often mistaken for a useless psyche-
delic effect because, as with Photoshop, it appears in
the Stylize menu). Check the Invert box for an edge A useful third-party alternative
to Minimax is Erodilation from
highlighted in white.
ObviousFX (www.obviousfx.com).
Minimax is useful to help choke or spread this edge It can help do heavier choking
(eroding) and hole-filling (dilat-
matte. The default setting under Operation in this ing), and its controls are simple and
effect is Maximum, which spreads the white edge intuitive (choose Erode or Dilate
pixels by the amount speciﬁed in the Radius setting. from the Operation menu and the
Minimum chokes the edge in the same manner. If the channel—typically Alpha).
result appears a little crude, an additional Fast Blur
will soften it (Figure 6.30).

Figure 6.30 An edge matte can
be used to blur background and
foreground together, or to match
the intensity and saturation to the
background. The matte can itself be
softened with a blur, and Minimax, set
to Maximum and Color, can be used
to grow the matte by increasing the
Radius setting.

3. Apply the result via a luma matte to an adjustment
layer. You should not need to precomp before doing so.

205


You can then use Fast Blur to soften the blend area between
the foreground and background, which often works better
than simply softening a chewy matte. A Levels adjustment
will darken or brighten the composited edge to better blend
it. Hue/Saturation can be used to desaturate the edge, simi-
lar to using a gray edge replacement color in Keylight.

Close Holes
When holes open up in your foreground or background,
the best ﬁx is usually to take care of these in the individual
Garbage and Core passes; in fact, hole-closing is the main
point of the three-pass method described earlier. The basic
method is as follows:
1. Choke (Garbage matte) or Spread (Core matte) the
holes until they disappear.
2. Spread or Choke (the opposite of the previous step) an
equivalent or greater amount.
This will of course destroy all edge subtlety, which is why it
only works well on a matte intended to be crude and inside
or outside the delicate foreground edges.
Alpha Cleaner, part of the Key Correct set from Red Giant
Software, has an automated tool to do this without destroy-
ing edges, but it can have unintended consequences, ﬁlling
small gaps that should remain transparent (Figure 6.31).

Figure 6.31 Mind the gaps; choking
and spreading a matte, or using tools
to do so automatically, such as the
third-party Key Correct tools, is liable
to close small holes like this; they may
require their own special hold-out
mattes.

206


Therefore, there are occasions where you have to roto-
scope, usually because the holes are too close to the edge.
It can help to create an animated mask (Chapter 7) or
track in a paint stroke (Chapter 8).

Color Spill
If Keylight (or another tool) is not doing spill suppression
as you would like, the following alternatives are available:
Effect > Keying > Spill Suppressor uses a simple channel
multiplication formula to pull the background color out of
the foreground pixels. All you need to do is select a sample
from the background color (you can sample the Keylight
Screen Colour selection) and leave the setting at 100%.
Effect > Color Correction > Hue/Saturation is the option
I use most nowadays. It takes practice, which can easily be
done; try the following on a source blue or green screen
image (no key—just so you can see everything):
1. Apply Effect > Color > Hue/Saturation.

2. Under Channel Control, choose your primary (Greens
or Blues—plural for some reason).
3. Lower the Saturation value for that channel to zero.

4. Use the Channel Range sliders to eliminate the key
color.
5. Boost Saturation to 100% to see if other colors are be-
ing desaturated that should be kept; adjust accordingly.
6. Now with the key applied, try some mixture of the fol-
lowing to eliminate spill:
. Lower Saturation (still on the individual color
channel).
. Shift Hue between about 30 and 90 degrees
to either direction (depending on your target
background).
Some combination of desaturation and hue shift with a
carefully targeted range usually does the trick; the worst
that will happen is that other color artifacts will appear,

207


in which case the trick is usually to go back and adjust the
range again.
The Channel Range is the most important part of this,
which is why I suggest cranking Saturation all the way down
and all the way up to see it clearly. The inside rectangular
sliders are the range, the outside, triangular sliders are
the threshold area; I tend to be generous with the latter
(Figure 6.32).

Figure 6.32 Isolate the Greens chan-
nel in Hue/Saturation to avoid killing
neighboring colors along with green
spill suppression. Expand the core
and threshold hash marks to a wider
range, and then both hue-shift and
desaturate the result. Desaturation
affects the core colors, and the hue
shifts those in the threshold, which are
thrown more to the range that would
otherwise be missing.

There may be cases where it is difﬁcult or impossible to
avoid affecting some part of a costume or set with spill
suppression; for example, a cyan colored shirt will change
color when the actor is corrected for green. The above
method should let you work around this better than most
of the automated tools, especially Keylight itself, but there
are cases where you might have to use loose roto to isolate
the costume and bring its color back.
To correct only the edges, see the section above about how
to isolate the edge and use it as a track matte.

208


Conclusion
Primatte Keyer (included as a demo in the book’s disc)
is a viable alternative to Keylight, especially because it
uses an entirely different internal model to create a key.
Whereas Keylight must, by design, be used only for blue,
green or red, Primatte is a geometric keyer that treats R,
G and B (or H, S and B) internally as a three-dimensional
array of color—a box of hues. As you add and subtract
from the key selection, a geometric blob grows and shrinks
inside that box. You can’t see this “blob” but the colors
fully within it are keyed, and it has an outer threshold
region as well. It takes some practice to master, but once
you do, you should ﬁnd more success getting a good key
on one pass (forgoing the three-pass method discussed
earlier) with Primatte than with Keylight.
The next chapter offers hands-on advice for situations
where procedural matte generation must be abandoned in
favor of hand matte generation, also known as rotoscoping.

209

CHAPTER

7
Rotoscoping and Paint

It’s a small world, but I wouldn’t want to paint it.
—Steven Wright

Rotoscoping and Paint

R otoscoping is something many artists strive to avoid,
yet denial can be costly; although usually quicker and
easier, procedural solutions can bog down. Adding roto
into the mix solves the problem straightforwardly. Rotoscop-
ing (or roto) is simply the process of adjusting a shot frame
by frame, generally using masks. Cloning and filling using
paint tools are variations on this task.
After Effects is not famed as a rotoscoping tool, but the
truth is that there is a lot of ignorance as to how to use it
most effectively for this purpose. Not only can the masking
“Rotoscoping” was invented by Max
Fleischer, the animator responsible and previewing tools be made less cumbersome if applied
for bringing Betty Boop and Popeye correctly, but you can often combine paint and roto with
to life, and patented in 1917. It tracking and keying.
involved tracing over live-action
movement, a painstaking form of Here are some overall guidelines for roto and paint:
motion capture. The term has come
to stand for any frame-by-frame . Keep it simple. In particular, pay attention to strategies
manipulation of a moving image. using the fewest possible number of keyframes.
. Consider your options in the following order of diffi-
culty and automation: keying, tracking, animated masks,
paint. Paint is almost always slower and more painstak-
ing than animating masks and should be reserved for
times when it’s truly the most expedient method.
. Review constantly at full speed (taking care to use the
setup described next).
. Combine strategies. A color key or hi-con matte may
take your matte 80 to 90% of the way there, so that
roto merely completes the job. Stabilizing an element
prior to rotoscoping it can save a lot of extra work. The
tracker can be combined with individual adjustments
rather than relying only on one or the other.
It can be satisfying to knock out a seamless rotoscope, espe-
cially when you avoid fighting the tools.

212


Articulated Mattes
An “articulated” matte is one in which individual mask
points are adjusted to detail a shape in motion.
Each adjustment that you make to an animated mask “Keyframing” began at Disney, in
requires the entire frame to redraw, whether in the Comp the 1930s, where top animators
or Layer viewer. Good rotoscoping requires eliminating would create the key frames—the
top of the heap, the moment of
this delay, as details are clear only when seen in motion. impact—and lower-level artists
Doing so is simple: would add the in-between frames
thereafter.
1. Create a comp containing only the source plate.

2. Add a solid layer above the plate.

3. Turn off the solid layer’s visibility.

4. Lock the plate layer.

5. Select the solid layer and draw the ﬁrst mask shape,
then press Option+Shift+M/Alt+Shift+M to keyframe it.
Now any changes you make to the masked solid have no
effect on the plate layer or the cache; you can RAM Pre-
view the entire section and it is preserved, as is each frame
There’s one major downside to
as you navigate to it and keyframe it. When it comes time masking on a layer with its visibility
to apply the masks, you can either apply the solid as a track off: You cannot drag-select a set
matte or copy the masks and keyframes to the plate layer of points (although you can Shift-
itself. Genius! select each of them).

Speciﬁc strategies make effective rotoscoping of complex
organic shapes possible, regardless of tools:
. Use multiple overlapping masks instead of a single
mask on a complex, moving shape (Figure 7.1).

Figure 7.1 It is crazy to mask a
complex articulate figure with a single
mask shape; the sheer number of
points will have you playing whack-
a-mole. Separated segments let you
focus on one area of high motion
while leaving another area that moves
more steadily more or less alone.

213

Chapter 7 Rotoscoping and Paint

One key to working quickly is to intentionally create
overlapping, separate mask shapes for different parts
of your masked area that you know will move inde-
pendently, or be revealed and concealed (in this case,
because of a simple perspective shift). A shape with 20
or more points will be difficult to manage over time.
. When working with multiple masks, you can lock, hide,
or show unselected masks to get them out of the way
(and bring them back): context-click > Mask > Lock
Always make certain that
Preferences > General > Preserve Other Masks/Hide Locked Masks/Show All Masks.
Constant Vertex Count when Editing . Define a shape beginning on a frame where it can be
Masks is toggled on, as it is by drawn with the fewest possible points, adding more
default.
points as needed as you go. As a rule of thumb, no
articulated mask should contain more than a dozen or
so points.
. Go through the clip and look for the natural keyframe
points, where a change of direction, speed, or shape
begins or ends, and block in keyframes before you ani-
mate them.
. To replace a Mask shape, in Layer view, select the shape
from the Target menu and start drawing a new one;
whatever you draw replaces the previous shape. Beware:
The first vertex point of the two shapes may not match,
creating strange in-between frames. To set the first ver-
tex point, context-click on any mask vertex.
. Use a mouse—a pen and tablet system makes exact
placement of points difficult.
. Use the arrow keys to move points. The increments
change according to the zoom level of the current view.
. Select a set of points and use the Tranform Box to
offset, scale, or rotate them together instead of moving
them individually (Figure 7.2). Most objects shift per-
spective more than they fundamentally change shape,
Enable Cycle Mask Colors in
and this method uses that fact to your advantage.
Preferences > User Interface Colors
to generate a unique mask color Block in and refine the first shape, then refine as neces-
with each new mask. You can sary. Check the result at full speed against a contrasting
customize the color if necessary
background, just as you do with color keys. Then move on
to make it visible by clicking its
swatch in the Timeline. to the next shape.

214


Figure 7.2 Select and double-click a
set of mask vertices and manipulate
them together using a transform box,
offsetting its center point as needed.

Rotobeziers
Rotobezier shapes are designed to animate a mask over
time; they’re like Bézier shapes (discussed in Chapter 3)
without the handles, which means less adjustment and
less chance of pinching or loopholes when points get
close together (Figure 7.3). Rotobeziers aren’t universally
beloved, partly because it’s difﬁcult to get them right in one
pass; adjoining vertices change shape as you add points.

Figure 7.3 Overlapping Bézier handles result in kinks and loopholes (left); switching the mask to Rotobezier (right) elimi-
nates the problem.

215


Activate the Pen tool (G key) and check the Rotobezier
box in the Tools menu, then click the layer to start drawing
points; beginning with the third point, the segments are,
by default, curved at each vertex.
You can freely toggle a shape from
Bézier to Rotobezier mode and The literal “key” to success with Rotobeziers is the Alt/
back, should you prefer to draw Option key. At any point as you draw the mask, or once
with one and animate with the you’ve completed and closed it by clicking on the ﬁrst
other.
point, hold Alt/Option to toggle the Convert Vertex
tool . Dragging it to the left increases tension, and makes
the vertex a sharp corner, like collapsed Bézier handles.
Drag in the opposite direction, and the curve rounds out.
You can freely add or subtract points as needed by
toggling the Pen tool (G key).
The real advantage of the Rotobezier is that it’s impossible
to kink up a mask as with long overlapping Bézier handles;
other than that, Rotobeziers are essentially what could be
If the Selection tool (V) is active,
Ctrl+Alt/Cmd+Option activates
called “automatic” Béziers (Figure 7.4). By drawing enough
the Adjust Tension pointer. Bézier points to keep the handles short, however, you may
ﬁnd that you don’t need them.

Figure 7.4 You can carefully avoid crossing handles with Béziers (left); convert this same shape to Rotobeziers (right) and
you lose any angles, direction, or length set with Bézier handles.

Beyond Built-in Limitations
As mentioned earlier, masks in After Effects haven’t evolved
as a feature in quite some time. Here are some common
limitations, followed by tips to work around them.

216


. After Effects has no built-in method for applying a
tracker to a mask. There is, however, a way to do this
thanks to at least one third-party script (credited in the
front of the book): TrackerViz, detailed in the following
chapter, lets you convert trackers to mask vertices.
. You can’t translate multiple mask keyframes at once;
even if you select multiple keyframes, the move is
Preferences > General includes
applied only at the current frame. a preference, on by default, to
. After Effects lacks the ability to specify whether a preserve a constant vertex count
feather is applied to the inside or outside of a mask, when editing masks. Practically
the only reason to disable this is if
nor can you vary feather settings on a per-vertex basis you’re creating keyframes on every
(Figure 7.5). If you really need this ability, check out PV frame, with no in-betweens.
Feather from RE:Vision Effects.

Figure 7.5 After Effects mask feather controls the entire mask; the PV Feather plug-in allows you to use two masks to con-
trol the inside and outside feather.

. Adding points to an animated mask has no adverse
effect on adjacent mask keyframes. Delete a point,
however, and it is removed from all keyframes, usually
deforming them.
. There is no dedicated morphing tool in After Effects.
The tools to do a morph, however, do exist, along with
solid deformation tools generally.

217


The following sections elaborate upon the above points
in depth.

Tracking and Translating
There seems to be no way to motion track a mask (other
than the aforementioned TrackerViz) nor to move a set
of Mask Path keyframes together. Here are some work-
arounds that help when a mask needs to move:
. If movement of a masked object is from camera motion
and occurs in the entire scene, you can essentially
stabilize the layer, animate the mask in place, and then
reapply motion to both. Chapter 8 goes into more
detail about this.
. To translate an entire animated mask, you can make
the masked layer a track matte applied to the same
layer (without masks); you can then track or translate
the entire track matte.
Mask shapes can be linked together directly with expres-
sions. Alt/Option-click the Mask Path stopwatch, then use
the pickwhip to drag to the target Mask Path. Only a direct
link is possible, no mathematical or logical operations, so
all linked masks behave like instances of the ﬁrst.

Mask Motion Blur
To mask motion blur would seem to be something of a
nightmare, as its edges defy careful observation. It’s not
easy by any means, but masks respect the motion blur
settings of the composition; a masked element in motion
can be set with motion blurred edges that obey the same
composition settings as any other animated element.
That means you have a chance of matching the motion
blur using a mask whose contours would ﬁt the shape of
the object in its stationary state (Figure 7.6). To make the
mask work with the moving element is a matter of getting
the composition’s Shutter Angle and Shutter Phase settings
to match. (For details, see Chapter 2.)

218


Figure 7.6 Animate a mask path and
enable motion blur, and the mask
obeys the setting.

Morph
Let’s talk about morphing—that’s right, that craze of the
early ’90s, the breakout success on Michael Jackson’s Black
or White video by PDI.
Thing is, morphing can come in handy, without being so
blatant and obvious, if you consider it a concealing tool.
It’s useful when you don’t want anyone to notice a transi-
tion between two objects or even a transition between one
object and itself at a different point in time. Or heck, go
crazy morphing together members of your extended family
to see if you look anything like the result.
What exactly is a morph? It is, quite simply, a combination
of two warps and a dissolve. Given two images, each with
a corresponding shape (say, the features of the face), you
warp the source image’s face shape to the face shape of
the target, warp the target from something that matches
the source to its face shape, fading the target in over the
source.
After Effects has no tool called “Morph,” and for a long
time, the program offered no good way to pull off this
effect. However Reshape is a warping tool that lays the
groundwork for simple morphs. You can build these up
into more complex morphs, with separate individual transi-
tions occurring to create an overall transition.

219


Reshape
Unfortunately, creating a morph with Reshape is nowhere
near a one-button process. This book generally steers away
from step-by-step recipes in favor of helping you solve
larger problems creatively, but in this case, it’s easy to get
confused. So here is a step-by-step, using a demonstration
of my own resemblance transitioning to that of my evil twin
(Figure 7.7):
1. Start with two elements that have some basic similarities
in their relative position of features. The subtler the
distortion required, the more you can get away with.
2. Isolate the two elements you will be morphing from
their backgrounds to prevent contamination. If you’re
working with elements shot against a blue screen, key
them out ﬁrst. Otherwise, mask them. In my example, I
created masks for each layer.
3. Ascertain that the two layers are the same size in the X
and Y dimensions and as closely aligned as possible. If
they are not, precompose one layer to match the other.
This step is important because it will make matching the source
and target masks much, much simpler. Name the two layers
Source and Target (as they are in the example) if it
helps you follow along.
4. Choose matching sections from your two clips for the
duration of your transition (say, 24 frames). If you
have still elements, no worries. The closer your moving
elements are to still, the more likely you’ll get a clean
result.

Figure 7.7 Me and my evil twin, and 5. Draw a mask around the boundary that is going to
that awkward adolescent phase in morph in both layers. Be as precise as possible, erring
between.
to the inside if at all (Figure 7.8). This could be the
masks created in step 2, if you created masks there.
It need not be active; if you don’t need it to mask out
the background, set it to None. Rename each mask
something descriptive like outer (the name used in
my example).

220


Figure 7.8 The outline of each head
is carefully masked. These two masks
need not correspond to one another;
they are defining the boundary for the
Reshape effect and can have differing
numbers and orders of points. This
mask prevents Reshape from pulling
bits of background into the morph
region.

6. Create a mask around the area of the Source layer that
is the focus of the morph. In my example, the focus is
facial features: the eyes, nose, and mouth (Figure 7.9).
Make this mask as simple as you can; use Rotobeziers
and as few points as will work to outline the features in
question.

Figure 7.9 Here’s the full setup along
with the top layer enabled at 100%
and color corrected. It begins the final
animation at 0% Opacity, reaching
100% at the end of the transition.

7. Set the mask mode to None; this mask is your ﬁrst
shape, you don’t need it to inﬂuence the layer at all;
you will use it for the Reshape effect only. Give it a
name (mine is called me).
8. Copy this mask shape and paste it in the Target layer. If
the two layers are the same dimensions, it should be an
identical size and position as it was in the Source layer.
As of version 7.0, mask names are
9. Duplicate the mask. Give it a different color and re- copied along with masks; the cop-
name it. In my case, I called the mask it. ied mask in step 8 remains named
me in the new layer.
10. First scale and rotate, then if necessary move indi-
vidual points (as little as possible!) so that the it mask
surrounds the equivalent area of the Target layer that

221


me does of the source: in my example, the eyes, nose,
and mouth.
11. Copy the resulting it mask shape, and paste it to the
Source layer to create a new, third mask.
12. You now have three masks for each of the Source and
Target layers. Now apply the Reshape effect to each
layer.
13. Starting with the Source layer, set the Source Mask
(me), the Destination Mask (it), and the Boundary
Mask (outer). Set Elasticity to Liquid (you can experi-
The Correspondence Points in
Reshape can be raised from the
ment with other settings later if need be). Set Interpo-
default of 1 to make the distortion lation to Smooth.
more precise (and in some cases,
less twisted). The downside is that 14. At the first frame of the morph transition, still in
this slow effect thus becomes even Source, set a keyframe for Percent (at the default,
slower. Better to simplify what 0.0%, meaning no reshaping is occurring). At the last
you’re attempting to do with your frame, set Percent to, you guessed it, 100.0%. Now sit
masks and fix things there; raise
this value only as a last resort.
back and wait for the gruesome transformation. Don’t
worry about how it looks yet.
15. Repeat steps 12 and 13 for the Target layer, with the
following changes: Set Source Mask to it and Destina-
tion Mask to me. The Percent keyframes should be set
to 0.0 at the last frame of the transition (where it is key-
framed to 100.0 on the Source layer) and, you guessed
it, 100.0 at the first frame.
16. You’ve created the warps, now you just need the cross
dissolve. Set an Opacity keyframe for Target at the
last frame of the transition (leaving it at the default
100.0%), and then add a 0.0% Opacity keyframe at the
first frame.
You should now be ready to preview. The preview may take
a long time to build the first frame; subsequent frames ren-
If at any point during setup it
becomes difficult to interact with der much more quickly, so be patient. The main question
the UI because After Effects is to resolve is whether the features line up properly; if not,
taking so long updating a frame, you must adjust the source and destination mask shapes
enable Caps Lock on your key- accordingly, watching the median frame to see whether the
board to prevent any further frame
rendering until you’re done. changes are improving matters.

222


At this point my example may be looking good in terms of
the face transition, but, of course, I’ve made life hard on
myself by transitioning from a head of one size to a larger
one, and the edges just kind of fade in. Therefore—and
this is where it can get really complicated—I add a sec-
ond morph, using the same steps as before, but with all
new masks.
Why complicated? Why new masks? These two questions
are interrelated. If you set a second Reshape effect, the
key is to avoid inﬂuencing the result of the ﬁrst Reshape Because I know this is complicated,
I’ve included the source and the
effect at all. Therefore, on the second instance the Bound- final result as projects on the book’s
ary mask should be the boundary of the object minus the DVD-ROM. Please do not use the
area occupied by the original source and destination masks. My result as the centerpiece of your
second set of shapes covers the ears and top of the head horror feature.
but avoids the area of the previous masks completely
(Figure 7.10).

Figure 7.10 The full setup to do more
than one morph on a single image
quickly becomes pretty gnarly. Here
are two non-overlapping holdout
areas containing two sets of transition
curves. This setup also takes exponen-
tially longer to render.

223


Now, a quick look at a detail I left out (by having you
duplicate the Source and Destination masks in step 9, for
example, rather than draw them from scratch).

First Vertex and Target Mask
The Reshape tool heavily relies on the First Vertex to
determine which point on the source mask corresponds to
which on the destination mask. In this morph example, it
is easiest to duplicate the source mask to create the desti-
nation mask to automatically satisfy the two criteria most
essential for a smooth Reshape effect:
. Placement of the First Vertex corresponds on both
masks
. Each mask has the same number of points
If either is not fulﬁlled, Reshape will execute some not-
so-nice compensatory measures, probably deforming the
in-between frames in undesirable ways. The easiest solution
is usually to duplicate the source mask and edit it, keep-
ing the same number of points. The next easiest method
would be to draw a new mask with the same number of
points, in the same direction (clockwise or counterclock-
wise), and to set the First Vertex where you need it (by
context-clicking on the mask and choosing Select First
Vertex from the menu).
To draw a new mask, go to the Layer viewer and choose
Masks from the View menu. If the layer has a mask, a Target
menu appears along the bottom of the viewer (Figure 7.11).
Choose the mask you want to replace from this menu and
draw or paste a new one. If the Mask Path has keyframes,
it will deform from the previous mask shape to the new
one, and if the First Vertex and vertex direction match, the
transition will be clean.

224


Figure 7.11 In the Layer panel, the
View menu (highlighted in yellow)
lets you choose any applied masks,
effects or paint strokes; if you want
to disable the effect of the selected
item, uncheck Render (in green). If
you select a Target mask (in orange)
you can then simply start drawing
or paste, and the existing mask is
replaced at that frame.

Puppet
Reshape is old-school compared with the Puppet tools.
To get started with Puppet, select a layer (preferably a
foreground element with a shape deﬁned by transparency,
such as an alpha channel or mask), and add the Puppet
Pin tool (Ctrl+P/Cmd+P) by clicking points on areas
you want to animate. Click two points, drag the third,
and behold: The layer is now pliable in a very organic,
intuitive way.

225


Figure 7.12 The outline of the pretzel
source can be seen in wireframe view;
this is the Puppet mesh. The pins, in
yellow, are effectively the joints.

Perhaps just because I could, I created a pretzel (Figure
7.12). Puppet can be used for much subtler stuff, but this
example clearly demonstrates something previously impos-
sible in After Effects.
Here are the basic steps to create a deformation animation:
1. Use the Puppet Pin tool to add three pins to a fore-
ground layer. Experimentation will tell you a lot about
where to place these, but they’re very similar to the
places on a marionette where the wires connect: the
center, joints, and ends.
2. Move one of them, and observe what happens to the
overall image.
3. Add points as needed to further articulate the
deformation.
4. To animate by positioning and timing keyframes:
Expose the numbered Puppet Pin properties in the
Timeline (UU); these have X and Y Position values
To animate an image from its
initial position once you’ve already
matching many other properties in After Effects.
deformed it, create keyframes for 5. To animate in real time: Hold Ctrl/Cmd as you move
the pins that have moved, go to the
frame that should have the initial your cursor over a pin, and a stopwatch icon appears;
position, and click Reset for those click and drag that pin, and a real-time animation of
pin properties. Only keyframes at your cursor movement records from the current time
the current time are reset. until you release the mouse. You can specify an alter-
nate speed for playback by clicking Record Options in
the toolbar.

226


Try these steps with virtually any matted image and see
what happens; you’ll quickly get the feel for how Puppet is
used. To refine the result, you have some options.
In the toolbar is a toggle to show the mesh. The mesh not
only gives you an idea how Puppet actually works, it can be
essential to allow you to properly adjust the result. I like to
leave it on.
The Expansion and Triangles properties use defaults that
are often just fine for a given shape. Raising Expansion can
help clean up edge pixels left behind in the source. Raising
Triangles makes the deformation more accurate, albeit
slower. The default number of triangles varies according to
the size and complexity of the source.

More Tools
The Puppet Pin tool does the heavy lifting, but Ctrl/
Cmd+P cycles through two other Puppet tools that help
in special cases.
Pins disappeared? To display them,
The pretzel example requires specific control over which three conditions must be satisfied:
parts of the deformation overlap as two regions cross; this the layer is selected, the Puppet
effect is active, and the Puppet Pin
is handled by the Puppet Overlap tool. The mesh must tool is currently selected.
be displayed to use Puppet Overlap, and you apply the
Overlap point on the original, undeformed mesh shape,
not the deformation (Figure 7.13).

Figure 7.13 The blue starch pin is
placed where the tail of the pretzel
had mistakenly been below all over-
lapping segments, and then its Extent
is raised near 500 (to highlight and
bring forward more polygons) and its
In Front setting raised to 100%.

Overlap is not a tool you animate (except perhaps its
numerical settings, which are found in the Timeline).
Instead, you place it at the center of the area that should

227


overlap others and then adjust the settings as to how much
“In Front” it’s meant to be and its Extent (how far from the
pin this overlap area reaches). You can leave the In Front
setting at the default until you start mixing more than one
overlap; the higher the value, the closer to the viewer (and
negative values will be further away than those with no set-
ting, which default to 0).
The Starch tool prevents an area from deforming. It’s
not meant to anchor a region of the image but instead to
sit between animated pins, preventing the highlighted area
(expanded or contracted with the tool’s Extent setting)
from being squished or stretched (Figure 7.14).

Figure 7.14 Without starch pins
applied to the hand and fingertips,
such a grotesque extension of the
fingers would distort the fingernails
and back of the hand along with
the fingers. Starch gives you control
over what should not be stretched or
distorted.

228


Paint and Cloning
Paint is generally a last resort when roto is impractical, and
for a simple reason: Paint work is typically painstaking and
more likely to show flaws than approaches involving masks.
Of course, there are exceptions. The ability to track clone
brushes offers a huge advantage over masks, which are not
so easy to track, and painting in the alpha channel is akin
to masking with paint.
For effects work, paint controls in After Effects have at least
a couple of predominant uses:
. Clean up an alpha channel mask by painting directly to
it in black and white
. Clone Stamp to overwrite part of the frame with alter-
nate source
Once you fully understand the strengths and limitations of
paint, you can come up with your own uses.

Paint Fundamentals
Two panels, Paint and Brush Tips, are essential to the three
brush-based tools in the Tools palette: Brush , Clone
Stamp , or Eraser . These can be revealed by choosing
the Paint Workspace.
The After Effects paint functionality is patterned after
Photoshop, but with a couple of fundamental differences.
After Effects offers fewer customizable options for its
brushes (you can’t, for example, design your own brush
tips). More significantly, Photoshop’s brushes are raster
based, while After Effects brushes are vector based, allow-
ing them to be edited and animated at any stage.
Suppose that you have an alpha channel in need of a
touch-up; for example, the matte shown in Figure 7.15 is a
difficult key without matte cleanup due to tracking markers
and shadows. With the Brush tool active, go to the Paint
palette and set Channels to Alpha (this palette remembers
the last mode you used); the foreground and background
color swatches in the palette become grayscale, and you
can make them black and white by clicking the tiny black-
over-white squares just below the swatches.

229


Figure 7.15 Touch up an alpha chan-
nel matte (for example, remove a
tracking marker): In the Paint palette,
select Alpha in the Channels menu,
then display the alpha channel (Alt/
Option+4).

To see what you are painting, switch the view to the Alpha
Channel (Alt+4/Option+4); switch back to RGB to check
the ﬁnal result.
When using the paint tools
. Brush-based tools operate only in the Layer panel.
. Brushes include a Mode setting (analogous to Transfer
Modes).
. With a tablet, you can use the Brush Dynamics settings
at the bottom of the Brush Tips palette to set how the
pressure, angle, and stylus wheel of your pen affect
strokes.
. The Duration setting and the frame where you begin
The older Vector Paint effect (Pro painting are crucial.
only) remains useful for painting in
. Preset brushes and numerical settings for properties
the context of a comp viewer, but it
won’t clone and it’s not intuitive; I such as diameter and hardness (aka feather) live in the
prefer to open two viewers (Comp Brush Tips panel.
and Layer).
Much more fun and interactive than the Brush Tips panel
are the following shortcuts. With the Brush tool active in
the Layer viewer

230


. Hold Ctrl/Cmd and drag to scale the brush.
. Add the Shift key to adjust in larger increments, and
Alt/Option for ﬁne adjustments.
. With the mouse button still held, release Ctrl/Cmd to
scale hardness (an inner circle appears representing
the inside of the threshold, Figure 7.16).
. Alt/Option-click to use the eyedropper (with brushes)
or clone source (with the clone brush).

Figure 7.16 Modifier keys (Ctrl/Cmd
to scale, Alt/Opt to feather, all with
mouse button held) let you define a
brush on the fly. The inner circle shows
the solid core; the area between it and
the outer circle is the threshold (for
feathering).

By default, the Duration setting in the Paint menu is set
to Constant, which means that any paint stroke you create
on this frame continues to the end of the layer. For clean-
There is a major gotcha with Con-
ing up an alpha channel, this is typically not a desirable stant (the default mode): Paint a
setting because you’re presumably painting stray holes in stroke at any frame other than the
your matte here and there, on single (or just a few) frames. first frame of the layer, and it does
The Single Frame setting conﬁnes your stroke to just the not appear until that frame during
playback. It’s apparently not a bug,
current frame on which you’re painting, and the Custom
but it is certainly an annoyance.
setting allows you to enter the number of frames that the
stroke will persist.
The other option, Write On, records your stroke in real
time, re-creating the motion (including timing) when you

231


replay the layer; this stylized option can be useful for such
motion graphics tricks as handwritten script.
The Brush Tips panel menu includes various display
options and customizable features: You can add, rename,
or delete brushes, as well. You can also name a brush
by double-clicking it. Brush names do not appear in the
default thumbnail view except via tooltips when you move
your cursor above each brush.
For an alpha channel, you will typically work in Single
Frame mode (under Duration in the Paint panel), looking
only at the alpha channel (Alt+4/Option+4) and progress-
ing frame by frame through the shot (pressing Page Down).
After working for a little while, your Timeline may contain
dozens of strokes, each with numerous properties of its
own. New strokes are added to the top of the stack and are
As in Photoshop, the X key swaps
the foreground and background
given numerically ordered names; to select one to edit or
swatches with the Brush tool active. delete, you may more easily ﬁnd it using the Selection tool
(V) to directly select it in a viewer panel.

Cloning Fundamentals
When you clone moving footage, the result retains grain
and other natural features that still images lack. Not only
can you clone pixels from a different region of the same
frame, you can clone from a different frame of a different
clip at a different point in time (Figure 7.17), as follows:
. Clone from the same frame: This works just as in Pho-
toshop. Choose a brush, Alt/Option-click on the area
of the frame to sample, and begin painting. Remem-
The Aligned toggle in the Paint
panel (on by default) preserves 1:1 ber that by default, Duration is set to Constant, so any
pixel positions even though paint stroke created begins at the current frame and extends
tools are vector-based. Nonaligned through the rest of the composition.
clone operations tend to appear
blurry. . Clone from the same clip, at a different time: Look at
Clone Options for the offset value in frames. Note that
there is also an option to set spatial offset. To clone
from the exact same position at a different point in
time, set the Offset to 0, 0 and change the Source Time.

232


Figure 7.17 Clone source overlay is checked (left) with Difference mode active, an “onion skin” that makes it possible to line
up precisely two matching shots (middle and right images). Difference mode is on, causing all identical areas of the frame
to turn black when the two layers are perfectly aligned.

. Clone from a separate clip: The source from which
you’re cloning must be present in the current compo-
sition (although it need not be visible and can even
be a guide layer). Simply open the layer to be used as
source, and go to the current time where you want to To clone from a single frame only
to multiple frames, toggle on Lock
begin; Source and Source Time Shift are listed in the Source Time in the Paint panel.
Paint panel and can also be edited there.
. Mix multiple clone sources without having to reselect
each one: There are five Preset icons in the Paint panel;
these allow you to switch sources on the fly and then
switch back to a previous source. Just click on a Preset
icon before selecting your clone source and that source
remains associated with that preset (including Aligned
and Lock Source Time settings).
That all seems straightforward enough; there are just a few
things to watch out for, as follows.
Clone is different from many other
Tricks and Gotchas
tools in After Effects in that Source
Suppose the clone source time is offset, or comes from Time Shift uses frames, not seconds,
a different layer, and the last frame of the layer has been to evaluate the temporal shift.
Beware if you mix clips with differ-
reached—what happens? After Effects helpfully loops back
ent frame rates, although on the
to the first frame of the clip and keeps going. This is dan- whole this is a beneficial feature.
gerous only if you’re not aware of it.
Edit the source to take control of this process. Time remap-
ping is one potential way to solve these problems; you can
time stretch a source clip or loop it intentionally.
You may need to scale the source to match the target.
Although temporal edits, including time remapping,
render before they are passed through, other types of

233


edits—even simple translations or effects—do not. As
always, the solution is to precompose; any scaling, rota-
tion, motion tracking, or effects to be cloned belong in the
subcomposition.
Finally, Paint is an effect. Apply your ﬁrst stroke and you’ll
see an effect called Paint with a single checkbox, Paint
on Transparent, which effectively solos the paint strokes.
You can change the render order of paint strokes relative
to other effects. For example, you can touch up a green-
screen plate, apply a keyer, and then touch up the resulting
alpha channel, all on one layer.
The View menu in the Layer panel (Figure 7.18) lists, in
order, the paint and effects edits you’ve added to the layer.
To see only the layer with no edits applied, toggle Render
off; to see a particular stage of the edit—after the ﬁrst
paint strokes, but before the effects, say—select it in the
View menu, effectively disabling the steps below it. These
settings are for previewing only; they will not enable or dis-
able the rendering of these items.

Figure 7.18 Isolate and solo paint strokes in the View menu of the Layer panel.

You can motion track a paint stroke. To do so requires
the tracker, covered in the next chapter, and a basic
expression.

Wire Removal
Wire removal and rig removal are two common visual
effects needs. Generally speaking, wire removal is cloning
over a wire (typically used to suspend an actor or prop in
mid-air). Rig removal, meanwhile, is typically just an ani-
mated garbage mask over any equipment that appeared
in shot.
After Effects has nothing to compete with the state-of-the-art
wire removal tool found in The Foundry’s Furnace plug-ins

234


(which sadly are available for just about every compositing
package besides After Effects, although Adobe did license
the Kronos toolset from Furnace to become Timewarp).
The CC Simple Wire Removal tool is indeed simple: It
replaces the vector between two points by either displacing
pixels or using the same pixels from a neighboring frame.
Photoshop Video
There are Slope and Mirror Blend controls allowing you a
Photoshop offers an intriguing alternative to
little control over the threshold and cloning pattern, and the After Effects vector paint tools; you can work
you can apply a tracker to each point via expressions and with moving footage in Photoshop. The After
the pickwhip (usage described in Chapter 10). Effects paint tools are heavily based on those in
Photoshop, but with one key difference: Photoshop
The net effect may not be so different from drawing strokes are bitmaps (actual pixels) and those from
a two-point clone stroke (sample the background by After Effects are vectors. This makes it possible to
Alt/Option-clicking, then click one end of the wire, and use custom brushes, as are common in Photoshop
(and which are themselves bitmaps). There’s not
Shift-click the other end). That stroke could then be as much you can do overall with the stroke once
tracked via expressions. you’ve painted it as in After Effects, but if you like
working in Photoshop, it’s certainly an option. After
Rig removal can very often be aided by tracking motion, Effects can open Photoshop files containing video.
because rigs themselves don’t move, the camera does. The
key is to make a shape that mattes out the rig, then apply
that as a track matte to the foreground footage and track
the whole matte.

Dust Bust
This is in many ways as nitty-gritty and low-level as rotoscop-
ing gets, although the likelihood of small particles appear-
ing on source footage has decreased with the advent of
digital shooting and the decline of film. Most of these flaws
can be corrected only via frame-by-frame cloning, some-
times known as dust busting. If you’ve carefully read this
section, you already know what you need to know to do this
work successfully, so get to it.

Conclusion
And so, like rain, into every effects artist’s life a little
rotoscoping must fall. The tools outlined here are mostly
sufficient for the type of rotoscoping work that composi-
tors will have to do. Dedicated rotoscope artists would
likely choose software other than After Effects or perhaps
employ Silhouette as an After Effects plug-in. As long as

235


rotoscoping isn’t your stock in trade, however, the After
Effects tools will usually allow you to ﬁnish the shot without
having to look for other software.
The next chapter completes the picture by adding motion
tracking to your areas of expertise. As mentioned,
motion tracking plus rotoscoping can equal a shortcut
around tedious tasks.

236

CHAPTER

8
Effective Motion Tracking

Never mistake motion for action.
—Ernest Hemingway

Effective Motion Tracking

T here is more that can be done with tracking than
simply sampling the motion of one layer and applying it
to another, even though that’s fundamentally what it does;
because you can track in several different ways using a vari-
ety of points, the methods and applications go far beyond
the obvious.
Native tracking possibilities have been further expanded in
After Effects CS4 thanks to the licensing and inclusion of
MochaAE from Imagineer Systems. Previously available as
a third-party application, Mocha uses a fundamentally dif-
ferent approach (planar tracking) than that of the built-in
After Effects point tracker and it solves a problem (corner
pinning) that had exposed weaknesses in that tracker.
It is also possible to use third-party 3D tracking software to
import 3D camera data matched to real-world motion into
After Effects, via a methodology that is described at the
end of this chapter.
All of these automated trackers are astoundingly accurate,
often sampling motion at a detail level that would be impos-
sible for you to pick up by eye. It’s not a fully automated
process, however, and so you must understand how to
. Choose effective track regions
. Customize settings based on the particular scene
. Rescue tracks that seem to have gone astray (or
abandon and restart, provided you understand what
to try instead)
. Handle motion blur

238


This chapter addresses these techniques and more. Once
you grasp these, you can use tracking to go beyond the
ordinary:
. Match multiple elements to a single track using the
3D camera
. Stabilize a handheld or other moving camera shot
. Continue tracks when the object being tracked leaves
the visible frame
. Import and use 3D tracking data
There are so many cases where effective tracking can help
accomplish the seemingly impossible; it’s worth knowing
this stuff inside out.

Point Tracking Essentials
Step one is to be able to get a good basic track right in
After Effects. You may ﬁnd your tracks going astray after
reviewing the clear After Effects documentation on how to
use the tracker. Here are some background fundamentals.
Tracking is a two-step process: The tracker analyzes the
clip and stores its analysis as a set of layer properties that
don’t actually do anything. The properties must be applied
to take effect. Both steps, setting the tracking target and
applying the track, occur in the Tracker Controls panel
when matching or stabilizing motion in After Effects.

Choose a Feature
Success with the After Effects tracker relies on your ability
to choose a feature that will track effectively (Figure 8.1).

Figure 8.1 Objects with clear con-
tours, definition, and contrast make
the best track targets.

239

Chapter 8 Eﬀective Motion Tracking

Ideally, the feature you plan to track
. Is unique from its surroundings
. Has high-contrast edges—typically a corner or point—
Search and feature regions don’t entirely within the feature region
have to be square! Widen the fea-
ture region to match a wide, short . Is identiﬁable throughout the shot
target feature. With unidirectional . Does not have to compete with similar distinct features
motion—say, a right-to-left within the search region at any point during the track
pan—widen and offset the search
region in the direction of the pan . Is close to the area where the tracked object or objects
(Figure 8.2). will be added

Figure 8.2 Offset the track region in
the direction of uniform motion, if
any exists.

Figure 8.3 shows the many draggable features of the
tracker control. It’s very easy to grab the wrong thing, so
pay close attention to the icon under your cursor as you set
up the track.

Figure 8.3 Many interactive controls
are clustered close together in the
tracker. Identified here are: A. Search
region, B. Feature region, C. Keyframe
marker, D. Attach point, E. Move
search region, F. Move both regions,
G. Move entire track point, H. Move
attach point, I. Move entire track point,
J. Resize region. Zoom in to ensure
you’re clicking the right one.

A B C D E F G H I J

240


The main decisions when setting up a track regard the size
and shape of the Search and Feature regions. Keep the fol-
lowing in mind:
. A large Feature region averages pixel data, producing a
smoother but possibly less accurate track.
. A small Feature region may pick up noise and grain as
much as trackable detail, creating an accurate but jit-
tery track.
. The bigger the Search region, the slower the track.
. The Feature region doesn’t have to contain the area of
frame you want to match. One way to offset a track is to
move the attach point—that little x at the center of the
tracker. A better solution is to apply the track to a null,
as is discussed later.
Most features that you’ll track are unreliable—they change
perspective, lighting or color throughout the course of the
shot. The following sections explain what to do when you
don’t have a constant, trackable feature exactly where you
want your target to go.

Tweak the Tracker
There are ﬁve types of track listed under the Track Type
menu in the Tracker panel (Figure 8.4).

Figure 8.4 The various available track
types are set in the Tracker Controls
panel.

Stabilize and Transform tracks are virtually identical until
applied. Edit Target shows the singular difference between
them: Stabilize tracks are always applied to the anchor point
of the tracked layer. Transform tracks are applied to the posi-
tion of a layer other than the tracked layer.

241


Using Stabilize, the animated anchor point (located at the
center of the image by default) moves the layer in opposi-
tion to Position. Increasing the anchor point’s X value
(assuming Position remains the same, which it does when
you adjust the Anchor Point value directly in the Timeline)
moves the layer to the left, just as decreasing the Position
value does.
Corner Pin tracks are very different; in After Effects these
require three or four points to be tracked, and the data is
applied to a Corner Pin plug-in to essentially distort the
perspective of the target layer. Because these tracks are
notoriously difﬁcult and unreliable, the happy truth is that
MochaAE, which also generates data that can be applied to
a corner pin, more or less supersedes Corner Pin tracking.
Raw tracks generate track data only, graying out the Edit
Target button. What good is track data that isn’t applied
anywhere? It can be used to drive expressions or saved to
be applied later. It’s no different than simply never clicking
Edit Target; the raw track data is stored within the source
layer (Figure 8.5).

Figure 8.5 Tracker data is stored
under the tracked layer and can be
accessed at any time.

Position, Rotation, and/or Scale
You can’t uncheck the Position toggle in the Tracker panel
(thus avoiding the unsolvable riddle, what is a motion
track without Position data?), but you can add Rotation
and Scale. Enable either toggle, and a second track point is
automatically added.

242


Figure 8.6 These two points are not suitable for a two-point track because they exist on different planes of depth relative to
the camera; the rear one covers more distance in this pan shot.

Additionally tracking rotation and scale data is straight-
forward enough, employing two track points instead of
one. Typically, the two points should be roughly equidis-
tant from the camera due to the phenomenon of parallax
(Figure 8.6).
Tracks that add rotation and scale usually have features
that change appearance through the course of the shot
(because they themselves may rotate and scale). Tracker >
Options contains the Adapt Feature on Every Frame toggle.
By default, the tracker is set to adapt the track feature
if conﬁdence slips below 80%. Adapt Feature on Every
Frame is like restarting the track on each and every frame,
comparing each frame to the previous one instead of the
one you originally chose. For ordinary tracks this adds an
unwanted margin of error, but in a case where a feature is
changing each frame anyhow, because the scale, orienta-
tion or blurriness is in constant ﬂux, this can help.

243


If a track keeps slipping going forward, try starting at the
end of the clip and tracking in reverse (note that the “play”
icons in Tracker Controls go both directions). It’s best to
start where your feature is largest and most prominent
Subpixel Motion
so that it doesn’t outgrow the boundaries of the Feature
The key feature of the After Effects tracker is
subpixel positioning, on by default in Motion Tracker region.
Options. You could never achieve this degree of
accuracy manually; most supposedly “locked off” Confidence
scenes require stabilization despite that the range
At the bottom of Motion Tracker Options is a submenu
of motion is less than a full pixel; your vision is
actually far more acute than that. of options related to After Effects’ mysterious Confidence
settings. Every tracked frame gets a Confidence setting, an
evaluation of how accurate the track was at that frame. This
As you watch a track in progress, the trackers move
in whole pixel values, bouncing around crudely, but may or may not be indicative of the actual accuracy, but my
unless you disable subpixel positioning, this does experience is that you’re almost guaranteed to be fine with
not reflect the final track, which is accurate to 1⁄10,000 values above 90%, and real problems will cause this value
of a pixel (four places below the decimal point). to drop way down, even to 30% or less (Figure 8.7).

Figure 8.7 It’s not hard to spot where this track lost its target, looking at the Confidence graph.

Depending on this setting, you can (Figure 8.8)
. Continue Tracking. Just power ahead no matter what.
. Stop Tracking. So you can reset the tracker manually
To reveal the current track in the right at the problem frame.
Timeline with the Track Controls . Extrapolate Motion. Let After Effects guess based on
active, use the SS (show selected) how it was moving on the previously tracked frames.
shortcut.

244


This is for cases where the tracked item disappears for
one or a few frames.
. Adapt Feature. The default behavior is to go ahead and
change the reference Feature region to the previous
frame if there is low Conﬁdence, indicating the feature
has suddenly changed appearance.

Figure 8.8 This menu specifies what
to do when the Confidence rating
drops below a certain threshold dur-
ing the track.

Whichever you choose, you also have the option to go back
to the frame where a track went wrong, reset the feature
region by hand, and restart the track.
Most of the other Motion Tracker
Options are rarely adjusted. If your
Solve Problems with Nulls footage has contrast only on one
I almost always apply track data to a null object layer and color channel, you can try RGB
instead of Luminance; I’ve never
then use parenting to apply the motion to other layers.
seen footage that has trackable
This offers a bunch of advantages: contrast in Saturation, although
it must exist. You don’t need to
. Offset and relocate tracks: It doesn’t matter whether Track Fields if you removed them
the track attach point is in the right location; the on import. You can blur a jittery
null picks up the relative motion, and any layer par- track before match, or enhance
ented to it can be repositioned or animated on its own low-contrast footage, but adjusting
the size of the Feature region or
(Figure 8.9).
choosing a different feature, can do
. Optionally lock the tracked layer once you’re happy a better job making your track more
with it, so you don’t inadvertently move it. or less precise.

. Use a Stabilization track to peg multiple object to a
scene (next section).
. Parent one motion track to another if you need to
switch to a different track target mid-shot.

245


One common problem is that a track is good only until a
given point in the scene, and if you then move the tracker
to search a different location, the whole track jumps. By
adding a new track and another null, you can create and
apply as many individual tracks as you like, parenting one
to the next.

Figure 8.9 The rope must be painted and tracked to reach the edge of the frame (top). The point where the rope meets the
edge of the set can be tracked (middle), and a two-point mask with a Stroke effect can be added to replace it (bottom left).
The final animation offsets the track to match the rope angle (bottom right).

246


Track Averaging and Mask Tracking
A couple of techniques have long been popular but
reserved to the select few willing to set them up; with the
addition of a script called TrackerViz (credited at the front
Lost track of a track? Tracker >
of the book), there’s no need to undergo the pain in order Motion Source and Tracker >
to get the benefit. Current Track offer a centralized
location for all tracks in a comp.
If you are having a difficult time getting an After Effects Eligible layers are listed in black
track to stick, there’s an alternative to continually rework- under Motion Source. To delete
ing a single track; several not-so-accurate tracks, averaged unused tracks, select them under
together, will often be more accurate than any alternative the source layer in the Timeline.
approach. The setup involved is a slight pain, however,
which is where the script comes in (Figure 8.10).

Figure 8.10 The TrackerViz script takes a set of passes tracking the same object
and averages them; the small red solid is a layer created as the average of the
three tracked nulls.

TrackerViz also lets you interchange tracked nulls and mask
points, making it possible to effectively track in a mask. You
first track a set of nulls, each with a unique name, corre-
sponding to vertices on the shape. Select them in the order

247


that you would draw them, and in TrackerViz choose and
apply Layers to Shape in the Function menu. The result is
not a shape layer but much more usefully, a solid with a key-
framed mask applied to it. The mask can be used to replace
any equivalent hand-drawn masks applied to layers.

Match Motion Blur
Don’t neglect the possibility of motion blur! It can be your
friend until it appears unwanted, like a malingering house-
guest, impossible to ignore.
A good track won’t look right until its motion blur also
matches that of the background plate. Matching motion
blur is similar to color or grain matching; you do it by eye,
most often zooming in to an area of the frame where it is
apparent in both the foreground and background.
Motion blur settings are not interactive; they reside in
Composition Settings (Ctrl+K/Cmd+K) > Advanced, where
you don’t see the effect of your changes until you clear
the dialog by clicking OK. As described back in Chapter 2,
adjust Shutter Angle and Shutter Phase until you see a good
match, raising (or in the odd case, lowering) Samples per
Frame and Adaptive Sample Limit to match (Figure 8.11).

Figure 8.11 Motion tracking can’t work without matching motion blur. This shot uses the standard film camera blur: a 180
degree shutter angle, with a phase of –90.

If you’re trying to stabilize a scene that contains motion
blur, you have a bigger problem, one that needs to be
avoided when shooting (by boosting the shutter speed of
the camera, where possible, Figure 8.12). In such cases it’s
no doubt preferable to smooth camera motion rather than
lock the stabilization (see “Smooth a Camera Move” later
in this chapter).

248


Figure 8.12 If shooting with the intention of stabilizing later, raise the film shutter speed and reduce camera motion to
minimize motion blur and a huge image gutter, which both soften the image beyond the point of being useful.

Match Multiple Objects
In the real world, objects sit in an environment, and if that
environment or the point of view changes, they remain in
place. You knew this. You may not know how to make a 2D
After Effects track re-create it.
The key is to stabilize the background layer and then par-
ent a camera to that stabilization, restoring the motion.
The motion of the source camera is captured and applied
to a virtual camera so that any elements you add to the
scene pick up on that motion. It’s quite cool.

The AE Camera as a Tracking Tool
Suppose you have an arbitrary number of foreground lay-
ers to match to a background plate: not just objects, but
color corrections, effects with hold-out masks, you name it.
Applying track data to each of those layers individually is
a time-consuming headache, and even parenting them all
to a tracked null may not work properly if there is rotation
or scale data, as that null then becomes the center of those
translations.
Instead, the following method allows you to stabilize the
background scene, add static foreground elements, and
then reapply the scene motion.

249


1. With the background layer selected, choose Stabilize
Motion (either by context-clicking the layer or by
choosing it from the Animation menu).
2. Stabilize the layer for Position, Rotation, and Scale,
using two points equidistant from the camera.
3. The stabilized layer offsets and rotates in the frame
(Figure 8.13). Return to the ﬁrst frame of the track
(quite possibly frame 0 of the comp). Turn on the stabi-
lized layer’s 3D switch.

Figure 8.13 Gaps open up around the
edges of the image as the track points
are held in place.

4. Add a 3D camera (context-click in an empty area of the
Timeline); in the Camera Settings, give it a name like
A 50 mm camera lens in After trackerCam, use the 50mm preset, and click OK.
Effects offers a neutral perspective;
toggle any layer to 3D and it should 5. Parent the camera layer to the stabilized layer
appear the same as in 2D. (Figure 8.14).

Figure 8.14 The relevant Transform
properties are copied and pasted to
a null, to which the camera is then
parented.

250


Everything now appears back to normal, with one intrigu-
ing twist: Any new item added to the scene picks up the
scene motion as soon as you toggle its 3D switch. All you
have to do is drop it in and enable 3D. Any layer that
shouldn’t follow the track, such as an adjustment layer, can
remain 2D (Figure 8.15).

Figure 8.15 Extra layers for a new
clock face and child’s artwork along
with a shadow are added as 3D layers,
so they pick up the motion of the
scene as captured by the tracked
camera.

2.5D Tracking
You can even fake 3D parallax by offsetting layers in Z
space. Any layer that is equidistant from the camera with the
motion track points has a Z-depth value of 0. Offsetting lay-
ers is tricky as there is no frame of reference for where they
should be placed in Z space—not even a grid (Figure 8.16).

Figure 8.16 Rotoscoped shapes can
be tracked so that they “stick” in a
scene; you don’t have to animate the
shape itself unless it is also a matte
that needs to be articulated around a
moving figure. Here it is even offset in
3D to match the parallax of the wall
nearer camera.

251


This is what I call “2.5D tracking”: a 3D camera paired with
two-dimensional tracking points and layers. Any 3D offset
derived from a 2D track is only approximately accurate, so
2.5D tracking will even stick this is a big cheat; if your scene has areas of frame closer or
foreground layers to a zoomed shot; further than the track points that you wish to match, you
the Scale stabilization scales the can try to guess where in 3D space to place a layer and you
parented camera, making it appear may just get lucky.
to zoom.
If your shot tracks (the camera moves), it probably requires
a real 3D camera track (see “Try It Out for Yourself” at the
end of this chapter).

Stabilize a Moving Shot
After Effects doesn’t provide you with a workﬂow to take
a shot with a moving camera—handheld, aerial, crane, or
dolly—and smooth bumps in the motion while retaining
the move. But you can do it.
Figure 8.17 features an aerial shot of Silicon Valley (look
closely and you’ll see Adobe headquarters). It wasn’t taken
in a helicopter with a gyro, but rather from a camera
attached to a 2 × 4 wooden plank, C-clamped to the open
door of a Cessna aircraft. The pilot did his best to keep the
plane steady, but even on a calm day that type of aircraft
bobs and weaves, which is why you might pay ten times
more money to get the helicopter with the gyro.

Figure 8.17 Track two co-planar
points at whatever distance the eye
is focused—here the near-horizon—
and stabilize for position, rotation
and scale.

252


Ah, but it’s so much fun to fix it in post. To stabilize a shot
like this, first decide where the viewer’s attention will be,
because that’s the area that you want to stabilize. Objects in
the near foreground move more than those in the dis-
tance; stabilizing them adds motion on the horizon. In this
case, the viewer’s attention is likely on the near-horizon,
below the foothills. Begin by creating a Position/Rotation/
Scale stabilization based on two points at that depth.
The basic setup replicates the first four steps from the last
section, but instead of parenting the camera to the stabi-
lized source layer, follow these steps:
1. Create a null named Stabilizer (or your choice), and
enable 3D.
2. Select all of the Anchor Point keyframes from the stabi-
lized background plate.
3. Copy and paste the Anchor Point keyframe data to
Stabilizer.
4. Apply smoothing to this duplicated smoothing data. Figure 8.18 The camera is parented to
a null whose transform data matches
5. Parent the camera to the Stabilizer layer (Figure 8.18). that of the stabilized plate.

You have a couple of options to “apply smoothing”:
Window > Smoother is an automatic but destructive solu-
tion; the smooth() expression is a vastly superior technique,
because you can view the smoothed motion in the Graph
Editor without affecting the source keyframes. If, later on, it
turns out you need to adjust it again, you can (Figure 8.19).

Figure 8.19 The smooth() expression
is displayed as a reference graph adja-
cent to the actual tracked keyframes
and does not rewrite them. The con-
trol to enable the expressions graph
is grouped with the other expression
controls under the property.

253


To apply a smooth() expression
1. Alt/Option-click on the Anchor Point stopwatch of the
layer to which the camera is parented.
2. With the default expression (anchorPoint) still high-
lighted, go to the Expressions menu icon and
under Property choose the smooth default:
smooth(width = .2, samples = 5, t = time).

This works, but as a starting setting, I recommend dis-
If expressions and arguments are
gobbledygook to you, take a look at
carding the third argument (“t = time”) and the other
Chapter 10. hints (“width =” and “samples =”) to get the leaner,
meaner and more powerful setting like smooth(2, 48).
The expression works as follows: It gives a command
(smooth()) followed by three settings known as arguments.
The third one, time, is used only to offset the result, and
it’s optional, so I have a habit of deleting it. The hints
for the other two (width = and samples =) are also not
needed to make the expression work—they are there just
to remind you what they do.
Width determines how much time (before and after the
current time) is averaged to create the result. A setting of
2 samples 2 seconds means 1 second before and 1 second
after the current time. The samples argument determines
how many individual points within that range are actually
sampled for the result; generally, the more samples, the
smoother the curve. A setting of 48 means that over that 2
seconds, 48 individual frame values will be averaged (the
maximum for 24 fps footage).
It’s also possible to smooth rotation in this manner,
although I ﬁnd a lighter touch (fewer samples) works best
with rotation. However, the best way to ﬁnd out for your
individual shot is by trying different settings, looking at
how smooth the resulting curve (not to mention the actual
motion) appears.
The same basic approach can desta- It’s a little hard to imagine that you can smooth the motion
bilize or “jitter” instead of smooth
data for the camera, causing it to go out of sync with the
camera motion, most likely with a
wiggle() expression, described background, and not have a mismatch. What is actually
in Chapter 10. happening, though, is that the scene motion is removed
completely and then restored in a smoother state.

254


Incorporate MochaAE
After Effects CS4 has added a whole separate tracking
application: MochaAE, licensed from Imagineer Systems.
MochaAE is not in any way integrated directly into After
Effects; it was originally sold as stand-alone software
and remains a separate application (now found in the
same folder as After Effects itself), which you can ﬁre up
as needed.
When is it needed? Mocha is a planar tracker, which means
that it does not analyze individual points or feature regions
like the After Effects tracker (and most other trackers) but
instead looks at entire plane of motion.
Look around the environment where you are right now
and you will see numerous two dimensional planes: walls,
table tops, the backs of chairs, or maybe the trunk of a
tree, or the side, front, top, wheel, hood, window (and on
and on) of a passing truck (Figure 8.20). If you were sit-
ting on the moon reading this book, the surface of earth,
though curved, would track as a plane; you wouldn’t track
the tip of Florida or New Zealand, but the whole planet.

Figure 8.20 A plane does not have
to be flat and rectilinear in order for
Mocha to track it; look around and you
will see many coplanar objects.

Thus Mocha is most plainly useful for Corner Pin track-
ing: replacing a surface with four discernable corners with
another such surface. This type of track has long been the
Achilles heel of the After Effects tracker, and Mocha pro-
vides such welcome relief that I’ve omitted coverage of cor-
ner pinning with the After Effects tracker from this book.

255


Mocha can, in fact, be used for just about any type of 2D
tracking. It’s less convenient to use outside software than to
track directly in After Effects, but once you become com-
fortable with Mocha, you may find that the extra precision
is worth the trouble of doing so.

The Basics
Mocha is an application all unto itself with its own manual,
longer than this chapter (just click Guide on the Mocha
Welcome dialog). It’s not a bad read. Because it’s a new
feature, here’s a look at how best to get up and running.
First of all, Mocha is limited to a subset of footage formats
that you might use in an After Effects project. Cineon log
files are okay, but several other higher-than-8-bpc formats
seem to be off-limits, so you may need to render a special
version of your source clip to be compatible. (I suggest Ani-
mation compressed QuickTime; it’s imperative to get the
frame rate correct and having it embedded in the source
file helps you avoid careless errors with image sequences.)
The wizard to set up a project results in the creation of
the Mocha project file before you get started—either in
a location you specify or adjacent to the shot. The most
crucial thing to get right in this three-page process is frame
rate (which is only an issue with an image sequence); the
difference between, for example, 23.976 and 24 frames per
second is crucial; a mismatch will result in a track that slips.
Once your shot appears in Mocha, the next step is to scrub
to a frame that clearly shows the planar area you wish to
track. The rules are entirely different than with a point
tracker: You don’t need to see the entire plane or any of
its boundaries; you’re just looking for coplanar details that
are most consistent through the shot. “Consistent” means
not obscured by foreground motion or light flares and
reflections, although there are workarounds for both.
On that starting frame, you use one of two tools to draw
out a mask shape: Bézier splines or the less familiar but
preferable Mocha X-splines . Select either and tap a few
points—it doesn’t matter how many—around the target
details (Figure 8.21). It’s helpful if these points correspond

256


to some recognizable detail for fine-tuning later, but they
don’t have to. Close the shape by clicking back on the
first point.

Figure 8.21 Mocha shows a close-up
of the selected point of the X-spline
shape around the target. Mocha is
a planar tracker and doesn’t track
this individual point like After Effects
would, but these points help you
gauge whether the shape continues
to line up, so it’s helpful to align them
with a feature you can easily identify.

At this point, with an easy track, you can use the forward
and backward controls and watch Mocha work its magic.
If the track sticks to its target, the next step is to fine-tune
it. Assuming you’re exporting corner pin data, you must
at minimum enable Surface under the View Controls and
place the corners where you want them in the result.
If all is still well, you can now click Export Tracking Data
on the Track panel and copy or save After Effects CS4
Corner Pin Data. If you copy, you can then paste directly Mocha is typically used for corner
into the target layer in After Effects at the first frame of the pinning, but you can instead choose
to export After Effects Transform
source; Mocha creates a keyframe on every frame (Figure
Data and use it like regular tracker
8.22). If you instead choose to save a text file, you can then data.
copy and paste its data from an ordinary text editor.

Figure 8.22 The bridge from Mocha
back to After Effects is this export
dialog. If you’ve never seen it, this is
the format that After Effects keyframe
data takes when it’s pasted into a text
file, which is how Mocha exports it;
you then open the text file, select all,
copy, and paste it into the target layer
to be corner-pinned.

257


The Nitty Gritty
It’s normal for a track to be slightly more complicated than
this, usually due to motion or perspective shifts within the
track area. This can be the result of foreground objects
passing across the track region or the appearance of the
region itself changing over time.
Figure 8.23 shows an otherwise straightforward track of a
handheld game unit with the following challenges: flares
and reflections play across the screen, the hands move
back and forth across the unit, and the perspective of the
screen changes from flat to face-on.

Figure 8.23 The tracking markers
on the screen are not necessary for
Mocha to track this handheld unit for
screen replacement, it’s the reflective
screen itself and the movement of the
thumbs across it that present Mocha
with a challenge.

258


There are two standard solutions to any track that slips:
. Sudden slippage is often the result of foreground
motion or light changing the appearance of the tracked
area; the solution is to mask out the area containing the
disturbance.
. Small, gradual slippage is often the result of shifts in
perspective and can be keyframed.
The clip shown in Figure 8.23 requires both techniques.
A track of the entire face of the unit shifts slightly as it is
tilted and a lot as the thumbs move across the track area
and reflections play across the screen.
Big shifts in the track region are caused by changes in the
track area, so I fix those first, adding an additional spline
(or splines) containing the interruptive motion. The Add
X-spline and Add Bézier Spline create a subtractive
shape (or shapes) around the areas of the first region that
contain any kind of motion. Figure 8.24 shows that these
can be oddly defined; they track right along with the main
planar track.

Figure 8.24 Holdout masks are
added to eliminate areas where the
screen picks up reflections and the
left thumb moves around. Notice
that the tracking markers aren’t even
used; there is plenty of other detail for
Mocha to track without them.

Retracking with these additional hold-out masks improves
the track a good amount; all that is required to perfect
this track is a keyframe at one point where the unit is tilted
about 15 degrees toward camera. I look for the frame where
the plane settles into its new, offset position and highlight
one corner of the main track mask (not the hold-outs).

259


When I highlight the point, zoomed views with crosshairs
appear at the upper-left or right corner (Figure 8.25).
At the right under Keyframe Controls, Autokey is on by
default. This is where it becomes helpful to have aligned
mask points with recognizable features, as the idea is to
move the points to match the adjacent position. This cre-
ates a green keyframe along the main Timeline. Mocha
uses these keyframes as extra points of comparison, rather
than simply averaging their positions.
You need not rely on luck and manual labor to get points
to match over time. Open the AdjustTrack panel. Under
Nudge are controls allowing Mocha to automatically guess
where the point should move to match the Master, along
with controls to move it in ﬁne increments. If Mocha isn’t
ﬁnding your point automatically, you can try adjusting the
neighboring AutoNudge controls. If it seems as though
you’ve started with a reference point that isn’t representa-
tive, you can reset the Master under Reference Point.

Figure 8.25 Mocha relies on you to fix the position of the mask vertex by hand if it slips, helpfully offering close-up views to
allow you to match the cross-hairs to the same feature, automatically creating a new keyframe when the Autokey button is
on (in red, at the right).

260


In this example it’s also helpful to check Perspective under
Motion in the Track tab; this allows the change in propor-
tions from the tilting of the screen to be included in the
Corner Pin export. Still in beta at this writing, the Red
Giant Corner Pin effect included in
Figure 8.26 The Red Giant Corner its Warp collection is ready-made for
Pin effect not only includes a usage with Mocha (Figure 8.26).
Mocha Import function, up top, it
allows “from” as well as “to” pins—
so your corner pin content can be
tracked from a moving source as
well—and even adds motion blur
capabilities. It puts After Effects’
built-in Corner Pin tool to shame.

If you get into trouble, you’ll want to know how to delete
keys (under Keyframe Controls) or reference points (in
the AdjustTrack tab). You also need to know a few new
keyboard shortcuts, such as X for the hand tool and arrow
keys to navigate one frame. This discussion omits the many
actual compositing features included with Mocha (the
Layer Controls), which you assumedly don’t need, because
you have After Effects.

Use Tracking with Expressions
Expressions and tracking data go together like peanut
butter and chocolate. You don’t even have to apply raw
tracking data in order to put it to use; you can pickwhip
any property containing X and Y position data directly to
other properties.
For example, to track in a paint clone operation in a
single layer
1. Set up a track with the paint target as the feature center
(the center of the Feature Region).

261


2. Move the attach point to the area from which you wish
to clone.
3. Track motion; you can set Track Type to Raw or simply
don’t apply it.
4. Add a clone stroke with appropriate settings.

5. Pickwhip Stroke Options > Clone Position to the Attach
Point and Transform: Clone 1 > Position to the Feature
Center.
This works equally well applied to another layer.

Continue Loop
Sometimes a track point will disappear before the track is
completed, either because it is obscured by a foreground
object or because it has moved offscreen. Mocha can
generally deal with this—if any part of the tracked plane
remains in frame, it is tracked.
Nonetheless it’s useful to do this right in After Effects. First
make certain there are no unwanted extra tracking key-
frames beyond where the point was still correctly tracked;
this expression uses the difference between the final two
keyframes to estimate what will happen next.
Reveal the property that needs extending (Position in this
case), and Alt/Option-click on its stopwatch. In the text
field that is revealed, replace the text (position) by typing
loopOut(“continue”). Yes, that’s right, typing; don’t worry,
you’re not less of an artist for doing it (Figure 8.27).
This expression uses the delta (velocity and direction) of
the last two frames. It creates matching linear motion (not
a curve) moving at a steady rate, so it works well if those
last two frames are representative of the overall rate and
direction of motion.
Chapter 10 offers many more ideas about how to go
beyond these simple expressions and to customize them
according to specific needs.

262


Figure 8.27 A continue loop is handy
anywhere you have motion that
should continue at the pace and
in the direction at the first or last
keyframe. Notice in this example that
although it could help as the skater
disappears behind the post, it doesn’t
do curves; motion continues along a
linear vector.

Import 3D Tracking Data
After Effects does not include a dedicated 3D tracker, but
it can import 3D tracking data created in a separate appli-
cation, such as 2D3’s Boujou, Pixel Farm’s PF Track, or
SynthEyes, from Andersson Technologies. Nor does it work
with 3D mesh objects (it’s the “postcards in space” model
of 2D in a 3D environment), so the 3D tracking workflow
operates as follows:
1. Track the scene with a 3D tracking application. This
generates 3D camera data, typically exportable in vari-
ous 3D formats; included with these is usually an option
to export a .ma file for After Effects.
The track will ideally also generate nulls corresponding
to the track points and an axis centered on the “floor,”
the ground plane of the shot, although you may have to
specify these.

263


2. (Optional) Import the camera data into a 3D anima-
tion program and render 3D elements. You can also
massage the 3D data for After Effects here, leaving, for
After Effects can also extract camera example, only helpful nulls.
data embedded in an RPF sequence
3. Import the camera data into After Effects; it imports in
(and typically generated in 3ds
Max or Flame). Place the sequence the form of a composition with an animated 3D camera
containing the 3D camera data in and nulls (potentially many layers of them). Add your
a comp and choose Animation > background plate and foreground elements; those that
Keyframe Assistant > RPF Camera are 2D can be freely matched with 3D elements, which
Import.
will pick up the camera movement.
Figure 8.28 shows the final result from the baseball plate;
the camera follows the pitch all the way to the plate, where
the batter hits it out of the park (of course). This shot is
a complete mishmash of 2D and 3D elements. The ball,
field, and front of the stands are computer generated, but
the crowd was lifted from footage of an actual game.

Figure 8.28 True 3D tracking (done in 2D3’s Boujou for this shot) is essential to this composite because the feet of the
characters must be locked to the computer-generated ground, and a camera on dolly tracks wobbles and bounces. The full
shot follows the ball to the plate.

Figure 8.29 shows a completely different type of shot that
also began with a 3D track in Boujou. The fires that you
see in the after shot are actually dozens of individual 2D
fire and smoke layers, staggered and angled in 3D space
as the camera flies over to give the sense of perspective.
You’ll find more on this shot and how it was set up in
Chapter 14.

264


Figure 8.29 Just because you’re stuck with 2D layers in After Effects doesn’t mean you can’t stagger them all over 3D space
to give the illusion of depth, as with this fly-by shot. Tracking nulls from Boujou helped get the relative scale of the scene,
this was important because the depth of the elements had to be to exact scale for the parallax illusion to work. (Final fire
image courtesy of ABC-TV.)

3D Tracking Data
Many people don’t realize that After Effects can import
Maya scenes (.ma ﬁles); but they have to be properly
prepped and include only rendering cameras (with transla-
Because After Effects offers no pro-
tion and lens data) and nulls. The camera data should be portional 3D grids in the viewers,
“baked,” which is Maya parlance meaning that it should nulls imported with a 3D scene can
have a keyframe at every frame (search on “baking Maya be essential to scale and position
camera data” in the online help for speciﬁcs on this). elements in 3D. The scale of the
scene is otherwise arbitrary.
3D trackers operate a bit differently than the After Effects
tracker. Generally you do not begin by setting tracking
points with these; instead, the software creates a swarm of
hundreds of points that come and go throughout the shot,
and it “solves” the camera using a subset of them.
Besides Position and Rotation, the Camera may also
contain Zoom keyframes. Unless Sergio Leone has started
making spaghetti westerns again, zoom shots are not the
norm and any zoom animation should be checked against
a camera report (or any available anecdotal data) and
eliminated if bogus (it might be a push or even an unstable
camera). Most 3D trackers allow you to specify that a shot
was taken with a prime lens (no zoom).

265


Import a Maya Scene
You import a .ma scene the same way you would any
element; make sure it has the .ma extension to be recog-
nized properly. After Effects will import either one or two
compositions: one for a Maya project with a square pixel
aspect ratio, and two for nonsquare (a square pixel version
is nested in the nonsquare one).
Your camera may be single-node (in which case the camera
holds all of the animation data) or targeted, in which case
the transformation data resides in a parent node to which
Instead of exporting a .ma scene,
Maxon’s Cinema 4D software offers the camera is attached.
its own support for After Effects Any null object with the word “null” in its name is also
integration via a plug-in available
from Maxon (www.maxon.net). imported. Depending on your tracker and your scene, you
may import so many nulls with the scene that it becomes
cumbersome. A composition with 500 layers, even nulls,
quickly becomes unwieldy, so if possible, weed out the
useless nulls, paring it down to a couple dozen of them
(adding descriptive names) in the tracking software or 3D
program. It’s usually easy to make out what the nulls cor-
respond to in the scene if you watch them over the back-
ground plate; they tend to cluster around certain objects.
If all of those nulls make their way into After Effects, ﬁnd
the dozen or two that are useful to you in the scene, and
select those in the Timeline along with the camera and its
parent null (if any). Context-click on the selected layers
and choose Invert Selection to select the potentially hun-
dreds of other unused nulls. Delete them.
If the nulls come in with tiny values—say, all in the single
digits—you can add a null, make it 3D and set its Position
to 0,0,0, parent all layers of the imported Maya comp to it,
3D tracking data uses 0,0,0 as a
center point; as you create new 3D
and scale up the null. Now hold down the Option/Alt key
layers in After Effects, zero out their and un-parent all of those layers; the scaled values stick.
position values to match. It’s much You can also use this method to invert a scene that comes
simpler to work this way than to in upside-down.
translate the tracked scene to the
comp center.
Try It Out for Yourself
If you’d like to try out a 3D tracker, look no further than
the book’s DVD-ROM. It includes a demo of SynthEyes, a
reasonably priced 3D tracker from Andersson Technologies

266


that has been used on feature ﬁlms (Figure 8.30). For about
the cost of a typical After Effects plug-in set, you can own
your own 3D tracking software (Mac or Windows). The
demo is the full version but with a time limit, and projects
cannot be saved, so you must execute and export your track
start to ﬁnish before quitting. The output data, however, is
fully usable.

Figure 8.30 SynthEyes is one of several third-party software options that will
output a Maya .ma file suitable for import as an After Effects composition. The
tracking markers are included as null objects that can help orient you when you
place objects into the resulting 3D space.

267


There are also sample ﬁles to try out before you create a
scene of your own. You may have luck simply importing
your shot into SynthEyes, clicking Full Automatic, and
exporting the result as AfterEffects via .ma. If there’s more
To learn more about the complex
involved in getting a good track, however, you will need
art of match moving, check out
Matchmoving: The Invisible Art of to learn a bit more about how the software works, and it’s
Camera Tracking (Sybex Inc.) by beyond the scope of this chapter to document it. SynthEyes
Tim Dobbert, a colleague from The is the type of application that yields much better results
Orphanage. if you carefully read the online documentation, which is
available from the Help menu.

Conclusion
Despite all attempts to make it standardized and auto-
matic, tracking remains as much art as it is science, which
is probably why most large effects facilities retain a staff of
match movers. Even if you’ve understood everything in this
chapter and followed along closely, working on your own
shots will open a process of trial and error.
The next chapter will delve further into the ways in which
After Effects can replicate what a physical camera can do,
expanding on some of the concepts touched on earlier in
the “2.5D Tracking” section.

268

CHAPTER

9
The Camera and Optics

A film is never really good unless the camera is an eye
in the head of a poet.
—Orson Welles

The Camera and Optics

I t seems as if visual effects is all about simulating the look
of the real world, but that’s not quite the goal; as a visual
effects compositor, your actual job is to simulate the real
world as it appears through the lens of a camera. The distinc-
tion is critical, because when photographed the world
looks different—more or less real, and possibly both.
It’s not too grandiose to say that cinematography is essen-
tial to compositing, because After Effects offers the oppor-
tunity to re-create and even change essential shooting
decisions long after the crew has struck the set and called
it a wrap. Your shot may be perfectly realistic on its own
merits, but it will only belong in the story if it works cin-
ematically. Factors in After Effects that contribute to good
cinematography include
. Field of view
. Depth of focus
. The shooting medium and what it tells about the
storyteller
. Planar perspective and dimensionality
. Camera motion (handheld, stabilized, or locked) and
what it implies about point of view
These seemingly disparate points all involve understand-
ing how the camera sees the world and how film and
video record what the camera sees. All of them transcend
mere aesthetics, influencing how the viewer perceives the
story itself.

270


Cameras: Virtual and Real
We begin our exploration of virtual cinematography with
the After Effects camera, which relates closely to an actual
motion picture camera without actually being anything like
one. Following is an examination of how 3D operates in
After Effects and how the application’s features—not only
the camera, but also lights and shading options—corre-
spond to real world counterparts.

See with the Camera
Toggle a layer to 3D and voila, its properties contain three
axes instead of two—but enabling 3D without a camera is a
little bit like racing a car with automatic transmission: You
can’t really maneuver, and before long you’re bound to
slam into something.
The Camera Settings dialog (Figure 9.1) uniquely includes a
physical diagram that helps tell you what you need to know
about how settings in the 3D camera affect your scene.

Figure 9.1 Artists love a visual UI, and
the Camera Settings dialog provides
one to help elucidate settings that
might otherwise seem a bit abstract.
The 50 mm preset selected in the
Preset menu is the neutral (default)
setting; use it when you want no
change in perspective by the simple
act of adding a camera.

Lens Settings
Although it is not labeled as such, and although After
Effects displays previous camera settings by default, the
true default lens preset in Camera Settings is 50 mm. This

271

Chapter 9 The Camera and Optics

setting (Figure 9.2) is neither wide (as with lower values,
Figure 9.3) nor long (as with higher values, Figure 9.4);
and it introduces no shift in perspective.

Figure 9.2 The default lens (50 mm setting). If the Z Position value is the exact inverse of the Zoom value, and all other
settings are at the default, this is the view you get, and it matches the appearance of setting no After Effects camera
whatsoever.

Figure 9.3 The extreme wide or fisheye lens pointed inside an evenly proportioned 3D box. Note that the “long” look of
the box is created by this “wide” lens, which tends to create very strange proportions at this extreme. A physical lens with
anything like this angle would include extremely distorted lens curvature.

Figure 9.4 A telephoto lens (using the 200 mm setting) pushes items together in depth space, shortening the distance
between the front and back of the box dramatically.

“50 mm” is a virtually meaningless term because vir-
tual space doesn’t contain millimeters any more than it

272


contains kilograms, parsecs, or bunny rabbits. This is the
median lens length of a 35 mm SLR camera, the standard
professional still image camera.
Motion picture cameras are not so standardized. The
equivalent lens on a 35 mm film camera shooting Acad-
emy ratio itself has a 35 mm length. A miniDV camera, on
the other hand, has a tiny neutral lens length of around
4 mm. The length corresponds directly to the size of the
backplate or video pickup, the area where the image is
projected inside the camera.
Lens length, then, is a somewhat arbitrary and made-up
value in the virtual world of After Effects. The correspond-
ing setting that applies universally is Angle of View, which
can be calculated whether images were shot in IMAX or
HDV or created in a 3D animation package.

Real Camera Settings
To understand the relationship of the After Effects camera
to those of a real-world camera, look again at the Camera
Settings diagram introduced in Figure 9.1. Four numeri-
A fifth numerical field in Camera
cal fields—Film Size, Focal Length, Zoom, and Angle of Settings, Focus Distance, is enabled
View—surround a common hypotenuse. by checking Enable Depth of
Field; it corresponds to a camera’s
A prime (or fixed) lens would have static values for all four. aperture setting.
A zoom lens would of course work with a fixed Film Size,
but would allow Zoom and Focal Length to be adjusted,
changing the Angle of View. These four settings, then, are
interrelated and interdependent, as the diagram implies,
and the relationship is just the same as with a real camera,
which the Film Size can even help emulate. Lengthen the
lens by increasing Focal Length and you decrease Angle
of View.
The settings you actually use are Zoom (to animate) and
Angle of View (to match the real-world source).
Angle of View is the radius, in degrees, that fits in the view.
If you’re matching it, note that Camera Settings lets you
specify a horizontal, vertical, or diagonal measurement in
the Measure Film Size menu.

273


In After Effects, the Zoom value is the distance of the
camera, in pixels, from the plane of focus. Create a camera
and its default Z Position is the inverse of the Zoom value,
perfectly framing the contents of the comp at their default
Z Position, 0.0 (Figure 9.5). This makes for easy reference
when you’re measuring depth of field effects, and it lets
you link camera position and zoom together via expressions
(for depth of field and multiplane effects, discussed later).

Figure 9.5 Comp Size (at the right) is
the frame size, in pixels. The diagram
shows it as vertical (right), but here
it is measured horizontally accord-
ing to Measure Film Size (left). Units
can be inches or millimeters, helpful
when matching a physical camera
(described later in this chapter).

Emulate a Real Camera
Other considerations when matching a real-world camera
include
. Depth of field: This is among the most filmic and
evocative additions you can make to a scene. Like any
computer graphics program, After Effects naturally
has limitless depth of field, so you have to re-create the
shallow depth of real-world optics to bring a filmic look
to a comp.
. Zoom or push: A move in or out is used for dramatic
effect, but a zoom and a push communicate very differ-
ent things about point of view.
. Motion blur and shutter angle: These are composition
(not camera) settings; introduced in Chapter 2 and
further explored here.
. Lens angle: The perspective and parallax of layers in 3D
space change according to the angle of the lens used to
view them.

274


. Lens distortion: Real lenses introduce curvature to
straight lines, which is most apparent with wide-angle
or “fisheye” lenses. An After Effects camera has no lens,
hence, no distortion, but it can be created or removed
(see the section “Lens Distortion”).
. Exposure: Every viewer in After Effects includes an
Exposure control (Aperture icon, lower right); this
(along with the effect with the same name) is mathe-
matically similar but different in practice from the aper-
ture of a physical camera. Exposure and color range is
detailed in Chapter 11.
. Boke, halation, flares: All sorts of interesting phenom-
ena are generated by light when it interacts with the
lens itself. The appeal of this purely optical phenom-
enon in a shot is subjective, yet they can offer a unique
and beautiful aesthetic and lend realism to a scene shot
under conditions where we would expect to see them
(whether we know it or not).
A camera report is a record of the settings used when the
footage was taken, usually logged by the camera assistant
(or equivalent).

The Camera Report
With accurate information on the type of camera and the
focal length of a shot, you know enough to match the lens
of that camera with your After Effects camera.
Table 9.1 on the next page details the sizes of some typical
film formats. If your particular brand and make of camera A potentially easier alternative to
the listed steps, for those who like
is on the list, and you know the focal length, use these to using expressions, is to use the fol-
match the camera via Camera Settings. The steps are lowing expression on the camera’s
Zoom property:
1. Set Measure Film Size to Horizontally. (Note that FocalLength = 35 //
hFilmPlane stands for “Horizontal Film Plane.”) ➥change to your value,
➥in mm
2. Set Units to Inches. hFilmPlane = 24.892
➥//change to film size,
3. Enter the number from the Horizontal column of the
➥in mm (horizontal);
chart that corresponds to the source film format. ➥multiply values in
➥inches by 25.4
4. Set Units to Millimeters. this_comp.width*(Focal
➥Length/hFilmPlane)
5. Enter the desired Focal Length.

275


TABLE 9.1 Typical Film Format Sizes
FORMAT HORIZONTAL VERTICAL
Full Aperture Camera Aperture 0.980 0.735
Scope Camera Aperture 0.864 0.732
Scope Scan 0.825 0.735
2:1 Scope Projector Aperture 0.838 0.700
Academy Camera Aperture 0.864 0.630
Academy Projector Aperture 0.825 0.602
1.66 Projector Aperture 0.825 0.497
1.85 Projector Aperture 0.825 0.446
VistaVision Aperture 0.991 1.485
VistaVision Scan 0.980 1.470
16 mm Camera Aperture 0.404 0.295
Super-16 Camera Aperture 0.493 0.292
HD Full 1.78 0.378 0.212 (Full Aperture in HD 1.78)
HD 90% 1.78 0.340 0.191 (90% Safe Area used in HD 1.78)
RED One Mysterium 0.960 0.539
Courtesy of Stu Maschwitz The Orphanage

Once the Angle of View matches the footage, any objects
that you track in (perhaps using techniques described in
Chapter 8) maintain position in the scene as the shot pro-
gresses. It’s vital to get this right when the camera moves
during the shot, especially if a particularly wide or long
lens was used.

Lens Distortion
A virtual camera with a wide-angle view (like the one back
in Figure 9.2) has a dramatically altered 3D perspective,
but it has no lens to distort objects the way a real camera
lens does. A virtual camera can widen the view area in a
completely linear fashion.

276


Vertical Viewing Angle
42.305

Focus Distance

Tilt 14.603
True Horizon Right VP
If you’re lacking Angle of View data
Left VP
from a shoot, it can be cleverly
6⁄.3073*7.5191=146.8 derived (Figure 9.6).

Focus Distance
Vanishing Point
Arc

Horizontal Viewing Angle
54.575
=34.893mm Lens (in 35mm terms)

Figure 9.6 This diagram shows the result of calculating the view angle and thus
the lens length used to take this shot. Check out the 12-minute video description
by Mike Seymour of how this was done (on the book’s disc). Image and video
excerpted from the Background Fundamentals course at fxphd.com.

A physical lens curves light to project it properly across the
camera backplate with physical width and height, no mat-
ter how small. To show up properly the reflected imagery
must be perpendicular to the surface of the lens glass, so
a wide-angle view requires not only a short lens length but
also a convex lens in order to gather the full range of view.
At the extremes, this causes easily visible lens distortion;
items in the scene known to contain straight lines don’t
appear straight at all, but bent in a curve (Figure 9.7). In a
fisheye lens shot, it’s as if the screen has been inflated like
a balloon.

Figure 9.7 The nearly psychedelic
look of extreme lens distortion; the
lens flare itself is extremely aberrated.
You can create just as wide a lens
with the 3D camera, but there would
be no lens distortion because there
is no lens.

277


As you refine your eye, you may notice that many shots that
aren’t as extreme as a fisheye perspective contain a degree
of lens distortion. Or you might find that motion tracks
match on one side of the frame but slip on the opposite
side, or proportions go out of whack, or things just don’t
quite line up as they should (Figure 9.8).
There’s no way to introduce lens distortion directly to a
3D camera, but the Optics Compensation effect is
designed to add or remove it directly on the 2D layer or
stack. Figure 9.9 shows this effect in action. Increasing the
Field of View makes the affected layer more fish-eyed in
Figure 9.8 It is simply not possible to appearance; to correct a shot coming in with lens distor-
make all four corners and edges of a tion, check Reverse Lens Distortion and raise the Field of
yellow solid line up properly with the
side of a distorted building. View (FOV) value.

Figure 9.9 Optics compensation takes
place in a composition larger than the
source; the padding gives the corners
of the image space. The Beam effect
can serve as a virtual plumb line, or
you can use Ctrl/Command+“ to
superimpose a grid.

This process is not exactly scientific. You typically eye-
match the setting, because Field of View doesn’t corre-
spond to measurable data in the camera report, such as the
Lens Angle.
1. Having identified lens distortion (Figure 9.8), precomp
the background into a new composition that is at least
20% larger than the plate (to allow room at the borders
to distort).

278


2. Add an adjustment layer above the plate layer, and
apply Optics Compensation to that layer. With a wide
angle lens, check Reverse Lens Distortion and raise
the Field of View (FOV) setting until all straight lines
appear straight.
3. Add a Beam effect to the adjustment layer (below the
Optics Compensation effect, unaffected by it). To get
away from the distracting light saber look, match Inside
Color and Outside Color to some easily visible hue,
then align the Starting Point and Ending Point along
an apparently straight line near the edge of the frame.
Fine-tune the Field of View setting a little more until
the line is plumb (Figure 9.9).
4. Precompose all of these layers and set this new compo-
sition as a guide layer. In Figure 9.10, you can see that
the corner pin is now successful.

Figure 9.10 Over the undistorted
background plate, you can freely
position, animate, and composite ele-
ments as if everything were normal.
Note that the perspective is still that
of a very wide-angle lens, but without
the curvature.

5. To complete the shot, restore the original ﬁeld of view,
including distortion. First create a new comp with
the original background plate (no Optics Compensa-
tion) and the precomp with the assembled foreground
elements.

279


6. Copy Optics Compensation with the settings added in
step 2 and paste it to the foreground element. Toggle
Reverse Lens Distortion off. The Field of View of the
background is restored, but the foreground elements
match (Figure 9.11).
Here is an original Stu Maschwitz haiku summing it up:
undistort, derive
reunite distorted things
with an untouched plate

Figure 9.11 The composited layer is
distorted to match the curvature of
the original background.

2D and 3D
The point of matching a 3D lens angle is typically to
composite in 3D over a 2D plate background. After Effects
helps this process in the following ways:
. A 2D background layer remains in place no matter
what happens with the camera and 3D layers.
. 2D adjustment layers set to comp size and default posi-
tion affect all layers below them, including 3D layers.
. 3D layers can use 2D Blending Modes (over 2D
elements, they obey layer order, and with other
3D elements, z-space depth).

280


You just need to be careful:
. Joining a 3D track matte to a 3D layer is almost never
the right idea, particularly with an animated camera. A
3D layer can use a 2D layer as a track matte; it is applied
just as it would be to a 2D layer. A 2D layer can use a 3D
layer as a track matte; the 3D perspective of the track
matte renders ﬁrst and is then applied. But combine
two 3D layers in this manner and the matte is ﬁrst trans-
lated by the camera perspective once, then applied, and
then effectively translated again as the affected layer
also obeys camera perspective.
. Paradoxically, the only layers in After Effects that them-
selves can contain true 3D effects are 2D layers (which
may nonetheless make use of the 3D camera perspec-
tive, Figure 9.12).

Figure 9.12 Particles generated by
Trapcode Particular fill the volume of
3D space, as is evident in a perspec-
tive view, although the effect is
applied to a 2D layer.

. A precomped set of 3D layers behaves like a single 2D
layer unless Collapse Transformations is enabled on
that layer. As described back in Chapter 4, this toggle
passes through all 3D data from the precomp as if those
layers lived right in the master composition.
Each point provides advantages provided you understand
how the image pipeline works.

281


Photoshop CS4 3D
One of the most intriguing new features of the CS4 version
of Adobe software is the ability to bring 3D models into
After Effects as Photoshop 3D layers. You might expect
this feature to radically change what can be done with an
After Effects 3D scene; however, there are many limitations
to its use—enough, in fact that it may rarely be worth the
trouble.
To really take advantage of using 3D models directly in
After Effects, you would want them to preview quickly
(perhaps using OpenGL), to allow ﬁne tuning of lighting,
texturing, anti-aliasing, and motion blur, not to mention
animating or deforming the mesh itself. Instead
. You can control texture, lighting, and anti-aliasing only
in Photoshop CS4.
. To adjust Photoshop-only features (even changing anti-
aliasing from Draft quality so you can work quickly to
Best quality for rendering) you must choose the image
layer and use Edit Original (Ctrl/Command+E).
. After Effects lighting, material options, and motion
blur have no effect.
Figure 9.13 shows a 3D model imported into After Effects;
although the source Photoshop ﬁle has a single layer, the
After Effects Comp contains three: a camera, a Control-
ler layer, and the 3D image itself. You can replace or even
eliminate the camera layer, but the other two must remain
together or the layer becomes ordinary again, like Cinder-
ella after midnight.
To transform the 3D object, you work with the controller
layer, a null. You can apply any standard image effects to
the layer that contains the image itself. More fundamental
changes to the appearance of the model are hardly more
convenient than if it were rendered in third-party software
such as Maya; lighting and surface changes can only hap-
pen in Photoshop.

282


Figure 9.13 The Photoshop Import dialog has changed to accom-
modate Live Photoshop 3D layers; to get this option, import the
PSD as a composition. The resulting comp contains a camera,
the image and a controller layer; the image has a Live Photoshop
3D effect applied to it, which links it to the Controller via a set of
expressions (in red).

283


The lack of motion blur seems like the main deal-killer;
there are workarounds if you’re willing to take the render
hit. You can try adding an adjustment layer at the top of
the comp containing your 3D animation and then
. Apply the Timewarp effect to that layer. Change speed
to 100 and toggle Enable Motion Blur, then set the
other Motion Blur settings to get the motion blur look
you want.
. For a less render-intensive approach, you can try apply-
ing CC TimeBlend; it won’t work with heavy motion
and is a bit eccentric to preview (if it looks strange, try
hitting the Clear button at the top of the effect and
re-previewing).
These are the same workarounds you would use if for
some reason your 3D render had no motion blur; it’s a
less accurate and, especially in the case of Timewarp, more
render-intensive approach. More about using Timewarp to
generate motion blur can be found in Chapter 2.

Storytelling and the Camera
Locked-off shots are essential to signature shots by Welles,
Hitchcock, Kubrick, and Lucas, among others, but they’re
the exception, not the norm. In contemporary films in
Always keep in mind where the
audience’s attention is focused—
particular, the camera point of view often implies a human
a successful visual effects artist point of view, hence the popularity of the hand-held shot
employs the magician’s technique, since the 1970s.
misdirection, all the time.
In the bad old days of optical compositing, it was scarcely
possible to composite anything but a static camera point
of view. Nowadays, most directors aren’t satisfied being
limited to locked-off shots, yet the decision to move the
camera might not happen on set, or it might have to be
altered in post-production.
It’s helpful to create a rough assemble with camera anima-
tion as early in the process of creating your shot as pos-
sible, because it will tell you a lot about what you can get
away with and what needs dedicated attention. The “Sky
Replacement” section in Chapter 13 contains an example
in which a flat card stands in for a fully dimensional skyline

284


(Figure 9.14). The audience should instead be focused
on watching the lead character walk through the lobby,
wondering what he has in his briefcase; if not, the ﬁlm
has more problems than can be ﬁxed with more elaborate
visual effects elements.

Figure 9.14 Prominent though it may
appear in this still image, the audience
isn’t focused on that San Francisco
skyline outside the window. There’s
no multiplaning as the camera moves
because the background skyline is a
still image; no one notices because
viewer attention is on the foreground
character. (Image courtesy of The
Orphanage.)

Camera Animation
The most common confusion about the After Effects cam-
era stems from the fact that by default, it includes a point
of interest, a point in 3D space at which the camera always
points, for auto-orientation. The point of interest is fully
optional, yet the setting is among the least discoverable in Figure 9.15 So many After Effects 3D
After Effects. To clarify camera tragedies could have been
avoided had only more users known
. Layer > Transform > Auto-Orient (Ctrl+Alt+O/ about this dialog box (Ctrl+Alt+O/
Cmd+Option+O). By disabling auto-
Cmd+Option+O) contains the toggles to disable auto-
orientation, you are free to move the
orientation and the point of interest (making the cam- camera anywhere without changing
era a free camera, Figure 9.15). its direction.

. In that same dialog, you can instead orient the camera
along its path of camera motion so that its rotation
maintains tangency; in other words, it follows the path.
. You might want to use the point of interest but trans-
form it and the camera together. To do this, don’t The new Unified Camera Tool (C)
attempt to match keyframes for the two properties— lets you use a three-button mouse
this is sheer madness! Parent the camera to a null and to orbit, track, and zoom the
translate that instead. camera without having to cycle
through the tools.

285


. Orientation works differently depending on whether
auto-orientation is on (causing it to revolve around
the point of interest) or not (in which case it rotates
around its own center).
. The auto-oriented camera always maintains an upright
position; cross over the X/Y plane above the center
and the camera ﬂips. To avoid this behavior, use a free
The Y axis is upside down in
After Effects 3D, just as in 2D; an camera.
increased Y value moves a layer The preceding points come into play only with more elabo-
downward. rate camera animations; more modest use of the 3D cam-
era, such as a simple camera push, raises other questions.

Push and Zoom
A camera push moves the camera closer to the subject; a
zoom lengthens the lens while the camera remains station-
ary. Figure 9.16 demonstrates the difference in perspective,
which is just as noticeable with multiple 3D elements in
After Effects as with objects in the real world. The zoom
has a more extreme effect on the foreground/background
composition of the shot and calls more attention to the
camera itself.

Figure 9.16 Frame a similar shot with a long (left) and wide (right) lens and you see the difference between a zoom and a
push. A zoomed image has a flattened perspective.

Dramatic zooms for the most part had their heyday in
1960’s-era Sergio Leone movies, although they have
recurred any time a ﬁlmmaker wants the live feel of a cam-
era operator reaching quickly for a shot. And that’s really
the point; because your eye does not zoom, this move calls

286


attention to the camera apparatus itself, and to the camera
operator. Its use is therefore limited.
The push, on the other hand, is a dramatic staple. The
question when creating one in After Effects is, does it
require a 3D camera when you can simply scale 2D layers?
Scaling a 2D layer (or several, parented to a null) works for
a small move; however, to re-create progression through
Z space, the rate of scaling must increase logarithmically,
Animation > Keyframe Assistant >
which complicates things. Not only does a 3D camera move Exponential Scale is the old-school,
create this effect naturally, it makes it simple to add eases, pre-3D way to fake the illusion
stops, and starts, a little bit of destabilization—whatever of a camera move on a 2D layer.
works. There is no good reason to employ
this feature when you can instead
Natural camera motion will contain keyframe eases (Chap- animate a 3D camera.
ter 2), for the human aspect. A little bit of irregularity lends
the feeling of a camera operator’s individual personality
(Figure 9.17), or even dramatic interest (hesitation, cau-
tion, intrigue, a leap forward—the possibilities are many).

Figure 9.17 The Graph Editor shows
where you’ve created organic motion
in ease curves, although the smooth-
ness of this camera push as it eases to
a stop may itself lack that extra human
imperfection, which would also show
up in the curves.

A move in or out of a completely 2D shot can easily look
wrong due to the lack of parallax. Likewise, all but the
subtlest tracking and panning shots, crane-ups, and other
more elaborate camera moves will blow the 2.5D gag, so if
you do them, that had better be your intention.

287


Certain types of elements—soft, translucent organic
shapes, such as clouds, fog, smoke, and the like—can be
layered together and staggered in 3D space, fooling the eye
into seeing 3D volume. Chapter 13 shows how to do this.
Do you have a bunch of coplanar
layers you’re making 3D just so you
can push in on them? Precomp Camera Projection
them together first to avoid little Camera projection (or camera mapping) typically begins with
rounding errors that can easily a still photo, which is then projected onto 3D objects that
occur where they overlap in 3D.
match the dimensions and placement of objects in the
photo. You can then move the 3D camera—typically only
along the Z axis—providing the illusion that the photo
is fully dimensional (right up until things start to tear
and break as areas and perspectives that aren’t there are
revealed).
Figure 9.18 shows a camera projection that ambitiously
features two military vehicles parked in the foreground.
A dozen separate white solids with masks were created
to form a crude 3D model, ready to receive a projected
image (Figure 9.19). This example shows both the magic
of this technique—deriving perspective shifts from a ﬂat,
still image—and the associated problems of image tearing
when an area of the frame is revealed that had previously
been obscured in the source photo.
The setup is key: How is it that the one “texture” (the
photo) sticks to the 3D objects? The fundamental concept
is actually relatively simple; it’s a virtual version of a slide
projector and a bunch of angled screens. Getting the setup
right is a question of managing details, and that part is
fairly advanced and not for the faint of heart (which is why
mention of a third-party option follows this description).
The steps to projecting any still image into 3D space (an
example of which, 09_cameraProjection.aep, can be found
on this book’s disc) are as follows:
1. Begin with an image that can be modeled as a series of
planes.
2. Create a white solid for each dimensional plane in
the image. Enable 3D for each, and under Material
Options, change the Accepts Lights option to Off.

288


Figure 9.18 The progression from the source image (left) through the camera move. By the final frame (right), image warp-
ing and tearing are evident, but the perspective of the image is essentially correct for the new camera position. The tearing
occurs simply because as the camera moves it reveals areas of the image that don’t exist in the source.

3. Add a camera named Projection Cam; if you know the
Angle of View of your source image, add that value.
4. Add a Point light called Projector Light. Set its position
to that of Projection Cam, then parent it to Projection
Cam. Set Casts Shadows to On.
5. Duplicate the source image, naming this layer Slide.
Enable 3D, and in Material Options, change Casts
Shadows to Only and Light Transmission to 100%.
6. Slide not located properly? Add a null object called
Slide Repo; set its position to that of Projection Cam,
and parent it to Projection Cam. Now parent Slide to
it, and adjust its scale downward until the image is cast
onto the white planes, as if projected.
7. Now comes the challenge: masking, scaling, and repo-
sitioning those white solids to build the model, ground
plane, and horizon onto which the slide is projected.
Toggle on the reference layer and build your model to
match that, checking it with the slide every so often.
Figure 9.19 This shows the rather
8. If planes that you know to be at perpendicular 90 complicated setup for this effect, now
aided by a script on this book’s disc.
degree angles don’t line up, you need to adjust the From the top and side views you can
Zoom value of the Projection Cam, scaling the model see the planes that stand in for the
and slide as needed to match the new Zoom value. vehicles and the orange cone, which
appears stretched along the ground
The example ﬁle includes an expression applied to the plane.
Scale value of the Slide layer so that the slide scales up
or down to match however you adjust the Zoom of the
camera, which is not necessary but is helpful.

289


9. Once everything is lined up, duplicate Projection Cam,
and rename the duplicate (the one on the higher layer)
Anim Cam. Freely move this camera to take advantage
of the new dimensional reality of the scene.
10. To get everything looking sharp, go to Composition
Settings > Advanced > and click Options; boost the
Shadow Map Resolution from its default Comp Size to
This edition of the book newly
includes a script that Jeff Almasol a higher number as needed (check the image at Full
and I designed to accomplish the resolution to see if it becomes sharper). I tend to max
basic camera projection setup out this setting and take the render hit.
automatically. You can find it on
the disc. The best way to learn about this is probably to study the
example file included on this book’s disc.

Camera Blur
Camera blur, also known as lens blur, results from objects
in a scene falling outside focal range. It has unique char-
acteristics when compared with an ordinary image blur
operation.
Ironically, although the standard consumer video camera
has a very wide depth of field and rarely produces lens
blur, this is considered a major disadvantage of video to
most filmmakers. Narrow focal depth not only produces a
beautiful image one would almost automatically call “cin-
ematic,” it focuses the viewer’s attention and thus provides
the director with a powerful storytelling tool.
It can be a lot of work to re-create depth of field effects in
After Effects; it’s better to get them in camera if possible.
Nonetheless you can create specific cinematic blur effects
such as a rack focus shot, in which the plane of focus shifts
from a subject at one distance to another. This is a device
to create anticipation and change the object of attention
while creating a beautiful aesthetic.
Limited focal range is a natural part of human vision, but
camera lenses contribute their own unique blur charac-
teristics that in the contemporary era are often consid-
ered aesthetically pleasing the world over. There is even a
Japanese term (literally meaning “fuzzy”) to describe the

290


quality of out-of-focus light as viewed through a lens, boke
(also bokeh, more phonetic but clunkier).
You can create these effects and more in After Effects,
using a combination of tools built in to the software, third-
party tools to support the process, and a careful observa-
A solid description of boke with
tion of optics and nature. links lives on the Web at http://
en.wikipedia.org/wiki/Bokeh.
Image Planes and Rack Focus
Any time you can divide a shot into distinct planes of depth
with each plane as its own layer, you can rack focus. All you
need is a focal point to animate and a depth of field nar-
row enough to create blur everywhere but the immediate
plane of focus.
Narrow depth of field is created on a real camera by lower-
ing the f-stop value, which lowers exposure as well. Not so
with the After Effects 3D camera. Its Aperture and F-Stop
settings (Figure 9.20) affect only focal depth, not exposure
or motion blur. The two settings have an inverse relation-
ship. F-Stop is the setting more commonly referenced
by camera operators, and yet only Aperture appears as a
property in the Timeline.

Figure 9.20 Check Enable Depth of
Field in Camera Settings to activate
Focus Distance (the distance in pixels
of the focal point, which can be
toggled to Lock to the Zoom). A low
F-Stop (or high Aperture) with a Blur
Level of 100% creates a shallow focal
effect.

After Effects depth of field settings can be matched to a
camera report, provided that it includes the f-stop setting
used when the footage was shot. If so, open up the Camera
Settings dialog (Ctrl+Shift+Y/Cmd+Shift+Y, or double-
click on the Camera in the Timeline panel), check the box
labeled Enable Depth of Field, and enter the value.

291


The key here is to offset at least one layer in Z space so
that it falls out of focal range. Now, in the Top view, set the
Focus Distance (under Options) to match the layer that
will be in focus at the beginning of the shot, add a key-
frame, then change the Focus Distance at another frame to
match a second layer later in the shot (Figure 9.21).
A static focus pull doesn’t look quite right; changing
Figure 9.21 With Enable Depth focus on a real camera will change the framing of the shot
of Field on, the Focus Distance is
slightly. To sell the example shot, which starts on a view of
denoted by a red boundary line,
easily viewed and animated in the city and racks focus to reveal a sign in the foreground,
isometric views. I animate the camera pulling back slightly, augmented by
a nice shift that then occurs in the offset planes of focus
(Figure 9.22).

Figure 9.22 The final shot combines a rack focus with a gentle pull-back, using ease keyframes to animate Position and
Focus Distance.

Boke Blur
Racking focus in this manner generates camera blur that is
accurate relative to the plane of focus, but it does not truly
create the look of a defocused lens, because the lens itself
is a major participant in that look.
Boke connotes the phenomenon whereby points of light
become discs of light (also called circles of confusion) that
take on the character of the lens itself as they pass through
the camera lens and aperture. Like lens ﬂares (covered in
Chapter 12) these are purely a phenomenon of the camera
lens, not human vision; they can add beauty and suspense
to a shot.
Illuminated out of focus elements in a shot are, after all,
mysterious. Visual intrigue is created as the shot resolves in
or out of a wash of color and light (Figure 9.23).

292


Figure 9.23 With shallow depth of field, highlights in the foreground retain blur even in a focused shot.

A perfect lens passes a defocused point of light to the
back of the camera as a soft, spherical blur. A bright point
remains bright, but is larger and softer. Ordinary blur of a
low-dynamic-range image in 8 or 16 bit per channel color
mode instead merely dims the highlights (Figure 9.24).

Figure 9.24 Begin with a source image that includes bright highlights (top left); blur it via conventional means and the
result is gray and desaturated (top right), unless the source image is HDR and the comp is 32 bpc (bottom left), which
approaches the look of real camera motion blur (bottom right).

293


Most camera lenses are not perfect, so instead of perfect
blurred spheres, boke spheres may be brighter toward the
edges than in the middle. An anamorphic lens will show
squashed spheres, and as with lens flares, the shape of the
aperture itself may be visible in the circles, making them
hexagonal (or pentagonal, and so on, depending on the
number of blades in the opening).

Go for Boke
To accurately create the bloom of highlights as they are
blurred requires 32 bit per channel color and source high-
lights that are brighter than what would be called full white
in 8 or 16 bpc. The process of creating such an image is
explored and explained in Chapter 11.
The Lens Blur effect does not operate in 32 bpc—it instead
mimics the behavior of bright highlights through a lens.
It’s more or less a direct port from Photoshop; as such, it
can be slow and cumbersome in After Effects. It won’t blur
beyond 100 pixels, and the effect does not understand
non-square pixels (it creates a perfect circle every time).
Instead of 3D camera or layer data, Lens Blur can use a
Depth Map Layer, using pixel values (brightness) from
a specified Depth Map Channel. You can rack focus by
adjusting Blur Focal Distance. Iris Shape defines polygons
around the highlights, corresponding to the number of
blades in the iris; these can also have a specified Iris Blade
Curvature and Iris Rotation (this rotates the polygon).
The actual amount of blur is determined by Iris Radius,
the bloom by Specular Threshold (all pixels above this
The most respected third-party value are highlights) and Specular Brightness, which
tool for lens blurs is Frischluft’s
Lenscare. The default settings are creates the simulation of highlight bloom. These are the
not reliable, but with adjust- controls you’ll tweak most (Figure 9.25).
ments and depth maps (for 3D
footage), you can derive some The Noise controls are designed to restore noise that
lovely results (you’ll find Lenscare would be removed by the blur operation; they don’t relate
at www.frischluft.com and on the to the blur itself and can be ignored in favor of grain tech-
book’s DVD). niques described in the following section.

294


Figure 9.25 The result of a Lens Blur effect (left) doesn’t look so hot compared it to the real thing (center), while the Lens-
care plug-in from Frischluft (right) is remarkably close. Sometimes the tools matter.

By no means do the settings in Lens Blur (or for that
matter, third-party alternatives such as Lenscare from
Frischluft) exhaust the possibilities for how defocused
areas of an image might appear, especially when illumi-
nated. Keep looking at the reference and thinking of ways
to re-create what you see in it (Figure 9.26).

Figure 9.26 Does an image with
shallow depth of field look more cin-
ematic? What do you see happening
in the defocused background?

The Role of Grain
Once the image passes through the lens and is recorded, it
takes on another characteristic: grain. Grain is essentially
high-frequency noise readily apparent in each channel of
most recorded footage, although progress in image gather-
ing technology has led to a gradual reduction of grain.
Grain can, however, be your friend, adding life to static
imagery and camouﬂaging edge detail.

295


Grain management is an essential part of creating high-
quality moving images; properly done, it is not simply
switched on or off, but requires careful per-channel adjust-
ment. There are two basic factors to consider:
. Size of the grain, per channel
. Amount of grain, or amount of contrast in the grain,
per channel
The emphasis here is that these factors typically vary from
channel to channel. Blue is almost universally the channel
likeliest to have the most noise; happily, the human eye is
Excessive grain is often triggered
by a low amount of scene light less sensitive to blue than red or green.
combined with a low-quality How much grain is enough? As with color in Chapter 5, the
image-gathering medium, such as
miniDV, whose CCD has poor light- goal is typically to match what’s there already. If your shot
gathering abilities. has a background plate with the proper amount of grain
in it, match foreground elements to that. A computer-
generated still or scene might have to be grain-matched to
surrounding shots.

Grain Management Strategies
After Effects Professional includes a suite of three tools for
automated grain sampling, grain reduction, and grain gen-
eration: Add Grain, Match Grain, and Remove Grain. Add
Grain relies on your settings only, but Match Grain and
Remove Grain can generate initial settings by sampling a
source layer for grain patterns.
I often caution against the automated solution, but not in
this case. Match Grain is not even appreciably slower with
grain sampling than Add Grain, which does not sample but
includes all of the same controls. Match Grain usually comes
up with a good ﬁrst pass at settings; it gets you 70–80% there
and is just as adjustable thereafter. In either case
1. Look for a section of your source footage with a solid
color area that stays in place for 10 to 20 frames. Most
clips satisfy these criteria (and those that don’t tend to
allow less precision).
2. Zoom to 200% to 400% on the solid color area, and
create a Region of Interest around it. Set the Work Area
to the 10 or 20 frames with little or no motion.

296


Figure 9.27 Even without slamming the image it’s clear that the added window looks too flat in this grainy scene, but
doing so with the addition of a gradient gives you a clear target.

3. Add a solid small enough to occupy part of the Region
of Interest. Apply a Ramp effect to the solid, and use
the eyedropper tools to select the darkest and light-
est pixels in the solid color area of the clip. The lack
of grain detail in the foreground gradient should be
clearly apparent (Figure 9.27).
4. Apply the Match Grain effect to the foreground solid.
Choose the source footage layer in the Noise Source
Layer menu. As soon as the effect ﬁnishes rendering
a sample frame, you have a basis from which to begin
ﬁne-tuning. You can RAM Preview at this point to see
how close a match you have. In most cases, you’re not
done yet.

297


5. Twirl down the Tweaking controls for Match Grain, and
then twirl down Channel Intensities and Channel Size.
You can save yourself a lot of time by doing most of
your work here, channel by channel.
6. Activate the green channel only in the Composition
panel (Alt+1/Option+1) and adjust the Green Intensity
and Green Size values to match the foreground and
background. Repeat this process for the green and blue
channels (Alt+2/Option+2 and Alt+3/Option+3). If you
don’t see much variation channel-to-channel, you can
instead adjust overall Intensity and Size (Figure 9.28).
RAM Preview the result.
7. Adjust Intensity, Size, or Softness controls under Tweak-
ing according to what you see in the RAM Preview. You
may also ﬁnd it necessary to reduce Saturation under
Color, particularly if your source is ﬁlm rather than
video.

Figure 9.28 In this unusual case there is little variation of grain channel to channel, and the automatic match is pretty good;
a slight boost to the overall Intensity setting under the “Tweaking” controls does the trick.

298


In most cases, these steps yield a workable result (Figure
9.29). The effect can then be copied and pasted to any
foreground layers that need grain. If the foreground layer
already contains noise or grain, you may need to adjust the
Compensate for Existing Noise percentage for that layer.

Figure 9.29 Even in this printed figure
the matching grain is somewhat
evident. Grain matching is often best
reviewed in motion with a boosted
exposure.

Use Noise as Grain
Prior to the addition of Add Grain and Match Grain
Obviously, whole categories of controls within Match to version 6.5 Professional, the typical way to
Grain remain untouched with this approach; the Applica- generate grain was to use the Noise effect. The
tion category, for example, contains controls for how the main advantage of the Noise effect over Match
grain is blended and how it affects shadows, midtones, and Grain is that it renders about 20 times faster. How-
ever, After Effects doesn’t make it easy for you to
highlights individually. These are typically overkill, as are separate the effect channel by channel, and scaling
the Sampling and Animation controls, but how far you go requires a separate effect (or precomping).
in matching grain before your eye is satisﬁed is, of course,
up to you.
You can employ three solid layers, with three
effects applied to each layer: Shift Channels,
Grain Removal Noise, and Transform. You use Shift Channels
Removing grain, or sharpening an image in general, is an to set each solid to red, green, or blue, respectively,
set Blending Modes to Add, and set their Opacity
entirely different process from adding grain. On a well- very low (well below 10%, adjusting as needed).
shot production, you’ll rarely have a reason to reach for Next, set the amount of noise and scale it via the
the Remove Grain tool. Transform effect.

If you do, the reason for doing so may be unique to your
particular footage. In such cases, you may very well ﬁnd If the grain is meant to affect a set of foreground
layers only, hold them out from the background
that leaving Remove Grain at the default settings gives you
plate either via precomping or track mattes. If this
a satisfactory result. If not, check into the Fine Tuning and sounds complicated, it is, which is why Match
Unsharp Mask settings to adjust it. Grain is preferable unless the rendering time is
really killer.

299


Remove Grain is often best employed stealthily—not neces-
sarily across the entire frame (Figure 9.30), or as part of
a series of effects. It is a reasonably sophisticated solution
(compared with the current alternatives) that can really
help in seemingly hopeless situations; this book’s technical
editor reports having used it extensively on a feature film
in which the aging lead actor needed a lot of “aesthetic”
facial work done (removing wrinkles and so on).

Figure 9.30 The left side of frame is
clearly less grainy than the right as a
result of applying Remove Grain and
letting it automatically sample the
footage.

When to Manage Grain
The most obvious candidates for grain addition are
If you’re using Remove Grain to computer-generated or still image layers that lack the mov-
improve a blue-screen or green- ing grain found in film or video footage. As soon as your
screen key, consider applying the shot has to match anything that came from a camera, and
result as an alpha track matte. This particularly in a large format such as HD or film, you must
offers the best of both worlds: a
clean matte channel and preserva-
manage grain.
tion of realistic grain on the source Blurred elements may also need grain addition, even
color layer.
if they originate as source footage. Blurry source shots
contain as much grain as focused ones because the grain
is an artifact of the medium recording the image, not the
subject itself. Elements that have been scaled down in After
Effects contain scaled-down grain, which may require resto-
ration. Color keying can also suppress grain in the channel
that has been keyed out.
Other compositing operations will instead enhance grain.
Sharpening, unless performed via Remove Grain, can

300


strongly emphasize grain contrast in an element, typically
in a not-so-desirable manner. Sharpening also brings out
any nasty compression artifacts that come with footage that
uses JPEG-type compression, such as miniDV video.
Lack of grain, however, is one of the big dead giveaways of
a poorly composited shot. It is worth the effort to match
the correct amount of grain into your shot even if the
result isn’t apparent as you preview it on your monitor.

Film and Video Looks
If you flipped to this section intentionally, you may be try-
ing to do one of two things with a given shot or project:
. Achieve the look of a different camera
. Maximize the production value of your video to make it
“filmic”
There are so many issues connected to the second one
above and beyond what you can achieve in an After Effects
comp that Stu Maschwitz went and wrote a whole book
about it. The DV Rebel’s Guide: An All-Digital Approach to Mak-
ing Killer Action Movies on the Cheap (Peachpit Press, 2006) is
an excellent resource, not only for After Effects knowledge,
but for the whole process of low-budget digital filmmaking.
The first chapter lists the major factors that influence pro-
duction value. Many of these, including image and sound
quality, location and lighting, cannot entirely be created
in After Effects, which must be why Stu’s book includes a
bunch of information on how to actually shoot. Stu is also
a big proponent of shallow depth of field as a necessary
component of visual storytelling.
Achieving the look of a different camera is well within the
realm of tricks you can pull off consistently in After Effects,
and some of the following play into both developing a look
and maximizing production value:
. Lens artifacts: In addition to those already discussed
in this chapter, such as boke and chromatic aberration,
are such filmic visual staples as the vignette and the
lens flare.

301


. Frame rate: Change this and you can profoundly alter
the viewer’s perception of footage.
. Aspect ratio: The format of the composition makes a
Garbage In, Garbage Out huge perceptual difference as well, although it’s not so
You don’t need me to tell you how difficult it is to
simple as “wider = better.”
bring a poorly shot image back from the dead, but . Color palette: Nothing affects the mood of a given shot
check The DV Rebel’s Guide for a thorough rundown like color and contrast. It’s a complex subject further
of factors that go into a well-shot image, and if
possible go on set to offer supervision and help
explored in Chapter 12.
eliminate flaws that will be difficult to fix in post.
Among the less obvious points from the book Lens Artifacts Aren’t Just Accidents
. When shooting digitally, keep the contrast low Because this chapter is all about reality as glimpsed by the
and overall light levels well below maximum; camera lens, several types of lens artifacts, visual phenom-
you are shooting the negative, not the final ena that occur only through a lens, have already appeared
(Figure 9.31).
in this chapter, including lens distortion and lens blur
. If using a small, light camera, mount it to
something heavy to move it; that weight reads
(or boke).
to the viewer as more expensive and more You won’t be surprised to hear that this isn’t all: potentially
natural motion.
in your palette are more phenomena of the type that pro-
fessional cinematographers tended to avoid until the 1970s
(when they started to be considered cool). These include
lens flares, vignettes, and chromatic aberration. None of
these occur with the naked eye, but remember, your target
is the look of the real world as seen through the camera.

Lens Flares
Optical flares are caused by secondary reflections bouncing
around between the camera elements. Because they occur
within the lens, they appear superimposed over the image,
even when partially occluded by objects in the foreground.
Unlike your eye, which has only one very flexible lens,
camera lenses are made up of a series of inflexible lens
elements; the longer the lens, the more elements within it.
Each element is coated to prevent reflection under normal
circumstances, but with enough light flooding directly in,
reflection occurs.
Artists sometimes like to get goofy and creative with lens
Figure 9.31 Shooting low-contrast (top) with a
flares; how many of us, after all, are experts in how they
camera that has a healthy contrast range allows
you to bring out hidden detail and color, even should look? And yet this is one more area where seem-
tone-mapping to do so only in specific areas of ingly unsophisticated viewers can smell something fake
frame (bottom). under their noses, so certain rules apply.

302


Zoom lenses contain many focusing elements and tend to
generate a complex-looking flare with lots of individual
reflections. Prime lenses generate fewer reflections and a
simpler flare.
Just as with boke, aperture blades within the lens can con-
tribute to the appearance of flares. Their highly reflective
corners often result in streaks, the number corresponding
to the number of blades. The shape of the flares some-
times corresponds to the shape of the aperture (a penta-
gon for a five-sided aperture, a hexagon for six). Dust and
scratches on the lens also reflect light.
You can create a lens flare look by hand using solids and
blending modes, but most people don’t have the time for
this. The Lens Flare effect that ships with After Effects is a
rather paltry offering and includes little in the way of cus-
tomization; you’re best off with Knoll Light Factory, which
is highly customizable and is derived from careful study of
lens behaviors, although if you already own Tinderbox 2
from The Foundry, the T_LensFlare is still a vast improve-
ment over the After Effects default.
More about the behavior and application of lens flares
appears in Chapter 12.

Vignettes
Vignetting is a reduction in image brightness around the
edges of an image. It’s generally an undesired by-product
of certain lenses (particularly wide-angle fisheyes), but it
is sometimes deliberately chosen because of how it helps
focus attention on the center of frame. I can say with
authority that several underwater shots from Pirates of the
Caribbean: At World’s End contain vignettes, because I added
them myself.
It’s an easy effect to create:
1. Create a black solid the size of your frame as the top
layer and name it Vignette.
2. Double-click the Ellipse tool in the toolbar; an elliptical
mask fills the frame.

303


3. Highlight the layer in the Timeline and press F to
reveal Mask Feather.
4. Increase the Mask Feather value a lot—somewhere in
the triple digits is probably about right.
5. Lower the Opacity value (T) until the effect looks right;
you might prefer a light vignette (10 to 15%) or some-
thing heavier (40 to 50%).
Note that the vignette is elliptical, not perfectly round, and
if your project is to be seen in more than one format (see
later in the chapter) you’ll have to decide which is the tar-
get (Figure 9.32). There would be no reason for a realistic
vignette to appear offset.

Chromatic Aberration
Even further down the road of questionably aesthetic visual
phenomena is chromatic aberration, a fringing or smear-
ing of light that occurs when a lens cannot focus various
Figure 9.32 A vignette is created with a feath- colors on the spectrum to a single point, because of the
ered mask applied to a solid (top). If the image is differing wavelengths. The effect is similar to that of light
reframed for display in another format, such as passing through a prism and dispersing into a rainbow of
anamorphic, you may have to use that framing
instead of the source (bottom).
colors.
Like vignettes, and optically related to lens ﬂares and boke,
chromatic aberration is something higher-end lenses are
designed to avoid, yet it can occur even under relatively
expensive and high-end shooting circumstances, particu-
larly if there is any type of lens conversion happening.
Unlike the others, it can really look like a mistake, so it’s
not the kind of thing you would probably add to a clip
in order to make it look cool; instead you might add it to
a shot or element to match another shot or background
plate in which it appears. My recommendation in such
a case?
1. Duplicate the layer twice and precompose all three.

2. Use the Shift Channels effect to leave only red, green
or blue on for each layer (so you end up with one of
each).
3. Set the top two layers to Add mode.

304


Figure 9.33 A simulation of chromatic aberration (right), the color ringing which is caused when different wavelengths of
light have different focal lengths; most lenses correct for it with an added diffractive element.

4. Scale the green channel to roughly 101% and the blue
channel to roughly 102%.
5. Add a small amount of Radial Blur (set to Zoom, not
the default Spin).
A before and after comparison appears in Figure 9.33.
29.97 fps Videotape Never Took Over
Frame Rate Isn’t Just Speed
The debate between using 24 fps film and 29.97
One could probably write a whole book or thesis on this fps videotape in the U.S. and other countries with
one topic alone, but it’s no accident that film images are NTSC has been raging since long before the digital
displayed at 24 frames per second and that newer digital era. It began with the advent of videotape in the
1950s, when tape was cheap and fast, if cumber-
formats, which could theoretically be optimized for just some by today’s standards.
about any frame rate, also aim for this rate (despite how
difficult it is to find a low-end camera that shoots 24p
natively, with no interlacing). One particular experiment from this era stands out.
For six episodes, the producers of The Twilight Zone
The question that would generate all of the chatter is tried tape before they evidently realized it was
ruining the show’s mystique.
“why?” There is no logical answer, and many attempts have
been made to explore alternatives. The simple truth seems
to be that frame rates of 30 fps and higher feel more like Video’s higher frame rate and harder look instantly
direct reality, but 24 fps is just above the threshold where turned one of the most intriguing and ironic series
of all time into something that looked more like a
persistence of vision breaks down, giving it a more ephem- soap opera. To judge for yourself, rent DVDs from
eral and dream-like quality, just as do other cinematic con- Season 2 that include the following videotaped
ventions such as light bloom and shallow depth of field. episodes: “Static,”“Night of the Meek,”“The Late-
ness of the Hour,”“The Whole Truth,”“Twenty-Two,”
If you have a choice on a given project and you want it to or “Long Distance Call.”
have a cinematic look, try creating it at 24 fps and judge
for yourself. After Effects is quite forgiving about letting
Even though videotape was always much cheaper
you change frame rates mid-stream compared with most and more convenient than film, there seem to be
video applications; details on how the conversion actually dramatic reasons for why it never took over.
works appeared back in Chapter 2.

305


If you have no choice but to work at 29.97 fps, you still
have a choice: progressive versus interlaced. It’s not neces-
sarily an error to render animation without adding inter-
The numbers “1.85” and “2.35” give lacing; in fact, step through your favorite animated series
the width, relative to a height of 1, on television and you may find that it’s animated at 15 fps
so it’s like saying 1.85:1 or 2.35:1. or less (and basically never at 59.94 fps, which is effectively
The 16:9 format, which has become what 29.97 fps interlaced means in animation terms). South
popular with digital video and
HD, is equivalent to a 1.77:1 ratio, Park doesn’t count.
slightly narrower than Academy,
but wide compared to the standard Format Isn’t Just Display Size
television format of 4:3 (1.33: 1).
As the world transitions from standard-definition to high-
definition broadcast television, formats are undergoing the
same transition that they made in film half a century ago.
The nearly square 4:3 aspect is being replaced as standard
by the wider 16:9 format, but 1.85 Academy aperture and
2.35 Cinemascope also appear as common “widescreen”
formats.
A lot of artists (students, particularly) fall in love with the
widescreen look for how it conjures Star Wars and Lawrence
of Arabia, but if these formats aren’t shown at 24 fps and
don’t obey other cinematic conventions outlined here,
the result tends to appear a bit cheesy. So remember, it’s a
convention we associate with film, whether or not we know
the following history.
In response to the growing popularity of television in
the 1950s, Hollywood conjured up a number of different
widescreen formats through experiments with anamorphic
lenses and film stocks as wide as 70 mm. These systems—
CinemaScope, VistaVision, Panavision, and so on—haven’t
completely faded away, but their presence in the modern
era is mostly felt in the way that films are displayed, not
how they are shot. 35 mm is once again the most popu-
lar shooting format, specifically the full-aperture version
known as Super 35 mm.
Standard 35 mm film has an aspect ratio of 4:3, which
is not coincidentally the same as a television. Almost all
current movies are filmed in this format as if originally
intended for the small screen. When shown in a theater
using a widescreen aspect of 1.85:1 (also known as 16:9,

306


the HDTV standard) or 2.35:1 (CinemaScope/Panavision),
the full 4:3 negative is cropped (Figure 9.34). Theater
patrons actually pay $10 to see less than if they waited for
the movie to get broadcast full screen on cable! 2.35:1
Less Color Can Be More
The influence of color decisions on the final shot, and
by extension on the story being told in the shot, is an
immense topic, hashed over by cinematographers and col- 1.85:1
orists the world over. Any attempt to distill this into a few
pithy paragraphs would be a disservice.
One thing you may notice as you study your favorite films 0.980”
2048
and videos is that they are limited to a distinct palette, and

0.735”
1536
scenes that you would think of as “full color” may in fact be 4:3
dominated by one, two or at most three basic hues.
Full Aperture
Thus, if you’re new to the idea of developing a color look
for a film or sequence, look at references. Study other peo-
ple’s work for the effect of color on the mood and story in Figure 9.34 The “wider” film formats
a shot, sequence, or entire film. There are also great color might more accurately be called
“shorter” because they typically involve
brainstorming tools such as the one shown in Figure 9.35, cropping the original 4:3 image.
designed specifically to give a particular look or mood to
your shot.

Figure 9.35 Looks 3, from Red Giant
Software, is both an After Effects
plug-in and a stand-alone tool called
Looks Builder. Not only does it have
an innovative color pipeline based on
real-world optics, it lets you window-
shop through dozens of strong, preset
looks (and create more of your own)
so it’s great for brainstorming.

307


Conclusion
And really, you’ve just scratched the surface of what’s pos-
sible. The inventive compositor can and should always look
for new methods to replicate the way that the camera sees
the world, going beyond realism to present what we really
want to see—realism as it looks through the lens.

308

CHAPTER

10
Expressions
Dan Ebberts

Music is math.
—Michael Sandison and Marcus Eoin
(Boards of Canada)

Expressions

E xpressions are very cool. You can use them to cre-
ate amazing procedural effects that would otherwise be
impossible (or at least impractical). You can also use them
to create complex relationships between various param-
eters. Unfortunately, many After Effects users are afraid of
expressions. Don’t be.
The fact that you’re reading this chapter indicates that you
are at least curious about expressions. That’s a good start.
By the end of the chapter, you’ll see how expressions can
open new doors for you, and, hopefully, you’ll have the
conﬁdence to give them a try.
The best way to learn about expressions is to examine
working examples to ﬁgure out what makes them tick. The
examples in this chapter focus on how you can use expres-
sions to create or control effects.
As you work through the examples (don’t be discouraged if
you need a couple passes to understand it all), please keep
in mind that I’m mainly a code guy—not a special effects or
motion graphics artist. My examples may not be very visually
impressive, but using these same techniques, I’m convinced
that you’ll be able to create your own dazzling effects.

What Expressions Are
The After Effects expression language is a powerful set of
tools with which you can control the behavior of a layer’s
properties. Expressions can range in complexity from
ridiculously simple to mind-numbingly complicated. At the
simple end of the spectrum, you can use expressions to
link one property to another or to set a property to a static

310


value. At the other extreme, you can create complex link-
ages, manipulate time, perform calculations in 3D space,
set up tricky procedural animations, and more.
Sometimes you’ll use expressions instead of keyframes
(most properties that can be keyframed can be controlled
by expressions). In other cases you’ll use expressions to aug-
ment the keyframed behavior. For example, you could use
keyframes to move a layer along a speciﬁc path and then
add an expression to add some randomness to the motion.

Expressions Have Limitations
Although the After Effects expression language presents you with an impressive
arsenal of powerful tools, it’s important to understand the limitations of expressions
so that you can avoid making assumptions that lead you astray.
. An expression may generally be applied only to a property that can be key-
framed, and it can affect only the value of that property. That is, a expression can
affect one, and only one thing: the value of the property to which it is applied.
This means there are no global variables. This also means that although an
expression has access to many composition and layer attributes (layer width and
height, for example) as well as the values of other properties, it can only read,
not change, them.
. Expressions can’t create objects. For example, an expression cannot spawn a new
layer, add an effect, create a paint stroke, change a blend mode—the list goes on
and on. Remember, if you can’t keyframe it, you can’t create an expression for it.
. Expressions can’t access information about individual mask vertexes.
. Expressions can’t access text layer formatting attributes, such as font face, font
size, leading, or even the height and width of the text itself.
. Expressions cannot access values they created on previous frames, which
means expressions have no memory. If you’ve had a little Flash programming
experience, you might expect to be able to increment a value at each frame.
Nope. Even though you can access previous values of the property using
valueAtTime(), what you get is the pre-expression value (the static value
of the property plus the effect of any keyframes). It’s as if the expression didn’t
exist. There is no way for an expression to communicate with itself from one
frame to the next. Note, however, just to make things more confusing, the post-
expression value of a property is available to any other expression, just not the
one applied to that property. In fact, the post expression value is the only value
available to expressions applied to other properties. To summarize: an expres-
sion has access only to the pre-expression value of the property to which it is
applied, and it only has access to the post-expression values for other properties
with expressions. It’s confusing at first, but it sinks in eventually.

311

Chapter 10 Expressions

Creating Expressions
The easiest way to create an expression is to simply
Alt/Option-click the stopwatch of the property where you
want the expression to go. After Effects then creates a
default expression, adds four new tool icons, changes the
color of the property value to red (indicating that the value
is determined by an expression), and leaves the expression
text highlighted for editing (Figure 10.1).

Figure 10.1 When you create an expression, After Effects creates a default
expression with the text highlighted for editing, changes the color of the
property value to red, and adds four new tool icons: an enable/disable toggle, a
Graph Editor toggle, a pickwhip, and an Expression Language menu fly-out.

At this point you have a number of options. You can simply
start typing, and your text will replace the default expres-
sion. Note that while you’re in edit mode, the Enter/
Return key moves you to a new line in the expression (this
is how you can create multiline expressions) and leaves you
in edit mode.
Another option while the text is highlighted is to paste in
the text of an expression that you have copied from a text
editor. This is the method I generally use if I’m working on
a multiline expression.
Instead of replacing all the default text by typing or past-
ing, you can click somewhere in the highlighted text to
create an edit point for inserting additional text.
Alternatively, you can drag the expression’s pickwhip
to another property or object (the target can even be in
another composition), and After Effects will insert the
appropriate text when you let go. Note that if an object or
property can be referenced using the pickwhip, a rounded
rectangle appears around the name as you drag the pick-
whip over it. If this doesn’t happen, you won’t be able to
pickwhip it.

312


Finally, you can also use the Expression Language menu to
insert various language elements.
After creating your expression, exit edit mode by clicking
somewhere else in the Timeline or pressing Enter on the
numeric keypad. If your expression text contains an error,
After Effects displays an error message, disables the expres-
sion, and displays a little yellow warning icon (Figure 10.2).
You can temporarily disable an expression by clicking on
the enable/disable toggle.

Figure 10.2 If your expression con-
tains an error, After Effects disables
the expression, changes the enable/
disable toggle to the disabled state,
returns the Property value to its nor-
mal color, displays an error icon, and
displays an error message dialog box.

Working with existing expressions is as easy as creating
them. Some common operations are
. Editing. Click in the expression text area to select the
entire expression; you now have the same options as
when creating a new expression. If your expression
consists of multiple lines, you may need to expand
the expression editing area to be able to see all (or at
least more) of it by positioning the cursor over the line
below the expression text until you see a double-ended
arrow and then clicking and dragging.
. Deleting. Simply Alt/Option-click the property’s stop-
watch, or you can delete all the text for the expression
and press Enter on the numeric keypad.
. Exposing. Select a layer in the Timeline and press EE to
expose any expressions applied to that layer.
. Copying. In the Timeline, select a layer property con-
taining an expression and choose Edit > Copy Expres-
sion Only to copy just the property’s expression. You
now can select as many other layers as you’d like and
Edit > Paste to paste the expression into the appropri-
ate property of the other layers.

313


The Language of Expressions
The After Effects expression language is based on a subset
of JavaScript. JavaScript is a scripting language used largely
for web page design and includes many features specifi-
cally aimed at that task. The JavaScript implementation for
expressions includes the core features only. That means
there’s a lot about JavaScript that you won’t need to know,
but it also means that any JavaScript reference you pick up
(and you’re going to need one if you really want to master
expressions) is going to have a lot of content that will be of
little or no use to you.
The rest of the expression language consists of extensions
that Adobe has added specifically for After Effects. This
means that in addition to a good JavaScript reference,
you’ll also be frequenting Adobe’s After Effects Expression Ele-
ment Reference. The most up-to-date version of this reference
can be found at Adobe’s Help on the Web. For machines
with an Internet connection, the After Effects Help menu
will take you there: Help > After Effects Help. Alternatively,
you can go to www.adobe.com/go/learn_ae_cs4helphome.
This chapter focuses on working examples rather than the
details of JavaScript. The book’s disc, however, contains an
abbreviated JavaScript guide, and I recommend that you
glance through it before you really dive into the sample
expressions discussed here. In addition, I’ll point you to
the appropriate sections of that guide as you encounter
new JavaScript elements for the first time.

Linking an Effect Parameter to a Property
Here’s the scenario: You want to link an effect to an audio
track. Specifically, you want to link the Field Of View
(FOV) parameter of the Optics Compensation effect to
the amplitude of an audio layer. Expressions can’t access
audio levels directly, so first you have to use a keyframe
assistant (Animation > Keyframe Assistant > Convert Audio
to Keyframes) to create a null layer named Audio Ampli-
tude with Slider Controls keyframed for the audio levels of
the Left, Right, and Both channels (for a stereo source).

314


Next, you just Alt/Option-click the stopwatch for the FOV
parameter of the Optics Compensation effect and drag
the pickwhip to the Both Channels Slider property of the
Audio Amplitude layer (Figure 10.3). Doing so generates
this expression
thisComp.layer("Audio Amplitude").effect("Both
➥Channels")("Slider")

Figure 10.3 Select the Both Channels
slider with the pickwhip to replace the
highlighted default expression text.

Take a closer look at its syntax: From JavaScript, the
After Effects expression language inherits a left-to-right
“dot” notation used to separate objects and attributes in
a hierarchy. If your expression references a property in a
different layer, you ﬁrst have to identify the composition.
You can use thisComp if the other layer happens to be in
the same composition (as in this example). Otherwise,
you would use comp("other comp name"), with the other
composition name in quotes. Next you identify the layer
using layer("layer name") and ﬁnally, the property, such
as effect("effect name")("property name") or possibly
transform.rotation.

In addition to objects and properties, the dot notation
hierarchy can include references to an object’s attributes
and methods. An attribute is just what you would guess: a
property of an object, such as a layer’s height or a composi-
tion’s duration. In fact, in JavaScript documentation, attri-
butes are actually referred to as properties, but in order to
avoid confusion with the layer properties such as Position
and Rotation (which existed long before expressions came
along), in After Effects documentation (and here) they’re
referred to as attributes. For example, each layer has a
height attribute that can be referenced this way:
comp("Comp 1").layer("Layer 1").height

315


Methods are a little harder to grasp. Just think of them
as actions or functions associated with an object. You can
tell the difference between attributes and methods by the
parentheses that follow a method. The parentheses may
enclose some comma-separated parameters.
It’s important to note that you don’t have to specify the full
path in the dot notation hierarchy if you’re referencing
attributes or properties of the layer where the expression
resides. If you leave out the comp and layer references,
After Effects assumes you mean the layer with the expres-
sion. So, for example, if you specify only width, After
Effects assumes you mean the width of the layer, not the
width of the composition.
Let’s forge ahead. You linked the amplitude of your audio
layer to your effect parameter, but suppose you want to
increase the effect that the audio level has on the param-
eter. You can use a little JavaScript math to multiply the
value by some amount, like this
If you’re not familiar with JavaScript
arithmetic operators (such as the thisComp.layer("Audio Amplitude").effect("Both
* for multiplication used in this ➥Channels")("Slider") * 3
example), you might want to take
a look at the “Operators” section of Towards the end of the chapter you’ll see a much more
the JavaScript guide. complicated and powerful way of linking an effect to audio.

Using a Layer’s Index
A layer’s index attribute can be used as a simple, but power-
ful tool that allows you to create expressions that behave
differently depending on where the layer is situated in the
layer stack. The index attribute corresponds exactly to the
number assigned to the layer in the Timeline window. So,
the index for the layer at the top of the stack is 1, and so on.

Time Delay Based on Layer Index
Suppose you keyframed an animation for one layer. Now
you want to create a bunch of identical layers, but you want
their animations to be delayed by an amount that increases
as you move down the layer stack. You also want to rotate
each copy by an amount proportional to its position in the

316


layer stack. To do so, you first apply an expression like this
to the top layer’s animated properties
delay = 0.15;
valueAtTime(time - (index-1)*delay)

Then you apply an expression like this to the Rotation
property:
offsetAngle = 3;
value +(index-1)*offsetAngle

Finally, duplicate the layer a bunch of times. The anima-
tion of each layer will lag behind the layer above it by 0.15
seconds and the rotation of each layer will be 3 degrees
more than the layer above (Figure 10.4).

Figure 10.4 Notice how the blaster
shot created by each layer lags that of
the previous layer and is at a slightly
different angle.

What’s going on here? In the first expression, the first line
defines a JavaScript variable named delay and sets its value
to 0.15 seconds. The second line is where all the action is
If you’re not familiar with JavaScript
and it’s packed with new things. For example, notice the variables, see the “Variables” sec-
use of time. It represents the current composition time, in tion of the JavaScript guide on the
seconds. In other words, time represents the time at which accompanying disc.
the expression is currently being evaluated.

317


You use valueAtTime() to access a property’s pre-expression
value at some time other than the current comp time (to
access the pre-expression value at the current comp time,
use value() instead, as in the Rotation expression). The
Remember, if you don’t specify a
comp and layer when referencing a
parameter passed to valueAtTime() determines that time:
property or attribute, After Effects time – (index-1)*delay
assumes you mean the layer with
the expression. When you refer- Subtracting 1 from the layer’s index and multiplying that
ence an attribute of the property result by the value of the delay variable (0.15) gives the
housing the expression, After
Effects makes a similar assumption, total delay (in seconds) for this layer. Subtracting 1 from
allowing you to specify only the index means that the delay will be zero for the first layer.
attribute name (without the entire So, for Layer 1, the total delay is zero, for Layer 2 it is 0.15,
comp/layer/property path). One for Layer 3 it is 0.30, and so on. You then subtract the total
side benefit of not having to specify
the entire path is that you can delay from the current comp time. The result of this is that
apply the same expression to any Layer 1’s animation runs as normal (not delayed). Layer 2’s
property, without having to modify animation lags behind Layer 1 by 0.15 seconds, and so on.
it at all.
The Rotation expression is very similar except that it
doesn’t reference time. The reason for this is that the first
expression is used to offset a keyframed animation in time,
while the second expression simply creates a static (not
animated) offset for the Rotation property. The first line of
the expression defines a variable named offsetAngle. This
variable defines the rotation amount (in degrees) by which
each layer will be offset from the layer above it. The second
line tells After Effects to calculate the layer’s offset and add
it to the pre-expression value of the property.
You’ll see other ways to use index in later examples.

Looping Keyframes
The expression language provides two convenient ways to
loop a sequence of keyframes: loopOut() and loopIn().
Suppose you keyframed a short animation and you want
that sequence to repeat continuously. Simply add this
expression to the keyframed property
loopOut("cycle")

and your animation will loop for the duration of the comp
(Figure 10.5).

318


Figure 10.5 The solid line in the graph represents the keyframed bounce action. The dotted line represents the subsequent
bounces created by loopOut("cycle").

There are three other variations of loopOut(), as well:
. loopOut("pingpong") Runs your animation alternately
forward then backward.
. loopOut("continue") Extrapolates the animation A small glitch in the cycle version
beyond the last keyframe, so the value of the prop- of loopOut() drops the first
keyframe from each of the loops. If
erty keeps moving at the same rate (and in the same you want the frame with the first
direction, if you’re animating a spatial property such as keyframe to be included, add a
Position) as the last keyframe. This can be useful, for duplicate of the first keyframe one
example, if you’re tracking an object that has moved off frame beyond the last keyframe.
screen and you want After Effects to extrapolate where
it would be if it kept moving at the same speed and in
the same direction.
. loopOut("offset") Works similarly to "cycle", except
that instead of returning to the value of the first
keyframe, each loop of the animation is offset by an
amount equal to the value at the end of the previous
loop. This produces a cumulative or stair-step effect.
loopIn() operates the same way as loopOut(), except that
the looping occurs before the first keyframe instead of
after the last keyframe. Both loopIn() and loopOut() will
accept a second, optional parameter that specifies how
many keyframes to loop. Actually, it’s easier to think of it
as how many keyframed segments to loop. For loopOut()
the segments are counted from the last keyframe toward
the layer’s In point. For loopIn() the segments are counted
from the first keyframe toward the layer’s Out point. If you
leave this parameter out (or specify it as 0), all keyframes
are looped. For example, this variation loops the segment
bounded by the last and next-to-last keyframes:
loopOut("cycle",1)

319


Two variations on the expressions—loopOutDuration()
and loopInDuration()—enable you to specify the time (in
seconds) as the second parameter instead of number of
keyframed segments to be looped. For loopOutDuration(),
the time is measured from the last keyframe toward the
layer’s In point. For loopInDuration(), the time is mea-
sured from the ﬁrst keyframe toward the layer’s Out point.
For example, this expression loops the two-second interval
prior to the last keyframe:
loopOutDuration("cycle",2)

If you leave out the second parameter (or specify it as 0),
the entire interval between the layer’s In point and the
last keyframe will be looped for loopOutDuration(). For
loopInDuration(), the interval from the ﬁrst keyframe to
the Out point will be looped.

Using Markers
The expression language gives you access to the attributes
of layer (and composition) markers. This can be extremely
useful for synchronizing or easily establishing timing rela-
tionships between animated events.
The marker attributes that appear most frequently in
expressions are time and index. As you might guess, the
time attribute represents the time (in seconds) where the
marker is located on the Timeline. The index attribute rep-
resents the marker’s order on the Timeline, where 1 repre-
sents the left-most marker. You can also retrieve the marker
nearest to a time that you specify by using nearestKey().
For example, to access the layer marker nearest to the cur-
rent comp time use
marker.nearestKey(time)

This can be handy, but more often you’ll want to know
the most recent previous marker. The code necessary to
retrieve it looks like this

320


n = 0;
if (marker.numKeys > 0){
n = marker.nearestKey(time).index;
if (marker.key(n).time > time){
n--;
}
}

Note that this piece of code by itself is not very useful.
When you do use it, you’ll always combine it with addi-
tional code that makes it suitable for the particular prop-
erty to which the expression will be applied. Because it’s so
versatile, and can show up in expressions for virtually any
property, it’s worth looking at in detail.
The first line creates a variable, n, and sets its value to 0. If
the value is still 0 when the routine finishes, it means that
at the current time no marker was reached or that there
are no markers on this layer.
The next line, a JavaScript if statement, checks if the layer
has at least one marker. If there are no layer markers,
After Effects skips to the end of the routine with the vari-
able n still set to 0. You need to make this test because the
next line attempts to access the nearest marker with the
statement

If After Effects attempted to execute this statement and
there were no layer markers, it would generate an error
and the expression would be disabled. It’s best to defend
against these kinds of errors so that you can apply the
expression first and add the markers later if you want to.
If there is at least one layer marker, the third line of the
expression sets n to the index of the nearest marker. Now
all you have to do is determine if the nearest marker
For more explanation of if state-
occurs before or after the current comp time with the ments, check out the “Conditionals”
statement and “Comparison Operators” sec-
tions of the JavaScript guide.
n--;
}

321


This tells After Effects to decrement n by 1 if the nearest
marker occurs later than the current time.
The result of all this is that the variable n contains the
index of the most recent previous marker or 0 if no maker
has yet been reached.
If you’re wondering about the
JavaScript decrement operator (--), So how can you use this little routine? Consider a simple
it’s described in the “Operators” sec- example.
tion of the JavaScript guide.

Trigger Animation at Markers
Say you have a keyframed animation that you want to
trigger at various times. All you need to do is drop a layer
marker (just press * on the numeric keypad) wherever you
want the action to be triggered. Then, apply this expres-
sion to the animated property:
n = 0;
if (marker.numKeys > 0){
n--;
}
}if (n == 0){
valueAtTime(0);
}else{
t = time - marker.key(n).time;
valueAtTime(t)
}

As you can see, it’s the previous marker routine with six
new lines at the end. These lines tell After Effects to use
the property’s value from time 0 if there are no previous
markers. Otherwise, variable t is deﬁned to be the time
since the most recent previous marker, and the value for
that time is used.
The result of this is that the animation will run, beginning
at frame zero, wherever there is a layer marker.

Play Only Frames with Markers
Suppose you want to achieve a stop motion animation
effect by displaying only speciﬁc frames of your footage, say

322


playing only the frames when your actor reaches the apex
of a jump so he appears to ﬂy or hover.
First enable time remapping for the layer, then scrub
through the Timeline and drop a layer marker at each
frame that you want to include. Finally, apply this expres-
sion to the Time Remap property:
n = marker.numKeys;
if (n > 0){
f = timeToFrames(time);
idx = Math.min(f + 1, n);
marker.key(idx).time
}else{
value
}

In this expression, the variable n stores the total number of
markers for the layer. The if statement next checks whether
there is at least one marker. If not, the else clause executes,
instructing After Effects to run the clip at normal speed. If
there are markers, the expression ﬁrst calculates the current
frame using timeToFrames(), which converts whatever time
you pass to it into the appropriate frame number. Here,
it receives the current comp time and returns the current
frame number, which is stored in variable f.
Next you need to convert the current frame number to a
corresponding marker index for the frame you actually
want to display. It turns out that all you need to do is add 1.
That means when the current frame is 0, you actually want
to show the frame that is at marker 1. When frame is 1, you
want to show the frame at marker 2, and so on. The line
idx = Math.min(f + 1, n);

calculates the marker index and stores it in the variable
idx. Using Math.min() ensures the expression never tries to
access more markers than there are (which would gener-
See “The Math Object” in the
ate an error and disable the expression). Instead, playback JavaScript guide for more informa-
freezes on the last frame that has a marker. tion on Math.min().
Finally, you use the idx variable to retrieve the time of the
corresponding marker. This value becomes the result of

323


the expression, which causes After Effects to display the
frame corresponding to the marker (Figure 10.6).

Figure 10.6 The bottom line in the graph represents how the Time Remap prop-
erty would behave without the expression. As you would expect, it is a linear,
gradual increase. The upper, stair-stepped line is the result of the expression.
Because the expression only plays frames with markers (represented in the graph
by small triangles) time advances much more quickly.

Time Remapping Expressions
There are many ways to create interesting effects with time
remapping expressions. You’ve already seen one (the last
expression in the previous section). Here are a few more
illustrative examples.

Jittery Slow Motion
Here’s an interesting slow motion effect where frames 0, 1,
2, and 3 play, followed by frames 1, 2, 3, and 4, then 2, 3,
4, and 5, and so on. First, enable time remapping for the
layer and then apply this expression to the Time Remap
property:
cycle = 4;
f = timeToFrames();
framesToTime(Math.floor(f/cycle) + f%cycle);

The ﬁrst line sets the value of the variable cycle to the
number of frames After Effects will display in succession (4
in this case). The second line sets variable f to the frame
For more detail on Math.
floor() and the % modulo
number corresponding to the current comp time. Next
operator, see “The Math Object” comes a tricky bit of math using JavaScript’s Math.floor()
and “Operators” sections of the method and its % modulo operator. The result is a repeat-
JavaScript guide. ing sequence (whose length is determined by the variable
cycle) where the starting frame number increases by 1 for
each cycle.

324


Wiggle Time
This effect uses multiple copies of the same footage to
achieve a somewhat creepy echo effect. This effect actually
involves three short expressions: one for Time Remap, one
for Opacity, and one for Audio Levels. First, you enable
time remapping for the layer. Then apply the three expres-
sions and duplicate the layer as many times as necessary to
create the look you want (Figure 10.7).

Figure 10.7 The time-wiggling effect
with multiple layers.

Note that this time-wiggling effect is interesting, even with
a single layer. The Opacity and Audio Levels expressions
are necessary only if you want to duplicate the layer.
The expression for the Time Remap property is
Math.abs(wiggle(1,1))

wiggle() is an extremely useful tool that can introduce
a smooth or fairly frenetic randomness into any anima-
tion, depending on your preference. wiggle() accepts
ﬁve parameters, but only frequency and amplitude are
required. Check the After Effects documentation for an
explanation of what the remaining three optional param-
eters do.
The ﬁrst parameter, frequency, represents the frequency
of the wiggle in seconds; wiggle(1,1) varies the playback

325


speed at the rate of once per second. The second param-
eter is the amplitude of the wiggle, given in the units of the
parameter to which wiggle() is applied, which in this case
is also seconds. So, wiggle(1,1) lets the playback time devi-
ate from the actual comp time by as much as one second in
either direction.
You use Math.abs() to make sure that the wiggled time
value never becomes less than zero, which would cause the
layer to sit at frame zero.
For more detail on Math.abs(),
see “The Math Object” section of the The Opacity expression gives equal visibility to each layer.
online JavaScript guide. Here’s what it looks like:
(index/thisComp.numLayers)*100

This is simply the ratio of the layer’s index divided by the
total number of layers in the comp, times 100%. That
means if you duplicate the layer four times (for a total of
five layers), the top layer will have an Opacity of 20%, the
second layer will have an Opacity of 40%, and so on, until
the bottom (fifth) layer, which will have an Opacity of
100%. This allows each layer to contribute equally to the
final result.
If the footage has audio, you have a couple of choices.
You can turn the audio off for all but one of the layers,
or you can use an expression for Audio Levels that nor-
malizes them so that the combined total audio level is
roughly the same as it would be for a single layer. I think
the second option enhances the creepiness of the effect;
here’s the Audio Levels expression for a stereo audio
source (for a mono source you could just leave out the
second line of the expression):
db = -10*Math.log(thisComp.numLayers)/Math.log(10);
[db,db]

This is just a little decibel math that reduces the level
of each layer based on how many total layers there are
(using the comp attribute numLayers). You’ll also notice
a couple of JavaScript elements you haven’t encountered
before: Math.Log() and an array (the second line of the

326


expression). In expressions, you specify and reference
the value of a multidimensional property, such as both
channels of the stereo audio level, using array square
bracket syntax.
For more information on Math.
log() see the “Math Object”
Random Time section of the JavaScript guide on
In this example, instead of having the time of each layer the accompanying disc; for more on
arrays see the “Arrays” section.
wander around, the expression offsets each layer’s play-
back time by a random amount. The expression you need
for the Time Remap property is
maxOffset = 0.7;
seedRandom(index, true);
time + random(maxOffset);

The first thing to notice about this expression is the use of
seedRandom() and random() and the relationship between
these functions. If you use random() by itself, you get a dif-
ferent random number at each frame, which is usually not
what you want. The solution is seedRandom(), which takes
two parameters. The first is the seed. It controls which
random numbers get generated by random(). If you specify
only this parameter, you will have different random num-
bers on each frame, but they are an entirely new sequence
of numbers. It’s the second parameter of seedRandom() that
enables you to slow things down. Specifying this parameter
as true tells After Effects to generate the same random
numbers on each frame. The default value is false, so if
you don’t specify this parameter at all, you get different
numbers on each frame. It’s important to note that seed-
Random() doesn’t generate anything by itself. It just defines
the subsequent behavior of random().
Here’s an example. This Position expression randomly
moves a layer to a new location in the comp on each frame:
random([thisComp.width,thisComp.height])

This variation causes the layer to stay in one random
location:
seedRandom(1,true);

327


This version is the same as the previous one, except that it
generates a different, single random location because the
value of the seed is different:
seedRandom(2,true);

Let’s get back to the Time Remap expression. The first line
creates the variable maxOffset and sets it to the maximum
value, in seconds, that each layer’s playback time can
More About random()
deviate from the actual comp time. The maximum for the
There are several ways to use random(). If
you call it with no parameters, it will gener- example is 0.7 seconds.
ate a random number between 0 and 1. If you The next line tells After Effects that you want the random
provide a single parameter (as in the Random
Time example), it will generate a random number number generator (random()) to generate the same ran-
between 0 and the value of the parameter. If you dom number on each frame.
provide two parameters, separated by a comma,
it will generate a random number between those The last line of the expression calculates the final Time
two parameters. It’s important to note that the Remap value, which is just the sum of the current comp
parameters can be arrays instead of numbers. For time plus a random offset between 0 and 0.7 seconds.
example, this expression will give you a random 2D
position somewhere within the comp: Next, you would apply the Opacity and Audio Levels expres-
random ([thisComp.width, sions from the wiggle() example so that each layer’s video
➥thisComp.height]) and audio will be weighted equally. Duplicate the layer as
many times as necessary to get the effect you like.
In addition to random(), After Effects provides
gaussRandom(), which operates in much the
same way as random() except that the results Layer Space Transforms
have more of a Gaussian distribution to them. That In the world of expressions, layer space transforms are
is, more values are clustered toward the center of
the range, with fewer at the extremities. Another indispensible, but they present some of the most difficult
difference is that with gaussRandom(), concepts to grasp. There are three coordinate systems in
sometimes the values may actually be slightly After Effects, and layer space transforms provide you with
outside the specified range, which never happens the tools you need to translate locations from one coordi-
with random().
nate system to another.
One coordinate system represents a layer’s own space. This
is the coordinate system relative (usually) to the layer’s
upper-left corner. In this coordinate system [0, 0] repre-
sents a layer’s upper-left corner, [width, height] represents
the lower-right corner, and [width, height]/2 represents
the center of the layer. Note that unless you move a layer’s
anchor point, it too will usually represent the center of the
layer in the layer’s coordinate system.

328


The second coordinate system represents world space. World
coordinates are relative to [0, 0, 0] of the composition.
This starts out at the upper-left corner of a newly created
composition, but it can end up anywhere relative to the
comp view if the comp has a camera and the camera has
been moved, rotated, or zoomed.
The last coordinate system represents comp space. In this
coordinate system, [0, 0] represents the upper-left corner
of the camera view (or the default comp view if there is no
camera), no matter where the camera is located or how it is
oriented. In this coordinate system, the lower-right corner
of the camera view is given by [thisComp.width, thisComp.
height]. In comp space, the Z coordinate really doesn’t
have much meaning because you’re only concerned with
the ﬂat representation of the camera view (Figure 10.8).

Figure 10.8 This illustration shows
the three coordinate systems of After
Effects. Positions in the yellow layer’s
coordinate system are measured
relative to its upper-left corner. The 3D
null is positioned at [0,0,0] in the comp
so that it shows the reference point of
the world coordinate system (which,
in this case is exactly the same as the
null’s layer coordinate system). The
comp’s coordinate system is always
referenced to the upper-left corner of
the Comp view, which in this case no
longer matches the world coordinate
system because the camera has been
moved and rotated.

So when would you use layer space transforms? One of the
most common uses is probably to provide the world coordi-
nates of a layer that is the child of another layer. When you
make a layer the child of another layer, the child layer’s
Position value changes from the world space coordinate
system to layer space of the parent layer. That is, the child
layer’s Position becomes the distance of its anchor point

329


from the parent layer’s upper-left corner. So a child layer’s
Position is no longer a reliable indicator of where the layer
is in world space. For example, if you want another layer
to track a layer that happens to be a child, you need to
translate the child layer’s position to world coordinates.
Another common application of layer space transforms
allows you to apply an effect to a 2D layer at a point that
corresponds to where a 3D layer appears in the comp view.
Both of these applications will be demonstrated in the fol-
lowing examples.

Effect Tracks Parented Layer
To start, consider a relatively simple example: You have
a layer named star that’s the child of another layer, and
you want to rotate the parent, causing the child to orbit
the parent. You have applied CC Particle Systems II to a
comp-sized layer and you want the Producer Position of
the particle system to track the comp position of the child
layer. The expression you need to do all this is
L = thisComp.layer("star");
L.toComp(L.transform.anchorPoint)

The ﬁrst line is a little trick I like to use to make the follow-
ing lines shorter and easier to manage. It creates a variable
L and sets it equal to the layer whose position needs to be
translated. It’s important to note that you can use variables
to represent more than just numbers. In this case the vari-
able is representing a layer object. So now, when you want
to reference a property or attribute of the target layer,
instead of having to preﬁx it with thisComp.layer("star"),
you can just use L.
In the second line the toComp() layer space transform
translates the target layer’s anchor point from the layer’s
own space to comp space. The transform uses the anchor
point because it represents the layer’s position in its own
layer space. Another way to think of this second line is
“From the target layer’s own layer space, convert the target
layer’s anchor point into comp space coordinates.”
This simple expression can be used in many ways. For exam-
ple, if you want to simulate the look of 3D rays emanating

330


from a 3D shape layer you can create a 3D null and make
it the child of the shape layer. You then position the null
some distance behind the shape layer. Then apply the CC
Light Burst 2.5 effect to a comp-sized 2D layer and apply
this expression to effect’s Center parameter (Figure 10.9):
L = thisComp.layer("source point");
L.toComp(L.anchorPoint)

Figure 10.9 The Center parameter for
the CC Light Burst Effect is obtained
by converting the null’s position in
3D world space to its corresponding
position in comp space.

(Notice that this is the same expression as in the previous
example, except for the name of the target layer: source
point, in this case). If you rotate the shape layer, or move a
camera around, the rays seem to be coming from the posi-
tion of the null.

Apply 2D Layer as Decal onto 3D Layer
Sometimes you may need to use more than one layer space
transform in a single expression. For example, you might
want to apply a 2D layer like a decal to a 3D layer using the
Corner Pin effect. To pull this off you need a way to mark
on the 3D layer where you want the corners of the 2D layer
to be pinned. Apply four Point Controls to the 3D layer,
and you can then position each of the 2D layer’s corners
individually on the surface of the 3D layer. To keep things
simple, rename each of the Point Controls to indicate the

331


corner it represents, making the upper-left one UL, the
upper-right UR, and so on. Once the Point Controls are
in place, you can apply an expression like this one for the
upper-left parameter to each parameter of the 2D layer’s
Corner Pin effect:
L = thisComp.layer("target");
fromComp(L.toComp(L.effect("UL")("Point")))

Expression Controls
Expression Controls are actually layer effects whose main purpose is to allow you to
attach user interface controls to an expression. These controls come in six versions:
. Slider Control
. Point Control
. Angle Control
. Checkbox Control
. Color Control
. Layer Control

All types of controls (except the Layer Control) can be keyframed and can themselves
accept expressions. The most common use, however, is to enable you to set or change
a value used in an expression calculation without having to edit the code. For example,
you might want to be able to easily adjust the frequency and amplitude
parameters of a wiggle() expression. You could accomplish this by applying two
Slider Controls to the layer with the expression (Effects > Expression Controls). It’s
usually a good idea to give your controls descriptive names; say you change the name
of the first slider to frequency and the second one to amplitude. You would then set
up your expression like this (using the pickwhip to create the references to the sliders
would be smart):
freq = effect(“frequency”)(“Slider”);
amp = effect(“amplitude”)(“Slider”);
wiggle(freq, amp)

Now, you can control the frequency and amplitude of the wiggle via the sliders. With
each of the control types (again, with the exception of the Layer Control) you can edit
the numeric value directly, or you set the value using the control’s gadget.

One unfortunate side note about expression controls is that because you can’t apply
effects to cameras or lights, neither can you apply expression controls to them.

332


The first line is just the little shorthand trick so that you
can reference the target layer (the 3D layer in this case)
more succinctly. The second line translates the position of
Point Controls from the 3D layer’s space to the layer space
of the 2D layer with the Corner Pin effect. There are no
layer-to-layer space transforms, however, so the best you
can do is transform twice: first from the 3D layer to comp
space and then from comp space to the 2D layer. (Remem-
ber to edit the expression slightly for each of the other
corner parameters so that it references the corresponding
Point Control on the 3D layer.)
So, inside the parentheses you convert the Point Control
from the 3D layer’s space into comp space. Then you con-
vert that result to the 2D layer’s space. Nothing to it, right?

Reduce Saturation Away From Camera
Let’s change gears a little. You want to create an expression
that reduces a layer’s saturation as it moves away from the
camera in a 3D scene. In addition, you want this expression
to work even if the target layer and the camera happen
to be children of other layers. You can accomplish this by
applying the Color Balance (HLS) effect to the target layer
and applying this expression to the Saturation parameter:
minDist = 900;
maxDist = 2000;

C = thisComp.activeCamera.toWorld([0,0,0]);
dist = length(toWorld(transform.anchorPoint), C);
ease(dist, minDist, maxDist, 0, -100)

The first two lines define variables that will be used to set
the boundaries of this effect. If the target layer’s distance
from the camera is less than minDist, you’ll leave the
Saturation setting unchanged at 0. If the distance is greater
than maxDist you want to completely desaturate the layer
with a setting of -100.
The third line of the expression creates variable C, which
represents the position of the comp’s currently active
camera in world space. It’s important to note that cameras
and lights don’t have anchor points, so you have to convert

333


a specific location in the camera’s layer space. It turns out
that, in its own layer space, a camera’s location is repre-
sented by the array [0,0,0] (that is, the X, Y, and Z coordi-
nates are all 0).
The next line creates another variable, dist, which rep-
resents the distance between the camera and the anchor
point of the target layer. You do this with the help of
length(), which takes two parameters and calculates the
distance between them. The first parameter is the world
location of the target layer and the second parameter is the
world location of the camera, calculated previously.
All that’s left to do is calculate the actual Saturation value
based on the layer’s current distance from the camera. You
do this with the help of ease(), one of the expression lan-
guage’s amazingly useful interpolation methods. What this
line basically says is “as the value of dist varies from minDist
to maxDist, vary the output of ease() from 0 to –100.”

Interpolation Methods
After Effects provides some very handy global interpola-
tion methods for converting one set of values to another.
Say you wanted an Opacity expression that would fade in
over half a second, starting at the layer’s In point. This is
very easily accomplished using the linear() interpolation
method:
linear(time, inPoint, inPoint + 0.5, 0, 100)

As you can see, linear() accepts five parameters (there is
also a seldom-used version that accepts only three param-
eters), which are in order
. Input value that is driving the change
. Minimum input value
. Maximum input value
. Output value corresponding to the minimum
input value
. Output value corresponding to the maximum
input value

334


In the example, time is the input value (first parameter),
and as it varies from the layer’s In point (second parame-
ter) to 0.5 seconds beyond the In point (third parameter),
the output of linear() varies from 0 (fourth parameter)
to 100 (fifth parameter). For values of the input parameter
that are less than the minimum input value, the output of
linear() will be clamped at the value of the fourth param-
eter. Similarly, if the value of the input parameter is greater
than the maximum input value, the output of linear() will
be clamped to the value of the fifth parameter. Back to the
example, at times before the layer’s In point the Opac-
ity value will be held at 0. From the layer’s In point until
0.5 seconds beyond the In point, the Opacity value ramps
smoothly from 0 to 100. For times beyond the In point +
0.5 seconds, the Opacity value will be held at 100. Some-
times it helps to read it from left to right like this: “As the
value of time varies from the In point to 0.5 seconds past
the In point, vary the output from 0 to 100.”
The second parameter should always be less than the third
parameter. Failure to set it up this way can result in some
bizarre behavior.
Note that the output values need not be numbers. Arrays
work as well. If you want to slowly move a layer from the
composition’s upper-left corner to the lower-right corner
over the time between the layer’s In point and Out point,
you could set it up like this:
linear(time, inPoint, outPoint, [0,0],
➥[thisComp.width, thisComp.height])

There are other equally useful interpolation methods in
addition to linear(), each taking exactly the same set of
parameters. easeIn() provides ease at the minimum value
side of the interpolation, easeOut() provides it at the
maximum value side, and ease() provides it at both. So if
you wanted the previous example to ease in and out of the
motion, you could do it like this:
ease(time, inPoint, outPoint, [0,0], [thisComp.width,
➥thisComp.height])

335


Fade as Move Away From Camera
Just as you can reduce a layer’s saturation as it moves away
from the camera, you can reduce its Opacity. The expres-
sion is, in fact, quite similar:
minDist = 900;
maxDist = 2000;

dist = length(toWorld(transform.anchorPoint), C);
ease(dist, minDist, maxDist, 100, 0)

The only differences between this expression and the pre-
vious one are the fourth and ﬁfth parameters of the ease()
statement. In this case, as the distance increases from 900
to 2000 the opacity fades from 100% to 0%.

From Comp Space to Layer Surface
There’s a somewhat obscure layer space transform that
you haven’t looked at yet, namely fromCompToSurface().
This translates a location from the current comp view to
the location on a 3D layer’s surface that lines up with that
point (from the camera’s perspective). When would that
be useful?
Imagine you have a 2D comp-sized layer named Beam, to
which you have applied the Beam Effect. You want a Lens
Flare effect on a 3D layer to line up with the ending point
of the Beam effect on the 2D layer. You can do it by apply-
ing this expression to the Flare Center parameter of the
Lens Flare effect on the 3D layer:
beamPos = thisComp.layer("beam").effect("Beam")
➥("Ending Point");
fromCompToSurface(beamPos)

First, store the location of the ending point of the Beam
effect into the variable beamPos. Now you can take a couple
of short cuts because of the way things are set up. First,
the Ending Point parameter is already represented as a
location in the Beam layer’s space. Second, because the
Beam layer is a comp-sized layer that hasn’t been moved or
scaled, its layer space will correspond exactly to the Cam-
era view (which is the same as comp space). Therefore, you

336


can assume that the Ending Point is already represented
in comp space. If the Beam layer were a different size than
the comp, located somewhere other than the comp’s cen-
ter, or scaled, you couldn’t get away with this. You would
More About sampleImage()
have to convert the Ending Point from Beam’s layer space
to comp space. You can sample the color and alpha data of a
rectangular area of a layer using the layer method
Now all you have to do is translate the beamPos variable sampleImage(). You supply up to four
parameters to sampleImage() and it returns
from comp space to the corresponding point of the sur-
color and alpha data as a four-element array ([red,
face of the layer with the Lens Flare, which is accomplished green, blue, alpha]), where the values have been
easily with fromCompToSurface(). normalized so that they fall between 0.0 and 1.0.
The four parameters are
You’ll look at one more example of layer space transforms
. Sample point
in the big ﬁnale “Extra Credit” section at the end of the
. Sample radius
chapter.
. Post-effect flag
. Sample time
Color Sampling and Conversion
Here’s an example that demonstrates how you work with The sample point is given in layer space
colors in a expression. The idea here is that you want to coordinates, where [0, 0] represents the center
of the layer’s top left pixel. The sample radius is
vary the opacity of an animated small layer based on the a two-element array ([x radius, y radius]) that
lightness (or luminosity) of the pixels of a background specifies the horizontal and vertical distance from
layer that currently happen to be under the moving layer. the sample point to the edges of the rectangular
The smaller layer will become more transparent as it passes area being sampled. To sample a single pixel, you
would set this value to [0.5, 0.5] (half a pixel in
over dark areas of the background and more opaque as it each direction from the center of the pixel at the
passes over lighter areas. Fortunately, the expression lan- sample point). The post-effect flag is optional (its
guage supplies a couple of useful tools to help out. default value is true if you omit it) and specifies
whether you want the sample to be taken after
Before examining the expression, we need to talk about masks and effects are applied to the layer (true) or
the way color data is represented in expressions. An indi- before (false). The sample time parameter specifies
vidual color channel (red, blue, green, hue, saturation, the time at which the sample is to be taken. This
parameter is also optional (the default value is the
lightness, or alpha) is represented as a number between current composition time), but if you include it, you
0.0 (fully off) and 1.0 (fully on). A complete color space must also include the post-effect flag parameter.
representation consists of an array of four such channels. As an example, here’s how you could sample the
Most of the time you’ll be working in red, blue, green, red value of the pixel at a layer’s center, after any
effects and masks have been applied, at a time one
and alpha (RGBA) color space, but you can convert to second prior to the current composition time:
and from hue, saturation, lightness, and alpha (HSLA)
mySample = sampleImage([width/
color space. This example uses sampleImage() to extract ➥height]/2, [0.5,0.5], true,
RGBA data from a target layer called background. Then ➥time – 1);
rgbToHsl() converts the RGBA data to HSLA color space myRedSample = mySample[0];

so that you can extract the lightness channel, which will
then be used to drive the Opacity parameter of the small
animated layer. Here’s the expression:

337


sampleSize = [width, height]/2;
target = thisComp.layer("background");
rgba = target.sampleImage(transform.position,
sampleSize, true, time);
hsla = rgbToHsl(rgba);
hsla[2]*100

First you create the variable sampleSize and set its value as
an array consisting of half the width and height of the layer
whose opacity will be controlled with the expression. Essen-
tially this means that you’ll be sampling all of the pixels of
the background layer that are under smaller layers at any
given time.
The second line just creates the variable target, which will
be a shorthand way to refer to the background layer. Then
sampleImage() retrieves the RGBA data for the area of the
background under the smaller layer and stores the result-
ing array in the variable rgba. See the sidebar “More About
sampleImage()” earlier in the chapter for details on all the
parameters of sampleImage().
Next rgbToHsl() converts the RGBA data to HSLA color
space and stores the result in variable hsla. Finally, because
the lightness channel is the third value in the HSLA array
you use the array index of [2] to extract it (see the “Arrays”
section of the JavaScript guide if this doesn’t make sense
to you). Because it will be a value between 0.0 and 1.0, you
just need to multiply it by 100 to get it into a range suitable
to control the Opacity parameter (Figure 10.10).

Figure 10.10 The small blue layer
becomes more transparent as it
passes over darker areas of the back-
ground image.

338


Extra Credit
Congratulations on making it this far. The remaining
examples build on concepts covered earlier, but I have
saved them for this section because they are particularly
tricky or involve some complex math. I’m presenting them
mainly to entice you to take some time and figure out how
they work.

Fade as Turn Away From Camera
Let’s briefly return to the world of layer space transforms
and examine a simple idea that requires only a short
expression, but one with a lot of complicated vector math
going on under the hood. The idea is that you want a 3D
layer to fade out as it turns away from the camera. This
needs to work not only when the layer rotates away from
the camera, but also if the camera orbits the layer. And of
course, it should still work if either the layer or the camera
happens to be the child of another layer. Take a look at an
expression for Opacity that will accomplish this:
minAngle = 20;
maxAngle = 70;

v1 = normalize(toWorld(transform.anchorPoint) – C);
v2 = toWorldVec([0,0,1]);
angle = radiansToDegrees(Math.acos(dot(v1, v2)));
ease(angle, minAngle, maxAngle, 100, 0)

The first two lines just create two variables (minAngle and
maxAngle) that establish the range of the effect. Here you
set their values so that when the layer is within 20 degrees
of facing the camera, it will be at 100% Opacity and its
Opacity will fade from 100% to 0% as the angle increase to
70 degrees. Beyond 70 degrees, Opacity will be 0%.
Next you create a variable C that represents the position of
the comp’s active camera in world space. You’ve seen this
before, in the expression where the layer fades as it moves
away from the camera.

339


Now starts the vector math. Things get a little bumpy from
here. Briefly, a vector is an entity that has a length and a
direction, but has no definite position in space. I like to
think of vectors as arrows that you can move around, but
they always keep the same heading. Fortunately the expres-
sion language provides a pretty good arsenal of tools to
deal with vectors.
To figure out the angle between the camera and the layer
with the expression, you’re going to need two vectors. One
will be the vector that points from the center of the layer
towards the camera. The other will be a vector that points
outward from the center of the layer along the Z axis.
To calculate the first vector (variable v1), convert the
layer’s anchor point to world space coordinates and
subtract from that value the location of the camera in
world space. What you’re doing is subtracting two points
in space. Remember, in After Effects, each 3D position
in space is represented by an array: [x,y,z]. The result of
subtracting two points like this gives you a vector. This
vector has a magnitude representing the distance between
the two points and a direction (in this case, the direction
from the layer to the camera). You can use normalize() to
convert the vector to what is known as a unit vector, which
maintains the direction of the original vector, but sets its
length to 1. This simplifies the upcoming determination of
the angle between two vectors.
Next you create the second vector (variable v2). You can
create the necessary unit vector in one step this time by
using toWorldVec([0,0,1]) to create a vector of length 1
pointed along the layer’s Z axis.
Now you have your two vectors. To calculate the angle
between two vectors, you use what is known as the vector
dot product. I won’t go into great detail about how it works
(there’s a lot of information on the Internet if you’re
curious), but it turns out that if you use unit vectors, the
vector dot product will directly give you the arc cosine of
the angle between the two vectors. Luckily, the expression
language gives us a built-in function, dot(), to calculate the
dot product.

340


So now you can calculate the angle you need (and store
it in variable angle) in three steps. First you take the dot
product of the two vectors, producing the arc cosine of
the angle. Then you use Math.acos() to convert that to an
angle (see the “Math Object” section of the JavaScript guide
for more information). Because the result of Math.acos()
will be in radians, you need to convert it to degrees so
that it will be in the same units as the limits minAngle and
maxAngle. Fortunately, the expression language provides
radiansToDegrees() to make the conversion.

The ﬁnal step is to use the interpolation method ease() to
smoothly execute the fade as the angle increases.

Audio Triggers Effect
Earlier, you learned about linking an effect to an audio
level. You can take that idea one step further and use audio
to trigger an animated effect. The difference is subtle, but
signiﬁcant. In the earlier examples, the effect tracked the
audio level precisely, leaving the result at the mercy of the
shape of the audio level’s envelope. Here, you’re going to
use the transitioning of the audio level above some thresh-
old to trigger an animation. The animation will run until
there is another trigger event, which will cause the anima-
tion to start again from the beginning.
This is a powerful concept and there are many ways to
use it. This example triggers a decaying oscillation that is
actually contained within the expression, but you could
easily adapt this to run a keyframed animation using
valueAtTime() or to run a time remapped sequence.

The heart of this expression is what I would call a “beat
detector.” The expression basically walks backward in time,
frame-by-frame, looking for the most recent event where
the audio level transitioned from below the threshold to
above the threshold. It then uses the difference in time
between the triggering event and the current comp time to
determine how far along it should be in the animation. At
each new beat, this time resets to zero and runs until the
next beat. Take a look at this monster:

341


threshold = 20.0;

A = thisComp.layer("Audio Amplitude").effect("Both
➥Channels")("Slider");

// beat detector starts here

above = false;
frame = timeToFrames();
while (true){
t = framesToTime(frame);
if (above){
if (A.valueAtTime(t) < threshold){
frame++;
break;
}
}else if (A.valueAtTime(t) >= threshold){
above = true;
}
if (frame == 0){
break;
}
frame--
}
if (! above){
t = 0;
}else{
t = time - framesToTime(frame);
}

// animation starts here

amp = 75;
freq = 5;
decay = 2.0;

angle = freq * 2 * Math.PI * t;
amp * (-Math.cos(angle)+1)/ Math.exp(decay * t);

This expression has three sections. The first section defines
the audio level that you want to trigger the animation
and stores it into the variable threshold. It then defines

342


variable A to use as shorthand notation for the Slider Con-
trol containing the keyframed data for the audio level.
The next section is the actual beat detector. In general, the
expression starts at the current comp time and determines
if the level is currently above the threshold. If it is, the
expression moves backward in time, frame-by-frame until
it finds the most recent frame where the audio level was
below the threshold. It then determines that the triggering
event occurred on frame after that (the most recent frame
where the level transitioned from below the threshold to
above it). That transition frame is converted to time using
framesToTime(), that value is subtracted from the current
comp time, and the result (the time, in seconds, since the
triggering event) is stored into variable t.
However, if instead, the audio level at the current comp
time is below the threshold, the expression has more work
to do. It first moves backward from the current comp time,
frame-by-frame until it finds a frame where the audio level
is above the threshold. Then it continues on, looking for
the transition from below the threshold to above it. The
elapsed time since the triggering event is then calculated
and stored in variable t.
There are some other things going on in this routine, but
they mostly have to do with special cases, such as the case
where there hasn’t yet been a triggering event (in which
case the animation is held at the first frame), or where the
level is above the threshold but it has been there since the
first frame.
There are some JavaScript elements in this section that you
haven’t seen before. // denotes the start of a comment.
The routine consists mainly of a giant while() loop. This
See the “Comments” section of the
loop is unusual in that its terminating condition is set to JavaScript guide for more details
true, so it will never end on its own. It will continue to loop on comments and the “Loops”
until one of the break statements is executed. section for more information about
while(), break , and loops in
When After Effects arrives at the last section of the expres- general.
sion, variable t contains the necessary information: how
long it has been since the last triggering event. The final
section uses it to drive a decaying oscillation routine with
Math.cos() and Math.exp(). First you define the amplitude

343


of the oscillation with the variable amp. Then you deﬁne the
frequency of the oscillation (oscillations per second) with
the variable freq. Variable decay determines how fast the
oscillation decays (a higher number means a faster decay).
You might want to visit “The Math Math.cos() creates an oscillating sine wave with amplitude
Object” section of the JavaScript
amp and frequency freq, then Math.exp() reduces the
guide for more information on
Math.cos() and Math.exp(). amplitude of the oscillating wave at a rate determined by
variable decay (Figure 10.11).

Figure 10.11 The graph shows the decaying oscillation triggered whenever the audio threshold level is crossed.

Conclusion
This chapter covered a lot of ground, but still it really only
provided a hint of what’s possible with expressions. Here
are a few resources where you can ﬁnd a lot of additional
information:
. www.aenhancers.com: A forum-based site where you
can get your questions answered and take a look at
expressions contributed by others
. http://forums.creativecow.net/forum/adobe_after_
effects_expressions: A forum dedicated to expressions
. www.adobeforums.com: Adobe’s own After Effects
forum, which has a sub-forum on expressions
. www.adobe.com/go/learn_ae_cs4helphome: The
online version of After Effects Help
. www.motionscript.com: My own Web site, which has a
lot of examples and analysis

344

CHAPTER

11
32-Bit HDR Compositing
and Color Management

True realism consists in revealing the surprising
things which habit keeps covered and prevents us
from seeing.
—Jean Cocteau (French director,
painter, playwright, and poet)

32-Bit HDR Compositing
and Color Management

W hether you are directly aware of limitations or not,
you no doubt realize that there are limits to the ways
images are processed and displayed on a computer. Your
monitor likely only displays 8 bits of color per channel, and
while its size (in pixel dimensions) has steadily increased
over the last few years, this color depth limitation has
hardly budged.
You may also be aware that although an After Effects proj-
ect, by default, operates in the same limited 8-bit-per-chan-
nel mode as your monitor, this is hardly the optimal way
to create an image. Other modes, models, and methods
for color are available, including high-bit depths, alternate
color spaces, and color management, and few topics in
After Effects generate as much curiosity or confusion as
these. Each of the features detailed here improves upon
the standard digital color model you know best, but at the
cost of requiring better understanding on your part.
In After Effects CS4 the process centers around Color Man-
agement, whose name would seem to imply that it is an
automated process to manage colors for you, when in fact
it is a complex set of tools allowing (even requiring) you to
effectively manage color.
On the other hand, 32-bit High Dynamic Range (HDR)
compositing is routinely ignored by artists who could

346


benefit from it, despite that it remains uncommon for
source files to contain over-range color data, which are
pixel values too bright for your monitor to display.
Film can and typically does contain these over-range color
values. These are most often brought into After Effects as
10-bit log Cineon or DPX files, and importing, converting,
and writing this format requires a bit of special knowledge.
It’s an elegant and highly standardized system that has rel-
evance even when you’re working with the most up-to-date,
high-end digital cameras.

Color Management: Why Bother?
It’s normal to wish Color Management would simply go
away. So many of us have produced footage with After
Effects for years and devised our own systems to manage
color through each stage of production. We’ve assumed,
naively perhaps, that a pixel is a pixel and as long as we
control the RGB value of that pixel, we maintain control
over the appearance of the image.
The problem with this way of thinking is that it’s tied to the
monitor. The way a given RGB pixel looks on your monitor
is somewhat arbitrary—I’m typing this on a laptop, and I
know that its monitor has higher contrast than my desktop
monitors, one of which has a bluer cast than the other if
I don’t adjust them to match. Not only that, the way that
color operates on your monitor is nothing like the way it
works in the real world, or even in a camera. Not only is
the dynamic range far more limited, but also an arbitrary
gamma adjustment is required to make images look right.
Color itself is not arbitrary. Although color is a com-
pletely human phenomenon—“color” as such does not
exist except in our vision system and that of other higher
primates—it is the result of measurable natural phenom-
ena. Because the qualities of a given color are measurable
to a large degree, a system is evolving to measure them,
and Adobe is attempting to spearhead the progress of that
system with its Color Management features.

347

Chapter 11 32-Bit HDR Compositing and Color Management

Completely Optional
The Color Management feature set in After Effects is
completely optional and disabled by default. Its features
become necessary in cases including, but not necessarily
limited to, the following:
. A project relies on a color managed file (with an
embedded ICC Profile). For example, a client provides
an image or clip with specific managed color settings
and requires that the output match.
. A project will benefit from a linearized 1.0 gamma
working space. If that means nothing to you, read on;
this is the chapter that explains it.
. Output will be displayed in some manner that’s not
directly available on your system.
. A project is shared and the color is adjusted on a
variety of workstations, each with a calibrated moni-
tor. The goal is for color corrections made on a given
workstation to match once the shot moves on from that
workstation.
To achieve these goals requires that some old rules be
broken and new ones established.

Related and Mandatory
Other changes introduced in After Effects CS4 seem tied
to Color Management but come into play even if you never
enable it:
. A video file in a DV or other Y’CrCb (YUV) format
requires (and receives) automatic color interpretation
upon import into After Effects, applying settings that
would previously have been up to you to add. This is
done by MediaCore, a little known Adobe application
that runs invisibly behind the scenes of Adobe video
applications (see “Input Profile and MediaCore,” later
in this chapter).
. QuickTime gamma settings in general have become
something of a moving target as Apple adds its own
form of color management, whose effects vary from
codec to codec. As a result, there are situations in

348


which imported and rendered QuickTimes won’t
look right. This is not the fault of Color Management,
although you can use the feature set to correct the
problems that come up (see “QuickTime,” below).
. Linear blending (using a 1.0 gamma only for pixel-
blending operations without converting all images to
linear gamma) is possible without setting a linearized
Project Working Space or enabling Color Management
(see the section “Blend Colors Using 1.0 Gamma”).
Because these issues also affect how color is managed, they
tend to get lumped in with the Color Management system
when in fact they can be unique from it.

A Pixel’s Journey Through After Effects
Join me now as we follow color through After Effects,
noting the various features that can affect its appearance
or even its very identity—its RGB value. Although it’s not
mandatory, it’s best to increase that pixel’s color ﬂexibility
and accuracy, warming it up to get it ready for the trip, by
raising project bit depth above 8 bpc. Here’s why.

16-Bit-Per-Channel Composites
A 16-bit-per-channel color was added to After Effects 5.0
for one basic reason: to eliminate color quantization, most
commonly seen in the form of banding where subtle gradi-
Many but not all effects and
ents and other threshold regions appear in an image. In 16 plug-ins support 16 bpc color. To
bpc mode there are 128 extra gradations between each R, discern which ones do, with your
G, B, and A value contained in the familiar 8 bpc mode. project set to the target bit depth
(16 bpc in this case), choose Show
Those increments are typically too ﬁne for your eye to 16 bpc-Capable Effects Only from
distinguish (or your monitor to display), but your eye eas- the Effects & Presets panel menu.
ily notices banding, and when you start to make multiple Effects that are only 8 bpc aren’t
off-limits; you should just be careful
adjustments to 8 bpc images, as may be required by Color to place them where they are least
Management features, banding is bound to appear in edge likely to cause banding—typically
thresholds and shadows, making the image look bad. either at the beginning or the end
of the image pipeline.
You can raise color depth in your project by either
Alt/Option-clicking on the color depth setting at the
bottom of the Project panel or via the Depth menu in
File > Project Settings. The resulting performance hit
typically isn’t as bad as you might think.

349


Most digital artists prefer 8 bpc colors because we’re so
used to them, but switching to 16 bpc mode doesn’t mean
you’re stuck with incomprehensible pixel values of 32768,
0, 0 for pure red or 16384, 16384, 16384 for middle gray.
In the panel menu of the Info panel, choose whichever
numerical color representation works for you; this setting
is used everywhere in the application, including the Adobe
color picker (Figure 11.1). The following sections use
8 bpc values despite referring to 16 bpc projects.

Figure 11.1 Love working 16 bpc but
hate analyzing 16-bit values that go
up to 32768? Choose 8 bpc in the Info
Panel Menu to display familiar 0 to
255 values. Or better yet, use Decimal
values in all bit depths.

Monitor Calibration
Sometimes it becomes obvious that RGB values alone can-
not describe pure colors; if you don’t know what I’m talk-
ing about, connect a still-working decade-old CRT monitor
to your system and see how it looks. You can imagine that a
255,255,255 white would be likely to look blue or yellow.
Assuming your monitor isn’t that far out of whack, third-
party color calibration hardware and software can be used
to generate a profile that is then stored and set as a system
preference. This monitor profile accomplishes two things:
. Defines a color space for compositing unique from
what is properly called monitor color space.
. Offers control over the color appearance of the compo-
sition. Each pixel has not only an RGB value but also an
actual precise and absolute color.

350


In other words, the color values and how they interrelate
change, as does the method used to display them.

Color Management: Disabled by Default
Import a file edited in another Adobe application such as
Photoshop or Lightroom and it likely contains an embed-
ded ICC color profile. This profile can tell After Effects
Is there an external broadcast moni-
how the colors should be interpreted and appear, instead tor attached to your system (set as
of remaining as raw electrical signals. an Output Device in Preferences >
Video Preview)? Color Management
A file called sanityCheck.tif can be found on the book’s settings do not apply to that device.
disc; it contains data and color gradients that will help you
understand linear color later in the chapter. Import this
file into After Effects and choose File > Interpret Footage >
Main (Ctrl+F/Cmd+F, or context-click instead). The Inter-
pret Footage includes a Color Management tab.
Figure 11.2 shows how this tab appears with the default set-
tings. Assign Profile is grayed out because, as the Descrip-
tion text explains, Color Management is off and color
values are not converted. Color Management is enabled as
soon as you assign a working space.

Figure 11.2 Until Color Management
is enabled for the entire project, the
embedded profile of a source image is
not displayed in the Project panel, nor
is it used.

Project Working Space
The proper choice of a working space is the one that
typically matches the “output intent,” the color space

351


corresponding to the target device. The Working Space
menu containing all possible choices is located in File >
Project Settings (Ctrl+Alt+K/Cmd+Opt+K, or just click
where you see the “bpc” setting along the bottom of the
Project panel).
Profiles above the line are considered by Adobe to be the
most likely candidates. Those below might include profiles
used by such unlikely output devices as a color printer
(Figure 11.3).

Figure 11.3 For better or worse, all of
the color profiles active on the local
system are listed as Working Space
candidates, even such unlikely targets
as the office color printer.

By default, Working Space is set to None (and thus Color
Management is off). Choose a Working Space from the
menu and Color Management is enabled, triggering the
following:
. Assigned profiles in imported files are activated and
displayed atop the Project panel when it’s selected.
. Imported files with no assigned profile are assumed to
have a profile of sRGB IEC61966-2.1, hereafter referred
to as simply sRGB.
. Actual RGB values can and will change to maintain con-
sistent color values.
Choose wisely; it’s a bad idea to change working space
mid-project once you’ve begun adjusting color, because
it will change the fundamental look of source footage
and comps.
Okay, so it’s a cop-out to say “choose wisely” and not give
actual advice. There’s a rather large document, included
on the disc and also available at www.adobe.com/devnet/
aftereffects/articles/color_management_workflow.html,
that includes a table itemizing each and every profile
included in After Effects.

352


We can just forego that for the moment in favor of a con-
cise summary:
. For HD display, HDTV (Rec. 709) is Adobe-sanctioned,
but sRGB is similar and more of a reliable standard.
. For monitor playback, sRGB is generally most suitable.
. SDTV NTSC or SDTV PAL theoretically let you forego a
preview broadcast monitor, although it’s also possible to
simulate these formats without working in them (“Dis-
play Management and Output Simulation,” below).
. Film output is an exception and is discussed later in
this chapter.
To say that a profile is “reliable” is like saying that a par-
ticular brand of car is reliable: It has been taken through
a series of situations and has caused the least problems
A small yellow + sign appears in the
under various types of duress. I realize that with color man- middle of the Show Channel icon to
agement allegedly being so scientific and all, this sounds indicate that Display Color Manage-
squirrelly, but it’s just the reality of an infinite variety of ment is active (Figure 11.4).
images heading for an infinite variety of viewing environ-
ments. There’s the scientifically tested reliability of the car
and then there are real-world driving conditions.
Gamut describes the range of possible saturation, keep- Figure 11.4 When Use Display Color
ing in mind that any pixel can be described by its hue, Management is active in the View
menu (the default after you set a
saturation, and brightness as accurately as its red, green,
working space) this icon adds a yellow
and blue. The range of hues accessible to human vision is plus symbol at its center.
rather fixed, but the amount of brightness and saturation
possible is not—32 bpc HDR addresses both. The idea is to
match, not outdo (and definitely not to undershoot) the
gamut of the target.
Working spaces change RGB values. Open sanityCheck.tif
in a viewer and move your cursor over the little bright red
square; its values are 255, 0, 0. Now change the working
space to ProPhoto RGB. Nothing looks different, but the
values are now 179, 20, 26, meaning that with this wider
gamut, color values do not need to be nearly as large in
order to appear just as saturated, and there is headroom
for far more saturation. You just need a medium capable of
displaying the more saturated red in order to see it prop-
erly with this gamut. Many film stocks can do it, and your
monitor cannot.

353


Input Profile and MediaCore
If an 8 bpc image file has no embedded profile, sRGB is
assigned (Figure 11.5), which is close to monitor color
space. This allows the file to be color managed, to preserve
its appearance even in a different color space. Toggle
Preserve RGB in the Color Management tab and the
Figure 11.5 Any imported image with
appearance of that image can change with the working
no color profile gets sRGB by default
to bring it into the color manage- space—not, generally, what you want, which is why After
ment pipeline. You can override this Effects goes ahead and assigns its best guess.
setting in Interpret Footage > Color
Management. Video formats (QuickTime being by far the most com-
mon) don’t accept color profiles, but they do require color
interpretation based on embedded data. After Effects
uses an Adobe application called MediaCore to interpret
these files automatically; it operates completely behind the
scenes, invisible to you.
In many ways, MediaCore’s automa- You know that MediaCore is handling a file when that file
tion is a good thing. After Effects has Y’CbCr in the Embedded Profile info, including DV
7.0 had a little checkbox at the
bottom of Interpret Footage labeled
and YUV format files. In such a case the Color Manage-
“Expand ITU-R 601 Luma Levels” ment tab is completely grayed out, so there is no option to
that obligated you to manage override the embedded settings.
incoming luminance range. With
MediaCore, however, you lose the Display Management and Output Simulation
ability to override the setting.
Expanded values above 235 and And in the middle of all of this great responsibility comes
below 16 are pushed out of range, a genuinely fun feature, Output Simulation, which simu-
recoverable only in 32 bpc mode. lates how your comp will look on a particular device. The
“device” in question can include film projection, and the
process of representing that environment on your monitor
works better than you might expect.
Suppose you need to know how an image (Figure 11.6)
would appear on NTSC and PAL standard definition televi-
sion, and you don’t have a standard def broadcast monitor
to preview either of those formats.
No problem. With the viewer selected choose View > Simu-
late Output > SDTV NTSC. Here’s what happens:
. The appearance of the footage changes to match the
output simulation. The viewer displays After Effects’
simulation of an NTSC monitor.

354


Figure 11.6 The source image
(courtesy of Michael Scott) is adjusted
precisely in a color managed project.

. Unlike when you change the working space, color val-
ues do not change due to output simulation.
. The image is actually assigned two separate color
profiles in sequence: a scene-referred profile to simu-
late the output profile you would use for NTSC (SDTV
Interpretation Rules
NTSC) and a second profile that actually simulates the
A file on your system named interpretation rules.
television monitor that would then display that ren- txt defines how files are automatically interpreted
dered output (SMPTE-C). To see what these settings as they are imported into After Effects. To change
are, and to customize them, choose View > Simulate anything in this file, you should be something of a
Output > Custom to open the Custom Output Simula- hacker, able to look at a line like
# *, *, *, “sDPX”, * ~ *, *, *,
tion dialog (Figure 11.7).
➥*, “ginp”, *
and, by examining surrounding lines and com-
ments, figure out that this line is commented
out (with the # sign at the beginning), and that
the next to last argument, “ginp” in quotes,
assigns the Kodak 5218 film profile if the file type
corresponds with the fourth argument, “sDPX”.
If this makes you squirm, don’t touch it, call a nerd.
In this case, removing the # sign at the beginning
would enable this rule so that DPX files would be
assigned a Kodak 5218 profile (without it, they are
assigned to the working space).

If this isn’t your cup of tea, as it won’t be for most
artists, leave it to someone willing to muck around
with this stuff.

Figures 11.7 This Custom Output Simulation dialog now nicely shows the four
stages from source RGB image to the monitor. The middle two stages are those
set by Output Simulation; the first occurs on import, the final when the image is
displayed.

355


Figure 11.8 The result of Output
Simulation shows bluer highlights,
deeper blacks (which may not read on
the printed page) and a less saturated
red dress. If you wanted the image to
appear different when projected, you
would now further adjust it with this
view active. It might then look “wrong”
with Output Simulation off, but “right”
when finally filmed out and projected.

This gets really fun with simulations of projected film
(Figure 11.8)—not only the print stock but the appear-
ance of projection is simulated, allowing an artist to work
directly on the projected look of a shot instead of waiting
until it is filmed out and projected.
Here’s a summary of what is happening to the source
image in the example project:
1. The source image is interpreted on import (on the
Having trouble with View > Footage Settings > Color Management tab) according
Simulate Output appearing grayed- to its Working Space setting.
out? Make sure a viewer window is
active when you set it; it operates 2. The image is transformed to the Project Working
on a per-viewer basis. Space; its color values will change to preserve its
appearance.
3. With View > Simulate Output and any profile selected

a. Color values are transformed to the specified Out-
put Profile.
b. Color appearance (but not actual values) is trans-
formed to a specified Simulation Profile.
4. With View > Display Color Management enabled
(required for step 3) color appearance (but not actual
values) is transformed to the Monitor Profile (the one
In Photoshop, there is no Project
Working Space, only the docu- that lives in system settings, that you created when you
ment Working Space, because there calibrated your monitor, remember?)
are no projects (no need to accom-
modate multiple sources together That leaves output, which relies only on steps 1 and 2. The
in a single nondestructive project). others are only for previewing, although you may wish to
render an output simulation (to show the filmed-out look

356


on a video display in dailies, for example). To replicate the
two-stage color conversion of output simulation:
1. Apply the Color Profile Converter effect, and match
the Output Profile setting to the one listed under
View > Simulate Output > Custom. Change the Intent
setting to Absolute Colorimetric.
2. Set a second Color Profile Converter effect, and match
the Input Profile setting to the Simulation Profile
under View > Simulate Output > Custom (leaving
Intent as the default Relative Colorimetric).
The Output Profile in the Render Queue then should
match the intended display device.
Now let’s leave simulation behind and look at what hap-
pens when you try to preserve actual colors in rendered
output. (Which is, after all, the whole point, right?)

Output Profile
By default, After Effects uses Working Space as the Output
Profile, usually the right choice. Place the comp in the
Render Queue and open the Output Module; on the Color
Management tab you can select a different profile to apply
on output. The pipeline from the last section now adds a
third step to the first two:
1. The source image is interpreted on import (on the
Footage Settings > Color Management tab).
2. The image is transformed to the working space; its
color values will change to preserve its appearance.
3. The image is transformed to the output profile speci-
fied in Output Module Settings > Color Management.
If the profile in step 3 is different from that of step 2, color
values will change to preserve color appearance. If the
output format supports embedded ICC profiles (presum-
ably a still image format such as TIFF or PSD), then a
profile will be embedded so that any other application
with color management (presumably an Adobe application
such as Photoshop or Illustrator) will continue to preserve
those colors.

357


In the real world, of course, rendered output is probably
destined to a device or format that doesn’t support color
management and embedded profiles. That’s okay, except
in the case of QuickTime, which may further change the
appearance of the file, almost guaranteeing that the output
won’t match your composition without special handling.

QuickTime
QuickTime continues to have special issues of its own
separate from, but related to Adobe’s color management.
Because Apple constantly revises QuickTime and the spec
has been in some flux, the issues particular to the version
of QuickTime at this writing (7.5.5) and how After Effects
handles it may continue to evolve.
The current challenge is that Apple has begun implement-
ing its own form of color management without sharing
the specification publicly or letting anyone know when it
changes. The gamma of QuickTime files can be specifi-
cally tagged, and the tag is then interpreted uniquely by
each codec, so files with Photo-JPEG compression have a
different gamma than files with H.264 compression. Even
files with the default Animation setting, which are effec-
tively uncompressed and assumedly neutral, display an
altered gamma, and at this writing, that gamma will display
differently depending on which application is displaying
it. Gamma handling is not even consistent among Apple
video applications.
The Match Legacy After Effects QuickTime Gamma Adjust-
ments toggle in Project Settings is not only the longest-
titled checkbox in the entire application, it is an option
you should not need, in theory at least, unless you’ve
opened up an old 7.0 (or earlier) project, or you need a
Composition to match what you see in QuickTime Player.
However, many of us deliver client review files as QuickTime
movies, so your best bet is to enable Color Management
for any project intended to output QuickTime video. The
option to disable the Match Legacy toggle is reserved for
cases in which that approach doesn’t work; these do unfor-
tunately crop up and remain a moving target as new versions
of QuickTime are released, further revising the standard.

358


To Bypass Color Management
Headaches like that make many artists long for the simpler
days of After Effects 7.0 and opt to avoid Color Manage-
ment altogether, or to use it only selectively. To completely
disable the feature and return to 7.0 behavior:
1. In Project Settings, set Working Space to None (as it is
by default).
2. Enable Match Legacy After Effects QuickTime Gamma
Adjustments.
Being more selective about how color management is
applied—to take advantage of some features while leav-
ing others disabled for clarity—is really tricky and tends to
stump some pretty smart users. Here are a couple of final
tips that may nonetheless help:
. To disable a profile for incoming footage, check Pre-
serve RGB in Interpret Footage (Color Management
tab). No attempt will be made to preserve the appear-
ance of that clip.
. To change the behavior causing untagged footage to be
tagged with an sRGB profile, in interpretation rules.txt
find this line
# soft rule: tag all untagged footage with an sRGB
➥profile
*, *, *, *, * ~ *, *, *, *, “sRGB”, *

and add a # at the beginning of the second line to
assign no profile, or change “sRGB” to a different format
(options listed in the comments at the top of the file).
. To prevent your display profile from being factored in,
disable View > Use Display Color Management and the
pixels are sent straight to the display.
. To prevent any file from being color managed, check
Preserve RGB in Output Module Settings (Color Man-
agement tab).
Note that any of the preceding tips may lead to unin-
tended consequences. Leaving a working space enabled
and disabling specific features is tricky and potentially
dangerous to your health and sanity. Add your own further
disclaimers here.

359


Film and Dynamic Range
The previous section showed how color benefits from
precision and flexibility. The precision is derived with the
steps just discussed; flexibility is the result of having a wide
dynamic range, because there is a far wider range of color
and light levels in the physical world than can be repre-
sented on your 8-bit-per-channel display.
However, there is more to color flexibility than toggling
16 bpc in order to avoid banding, or even color manage-
ment, and there is an analog image medium that is capable
of going far beyond 16 bpc color, and even a file format
capable of representing it.

Film and Cineon
To paraphrase Mark Twain, reports of film’s demise have
been exaggerated; not only that, but new formats make use
of tried and true filmic standards. Here’s a look at the film
process and the digital files on which it relies.
After film has been shot, the negative is developed, and
shots destined for digital effects work are scanned frame
by frame. During this, the Telecine process, some initial
color decisions are made before the frames are output as a
numbered sequence of Cineon files, named after Kodak’s
now-defunct film compositing system. Both Cineon files
and the related format, DPX, store pixels uncompressed at
10 bits per channel. Scanners are usually capable of scan-
ning 4 K plates, and these have become more popular for
visual effects usage, although many still elect to scan at half
resolution, creating 2 K frames around 2048 by 1536 pixels
and weighing in at almost 13 MB.

Working with Cineon Files
Because the process of shooting and scanning film is
pretty expensive, almost all Cineon files ever created are
the property of some Hollywood studio and unavailable
to the general public. The best known free Cineon file is
Kodak’s original test image, affectionately referred to as
Marcie (Figure 11.9) and available from Kodak’s Web site
(www.kodak.com/US/en/motion/-support/dlad/) or the

360


book’s disc. To get a feel for working with film, drop the
file called dlad_2048X1556.cin into After Effects, which
imports Cineon files just fine.

Figure 11.9 This universal sample
image has been converted from film
of a bygone era to Cineon format
found on the book’s disc.

The first thing you’ll notice about Marcie is that she looks
funny, and not just because this photo dates back to the
’80s. Cineon files are encoded in something called log
Also included on the book’s disc
color space. To make Marcie look more natural, open the is a Cineon sequence from the
Interpret Footage dialog, select the Color Management RED Camera (and courtesy of
tab, click Cineon Settings and choose the Over Range fxphd.com), showing off that digital
preset (instead of the default Full Range). Ah, that looks camera’s dynamic range and overall
image quality. This is one example
better; the log image is now converted to the monitor’s of Cineon format that is remaining
color space. viable with digital source.
It would seem natural to convert Cineon files to the moni-
tor’s color space, work normally, and then convert the end
result back to log, but to do so would be to throw away
valuable data. Try this: Apply the Cineon Converter effect
and switch the Conversion Type from Linear to Log. This
is a preview of how the file would be written on output
back to a Cineon log file. Upon further examination of
this conversion, you see a problem: in an 8 bpc (or even
16 bpc) project, the bright details in Marcie’s hair don’t
survive the trip (Figure 11.10).

361


Figure 11.10 When you convert an image from log space (left) to linear (center) and then back to log (right), the brightest
details are lost.

What’s going on with this mystical Cineon ﬁle and its log
color space that makes it so hard to deal with? And more
importantly, why? Well, it turns out that the engineers at
As becomes evident later in the Kodak know a thing or two about ﬁlm and have made no
chapter, the choice of the term “lin- decisions lightly. But to properly answer the question, it’s
ear” as an alternative to “log” space necessary to discuss some basic principles of photography
for Cineon Converter is unfortunate, and light.
because “linear” specifically means
neutral 1.0 gamma; what Cineon
Converter calls “linear” is in fact Dynamic Range
gamma encoded. The pictures shown in Figure 11.11 were taken in sequence
from a roof on a winter morning. Anyone who has ever
tried to photograph a sunrise or sunset with a digital cam-
era should immediately recognize the problem at hand.
With a standard exposure, the sky comes in beautifully, but
foreground houses are nearly black. Using longer expo-
sures you can bring the houses up, but by the time they are
looking good the sky is completely blown out.

Figure 11.11 Different exposures when recording the same scene clearly produce widely varying results.

362


The limiting factor here is the digital camera’s small
dynamic range, which is the difference between the bright-
est and darkest things that can be captured in the same
image. An outdoor scene has a wide array of brightnesses,
but any digital device can read only a slice of them. You
can change exposure to capture different ranges, but the
size of the slice is fixed.
Our eyes have a much larger dynamic range and our brains
have a wide array of perceptual tricks, so in real life the
houses and sky are both seen easily. But even eyes have
limits, such as when you try to see someone behind a bright
spotlight or use a laptop computer in the sun. The spotlight
has not made the person behind any darker, but when eyes
adjust to bright lights (as they must to avoid injury), dark
things fall out of range and simply appear black.
White on a monitor just isn’t very bright, which is one
reason we work in dim rooms with the blinds pulled
down. When you try to represent the bright sky on a dim
monitor, everything else in the image has to scale down in
proportion. Even when a digital camera can capture extra
dynamic range, your monitor must compress it in order to
display it.
A standard 8-bit computer image uses values 0 to 255 to
represent RGB pixels. If you record a value above 255—say
285 or 310—that represents a pixel beyond the monitor’s
dynamic range, brighter than white or overbright. Because
8-bit pixels can’t actually go above 255, overbright infor-
mation is stored as floating point decimals where 0.0 is
black and 1.0 is white. Because floating point numbers are
virtually unbounded, 0.75, 7.5, or 750.0 are all acceptable
values, even though everything above 1.0 will clip to white
on the monitor (Figure 11.12).
In recent years, techniques have emerged to create HDR Figure 11.12 Monitor white represents
the upper limit for 8-bit and 16-bit
images from a series of exposures—floating point files that pixels, while floating point can go arbi-
contain all light information from a scene (Figure 11.13). trarily higher (depending on format) or
The best-known paper on the subject was published by lower; the range also extends below
absolute black, 0.0—values that are
Malik and Debevec at SIGGRAPH ’97 (www.debevec.org
theoretical and not part of the world
has details). In successive exposures, values that remain you see (unless you’re in outer space,
within range can be compared to describe how the camera staring into a black hole).

363


is responding to different levels of light. That information
allows a computer to connect bright areas in the scene to
the darker ones and calculate accurate floating point pixel
values that combine detail from each exposure.

Darker Sky: 1.9

Bright Sky: 7.5

Dark Tree: 0.03

Houses: 0.8

Figure 11.13 Consider the floating point pixel values for this HDR image; they relate to one
another proportionally, and continue to do so whether the image is brightened or darkened,
because the values do not need to clip at 1.0.

But with all the excitement surrounding HDR imaging
and improvements in the dynamic range of video cam-
eras, many forget that for decades there has been another
medium available for capturing dynamic range far beyond
Photoshop’s Merge to HDR feature
allows you to create your own HDR what a computer monitor can display or a digital camera
images from a series of locked-off can capture.
photos at varied exposures.
That medium is film.

Cineon Log Space
A film negative gets its name because areas exposed to
light ultimately become dark and opaque, and areas
unexposed are made transparent during developing. Light
makes dark. Hence, negative.

364


Dark is a relative term here. A white piece of paper makes
a nice dark splotch on the negative, but a lightbulb dark-
ens the film even more, and a photograph of the sun
causes the negative to turn out darker still. By not com-
pletely exposing to even bright lights, the negative is able
to capture the differences between bright highlights and
really bright highlights. Film, the original image capture
medium, has always been high dynamic range.
If you were to graph the increase in film “density” as
increasing amounts of light expose it, you’d get something
like Figure 11.14. In math, this is referred to as a logarith-
mic curve. I’ll get back to this in a moment.

Figure 11.14 Graph the darkening (density) of film as increasing amounts of
light expose it and you get a logarithmic curve.

Digital Film
If a monitor’s maximum brightness is considered to be 1.0,
the brightest value film can represent is officially consid-
ered by Kodak to be 13.53 (although using the more effi-
cient ICC color conversion, outlined later in the chapter,
reveals brightness values above 70). Note this only applies
to a film negative that is exposed by light in the world as
opposed to a film positive, which is limited by the bright-
ness of a projector bulb, and is therefore not really consid-
ered high dynamic range. A Telecine captures the entire
range of each frame and stores the frames as a sequence of
10-bit Cineon files. Those extra two bits mean that Cineon
pixel values can range from 0 to 1023 instead of the 0 to
255 in 8-bit files.

365


Having four times as many values to work with in a Cineon
file helps, but considering you have 13.53 times the range
to record, care must be taken in encoding those values.
The most obvious way to store all that light would simply
be to evenly squeeze 0.0 to 13.53 into the 0 to 1023 range.
The problem with this solution is that it would only leave
75 code values for the all-important 0.0 to 1.0 range, the
same as allocated to the range 10.0 to 11.0, which you are
far less interested in representing with much accuracy.
Your eye can barely tell the difference between two high-
lights that bright—it certainly doesn’t need 75 brightness
variations between them.
A proper way to encode light on film would quickly fill up
the usable values with the most important 0.0 to 1.0 light
and then leave space left over for the rest of the negative’s
range. Fortunately, the film negative itself with its logarith-
mic response behaves just this way.
Cineon files are often said to be stored in log color space.
Actually it is the negative that uses a log response curve
and the file is simply storing the negative’s density at each
pixel. In any case, the graph in Figure 11.15 describes how
light exposes a negative and is encoded into Cineon color
values according to Kodak, creators of the format.

1023
10-bit Cineon Code Values

685 (white)

95 (black)

0
0.0 1.0 2.0 4.0 6.0 8.0 10.0 12.0 13.53

Figure 11.15 Kodak’s Cineon log encoding is also expressed as a logarithmic
curve, with labels for the visible black and white points that correspond to
0 and 255 in normal 8-bit pixel values.

366


One strange feature in this graph is that black is mapped
to code value 95 instead of 0. Not only does the Cineon
file store whiter-than-white (overbright) values, it also has
some blacker-than-black information. This is mirrored in
the film lab when a negative is printed brighter than usual
and the blacker-than-black information can reveal itself.
Likewise, negatives can be printed darker and take advan-
tage of overbright detail. The standard value mapped to
monitor white is 685, and everything above is considered All About Log
overbright. You may first have heard of logarithmic curves in
high school physics class, if you ever learned about
Although the Kodak formulas are commonly used to the decay of radioactive isotopes.
transform log images for compositing, other methods have
emerged. The idea of having light values below 0.0 is dubi- If a radioactive material has a half-life of one year,
ous at best, and many take issue with the idea that a single half of it will have decayed after that time. The
curve can describe all film stocks, cameras, and shooting next year, half of what remains will decay, leaving
environments. As a different approach, some visual effects a quarter, and so on. To calculate how much time
has elapsed based on how much material remains,
facilities take care to photograph well-defined photo- a logarithmic function is used.
graphic charts and use the resultant film to build custom
curves that differ subtly from the standard Kodak one.
Light, another type of radiation, has a similar effect
As much as Cineon log is a great way to encode light on film. At the molecular level, light causes silver
captured by film, it should not be used for compositing or halide crystals to react. If film exposed for some
short period of time causes half the crystals to
other image transformations. This point is so important
react, repeating the exposure will cause half of the
that it just has to be emphasized again: remaining to react, and so on. This is how film gets
its response curve and the ability to capture even
Encoding color spaces are not compositing color spaces.
very bright light sources. No amount of exposure
To illustrate this point, imagine you had a black pixel with can be expected to affect every single crystal.
Cineon value 95 next to an extremely bright pixel with
Cineon’s highest code value, 1023. If these two pixels were
blended together (say, if the image was being blurred), the
result would be 559, which is somewhere around middle
gray (0.37 to be precise). But when you consider that the
extremely bright pixel has a relative brightness of 13.5, that
black pixel should only have been able to bring it down
to 6.75, which is still overbright white! Log space’s extra
emphasis on darker values causes standard image process-
ing operations to give them extra weight, leading to an
overall unpleasant and inaccurate darkening of the image.
So, final warning: If you’re working with a log source, don’t
do image processing in log space!

367


Video Gamma Space
Because log space certainly doesn’t look natural, it proba-
bly comes as no surprise that it is a bad color space to work
in. But there is another encoding color space that you have
The description of gamma in video
is oversimplified here somewhat been intimately familiar with for your entire computer-
because the subject is complex using life and no doubt have worked in directly: the video
enough for a book of its own. An space of your monitor.
excellent one is Charles Poynton’s
Digital Video and HDTV Algorithms You may have always assumed that 8-bit monitor code value
and Interfaces (Morgan Kaufmann). 128, halfway between black and white, makes a gray that is
half as bright as white. If so, you may be shocked to hear
that this is not the case. In fact, 128 is much darker—not
even a quarter of white’s brightness on most monitors.
A system where half the input gives you half the output is
described as linear, but monitors (like many things in the
real world) are nonlinear. When a system is nonlinear, you
can usually describe its behavior using the gamma func-
tion, shown in Figure 11.16 and the equation
Output = inputgamma 0 <= input <= 1

In this function, the darkest and brightest values (0.0 and
1.0) are always ﬁxed, and the gamma value determines
how the transition between them behaves. Successive
applications of gamma can be concatenated by multiplying
them together. Applying gamma and then 1/gamma has
the net result of doing nothing. Gamma 1.0 is linear.

Figure 11.16 Graph of monitor 1
gamma (2.2) with file gamma (0.4545)
and linear (1.0). These are the color
45
curves in question, with 0.4545 and .45
m a0
2.2 each acting as the direct inverse of gam
the other. .0
a1
m
am
.2
a2

g
mm
ga

0
0 1

Mac monitors have traditionally had a gamma of 1.8, while
the gamma value for PCs is 2.2. Because the electronics
in your screen are slow to react from lower levels of input

368


voltage, a 1.0 gamma is simply too dark in either case;
boosting this value compensates correctly.
The reason digital images do not appear dark, however,
is that they have all been created with the inverse gamma
Gamma-rama
function baked in to prebrighten pixels before they are
In case all this gamma talk hasn’t already blown
displayed (Figure 11.17). Yes, all of them. your mind, allow me to clarify how monitor
gamma and human vision work together. The
question often comes up—why is middle gray
18% and not 50%? And why does 50% gray look
like middle gray on my monitor, but not on a linear
color chart?

It turns out that your eyes also have a nonlinear
response to color—your vision brightens low light,
which helps you to see where it’s dim, a survival
advantage. The human eye is very sensitive to
small amounts of light, and it gets less sensitive as
brightness increases. Your eye effectively brightens
the levels, and objects in the world are, in fact
darker than they appear—or, they become darker
when we represent their true linear nature.

The fact that 18% gray (or somewhere around that
level; there is some disagreement about the exact
number) looks like middle gray to your eye tells
us that the eye does its own gamma correction of
roughly 0.36: 50% x 0.36 = 18%.
Figure 11.17 The gamma settings in the file and monitor complement one
another to result in faithful image reproduction.

Because encoding spaces are not compositing spaces,
working directly with images that appear on your moni-
tor can pose problems. Similar to log encoding, video
gamma encoding allocates more values to dark pixels, so
they weigh more than they should. Video color space is not
much more valid than Cineon color space for recreating
the way light behaves in the world at large.

Linear Floating Point HDR
In the real world, light behaves linearly. Turn on two
lightbulbs of equivalent wattage where you previously had
one and the entire scene becomes exactly twice as bright.

369


A linear color space lets you simulate this effect simply by
doubling pixel values. Because this re-creates the color
space of the original scene, linear pixels are often referred
to as scene-referred values, and doubling them in this man-
ner can easily send values beyond monitor range.
The Exposure effect in After Effects converts the image
to which it is applied to linear color before doing its work
unless you specifically tell it not to do so by checking
Bypass Linear Light Conversion. It internally applies a
.4545 gamma correction to the image (1 divided by 2.2,
inverting standard monitor gamma) before adjusting.
A common misconception is that if you work solely in the
domain of video you have no need for floating point. But
just because your input and output are restricted to the 0.0
To follow this discussion, choose
Decimal in the Info panel menu to 1.0 range doesn’t mean that overbright values above 1.0
(this is the default for 32 bpc). 0.0 won’t figure into the images you create. The 11_sunrise.aep
to 1.0 values are those falling in project included on your disc shows how they can add to
Low Dynamic Range, or LDR— your scene even when created on the fly.
those values typically described in
8 bit as 0 to 255. Any values outside The examples in Table 11.1 show the difference between
this range are HDR, 32 bpc only. making adjustments to digital camera photos in their
native video space and performing those same operations
in linear space. In all cases, an unaltered photograph
featuring the equivalent in-camera effect is shown for
comparison.
The table’s first column contains the images brightened
by one stop, an increment on a camera’s aperture, which
controls how much light is allowed through the lens. Wid-
ening the aperture by one stop allows twice as much light
to enter. An increase of three stops brightens the image by
a factor of eight (2 × 2 × 2, or 23).
To double pixel values in video space is to quickly blow out
bright areas in the image. Video pixels are already encoded
with extra brightness and can’t take much more.
The curtain and computer screen lose detail in video space
that is retained in linear space. The linear image is nearly
indistinguishable from the actual photo for which cam-
era exposure time was doubled (another practical way to
brighten by one stop).

370


TABLE 11.1 Comparison of Adjustments in Native Video Space and in Linear Space
BRIGHTEN ONE STOP LENS DEFOCUS MOTION BLUR
Original Image

Filtered in
Video Space

Filtered in
Linear Space

Real-World
Photo

The second column simulates an out-of-focus scene using
Fast Blur. You may be surprised to see an overall darken-
ing with bright highlights fading into the background—at
least in video space. In linear, the highlights pop much
better. See how the little man in the Walk sign stays bright

371


in linear but almost fades away in video because of the
extra emphasis given to dark pixels in video space. Squint
your eyes and you notice that only the video image darkens
overall. Because a defocused lens doesn’t cause any less
light to enter it, regular 8 bpc blur does not behave like a
true defocus.
The table’s third column uses After Effects’ built-in motion
blur to simulate the streaking caused by quick panning as
the photo was taken. Pay particular attention to the high-
light on the lamp; notice how it leaves a long, bright streak
in the linear and in-camera examples. Artiﬁcial dulling of
highlights is the most obvious giveaway of nonlinear image
processing.
Artists have dealt with the problems of working directly
in video space for years without even knowing we’re com-
pensating all the time. A perfect example is the Screen
transfer mode, which is additive in nature but whose cal-
culations are clearly convoluted when compared with the
pure Add transfer mode. Screen uses a multiply-toward-
white function with the advantage of avoiding the clipping
associated with Add. But Add’s reputation comes from
its application in bright video-space images. Screen was
invented only to help people be productive when working
in video space, without overbrights; Screen darkens over-
brights (Figure 11.18). Real light doesn’t Screen, it Adds.
Add is the new Screen, Multiply is the new Hard Light, and
many other blending modes fall away completely in linear
ﬂoating point.

Figure 11.18 Watch those highlights. Adding in video space blows out (left), but Screen in video looks better (center). Add-
ing in linear is best (right).

372


HDR Source and Linearized Working Space
Should you in fact be fortunate enough to have high-bit-
depth source images with over-range values, there are
indisputable benefits to working in 32-bit linear, even if
Linear blending is available
your final output uses a plain old video format that cannot without 32 bpc HDR; in Project Set-
accommodate these values. tings, choose Blend Colors using 1.0
Gamma. This feature is described in
In Figure 11.19, the lights are severely clipped by video detail near the end of the chapter.
space, which is not a problem so long as the image is only
displayed; all of the images look fine printed on this page
or displayed on your monitor. Add motion blur, however,
and you see the problem at its most exaggerated; the
points of light should not lose their intensity simply by
being put into motion.

Figure 11.19 An HDR image is blurred without floating point (left) and with floating point (center), before being shown as
low dynamic range (right).

The benefits of floating point aren’t restricted to blurs,
however; they just happen to be an easy place to see the
difference. Every operation in a compositing pipeline
gains extra realism from the presence of floating point
pixels and linear blending.
Terminology
Figure 11.20 features an HDR image on which a simple Linear floating point HDR compositing uses radio-
composite is performed, once in video space and once metrically linear, or scene-referred, color data. For
using linear floating point. In the floating point version, the purposes of this discussion, this is perhaps best
called “linear light compositing,” or “linear floating
the dark translucent layer acts like sunglasses on the bright point,” or just simply, “linear.” The alternative mode
window, revealing extra detail exactly as a filter on a camera to which you are accustomed is “gamma-encoded,”
lens would. The soft edges of a motion-blurred object also or “monitor color space,” or simply, “video.”
behave realistically as bright highlights push through. With-
out floating point there is no extra information to reveal, so

373


Figure 11.20 A source image with
over-range values in the highlights
(top) is composited without floating
point (bottom left) and with floating
point (bottom right).

the window looks clipped and dull and motion blur doesn’t
interact with the scene properly.

32 Bits per Channel
Although it is not necessary to use HDR source to take
advantage of an HDR pipeline, it offers a clear glimpse
of this brave new world. Open 11_treeHDR_lin.aep; it
Included on the disc are two
similar images, sanityCheck.exr and contains a comp made up of a single image in 32 bit EXR
sanityCheck.tif. The 32 bpc EXR file format (used to create Figure 11.19). With the Info panel
is linearized, but the 8 bpc TIFF file clearly visible, move your cursor around the frame.
is not. Two corresponding projects
are also included, one using no color As your cursor crosses highlights—the lights on the tree,
profile, the other employing a linear specular highlights on the wall and chair, and most espe-
profile. These should help illustrate cially, in the window—the values are seen to be well above
the different appearances of a linear
and a gamma-encoded image. 1.0, the maximum value you will ever see doing the same
in 8 bpc or 16 bpc mode. Remember that you can quickly
toggle between color spaces by Alt/Option-clicking
the project color depth identiﬁer at the bottom of the
Project panel.

374


Any experienced digital artist would assume that there is
no detail in that window—it is blown out to solid white
forevermore in LDR. However, you may have noticed an
extra icon and accompanying numerical value that appears
at the bottom of the composition panel in a 32 bpc proj-
ect (Figure 11.21). This is the Exposure control; its icon
looks like a camera aperture and it performs an analogous
function—controlling the exposure (total amount of light)
of a scene the way you would stop a camera up or down (by
adjusting its aperture). Figure 11.21 Exposure is an HDR
preview control that appears in the
Drag to the left on the numerical text and something
Composition panel in 32 bpc mode.
amazing happens. Not only does the lighting in the scene
decrease naturally, as if the light itself were being brought
down, but at somewhere around -10.0, a gentle blue gradi-
ent appears in the window (Figure 11.22, left).

Figure 11.22 At –10 Exposure, the room is dark other than the tree lights and detail becomes visible out the window. At +3,
the effect is exactly that of a camera that was open 3 stops brighter than the unadjusted image.

Drag the other direction, into positive Exposure range,
and the scene begins to look like an overexposed photo;
the light proportions remain and the highlights bloom
outward (Figure 11.22, right).
Keep in mind that for each 1.0
The Exposure control in the Composition panel is a adjustment upward or downward
preview-only control (there is an effect by the same name of Exposure, you double (or halve)
the light levels in the scene.
that renders); scan with your cursor, and Info panel values Echoing the earlier discussion, a
do not vary according to its setting. This control offers +3.0 Exposure setting sets the light
a quick way to check what is happening in the out-of- levels 8× (or 23) brighter.
range areas of a composition. With a linear light image,

375


each integer increment represents the equivalent of one
photographic stop, or a doubling (or halving) of linear
light value.

Mixed Bit Depths and Compander
Floating Point Files Most effects don’t, alas, support 32 bpc, although there
As you’ve already seen, there is one class of files are dozens that do. Apply a 16 bpc or (shudder) 8 bpc
that does not need to be converted to linear space:
effect, however, and the overbrights in your 32 bpc project
floating point files. These files are already storing
scene-referred values, complete with overbright disappear—all clipped to 1.0. Any effect will reduce the
information. Common formats supported by After image being piped through it to its own color space limita-
Effects are Radiance (.hdr) and floating point .tiff, tions. A small warning sign appears next to the effect to
not to mention .psd, but the most universal and remind you that it does not support the current bit depth.
versatile is Industrial Light + Magic’s OpenEXR
format. OpenEXR uses efficient 16-bit floating You may even see a warning explaining the dangers of
point pixels, can store any number of image chan- applying this effect.
nels, supports lossless compression, and is already
supported by most 3D programs thanks to being an Of course, this doesn’t mean you need to avoid these
open source format. effects to work in 32 bpc. It may mean you have to cheat,
and After Effects includes a preset allowing you to do just
After Effects CS4 offers expanded support of the
that: Compress-Expand Dynamic Range (contained in
OpenEXR format by bundling plug-ins from fnord Effects & Presets > Animation Presets > Image – Utilities.
software, which provide access to multiple layers Make certain Show Animation Presets is checked in the
and channels within these files. EXtractoR can panel menu).
open any floating point image channel in an EXR
file, and Identifier can open the other, non-image This preset actually consists of two instances of the HDR
channels such as Object and Material ID. Compander effect, which was specifically designed to
bring floating point values back into LDR range. The first
instance is automatically renamed Compress, and the sec-
ond, Expand, which is how the corresponding modes are
set. You set the Gain of Compress to whatever is the bright-
est overbright value you wish to preserve, up to 100. The
values are then compressed into LDR range, allowing you
to apply your LDR effect. The Gain (as well as Gamma) of
Expand is linked via an expression to Compress so that the
values round-trip back to HDR (Figure 11.23).
Additionally, there are smart ways to set up a project to
8 and 16 bpc effects clip only the ensure that Compander plays the minimal possible role. As
image that they process. Apply one much as possible, group all of your LDR effects together,
to a layer, you clip that layer only. and keep them away from the layers that use blending
Apply one to an Adjustment Layer
and you clip everything below. Add
modes where float values are most essential. For example,
an LDR comp or layer to an other- apply an LDR effect via a separate adjustment layer instead
wise HDR comp in a 32 bpc project, of directly on a layer with a blending mode. Also, if pos-
it remains otherwise HDR. sible, apply the LDR effects first, then boost the result

376


into HDR range to apply any additional 32 bpc effects and
blending modes.

Blend Colors Using 1.0 Gamma
After Effects CS4 contains a fantastic option to linearize
image data only when performing blending operations:
the Blend Colors Using 1.0 Gamma toggle in Project Set-
tings. This allows you to take advantage of linear blending,
which makes Add and Multiply blending modes actually
work properly, even in 8 bpc or 16 bpc modes. Figure 11.23 Effects & Presets > Animation Presets >
Presets > Image - Utilities includes the Compress-
The difference is quite simple. A linearized working space Expand Dynamic Range preset, also known as the
does all image processing in gamma 1.0, as follows Compander (Compressor/Expander). It consists of
two instances of the HDR Compander effect, one
footage --> to linear PWS -> named Compress and the other, Expand, wired with
Layer -> an expression to undo whatever Compress does.
The idea is to place your 8 or 16 bpc effect (such as
Mask -> Effects -> Transform ->
Match Grain, shown) between the two of them. This
Blend With Comp -> prevents clipping of over-range values. If banding
Comp -> from linear PWS to OM space -> appears as a result of Compress-Expand, Gamma
output can be adjusted to weight the compressed image
more toward the region of the image (probably
whereas linearized blending performs only the blending the shadows) where the banding occurs. You are
sacrificing image fidelity in order to preserve a
step, where the image is combined with the composition, compressed version of the HDR pipeline.
in gamma 1.0
footage --> to PWS ->
Layer ->
Mask -> Effects -> Transform -> to linear PWS ->
Blend With Comp -> to PWS ->
Comp -> from PWS to OM space ->
output

(Special thanks to Dan Wilk at Adobe for detailing this out.)
Because effects aren’t added in linear color, blurs no
longer interact correctly with overbrights (although they
do composite more nicely), and you don’t get the subtle
benefits to Transform operations; After Effects’ much
maligned scaling operations are much improved in linear
floating point. Also, 3D lights behave more like actual
lights in a fully linearized working space.
I prefer the linear blending option when in lower bit
depths and there is no need to manage over-range values;
it gives me the huge benefit of more elegant composites

377


and blending modes without forcing me to think about
managing effects in linear color. Certain key effects, in par-
ticular Exposure, helpfully operate in linear gamma mode.

Output
Finally, what good is it working in linear floating point if
the output bears no resemblance to what you see in the
composition viewer? Just because you work in 32-bit float-
ing point color does not mean you have to render your
images that way.
Keeping in mind that each working space can be linear or
not, if you work in a linearized color space and then ren-
der to a format that is typically gamma encoded (as most
are), the gamma-encoded version of the working space will
also be used. After Effects spells this out for you explicitly
in the Description section of the Color Management tab.
To this day, the standard method to pass around footage
with over-range values, particularly if it is being sent for
film-out, is to use 10-bit log-encoded Cineon/DPX. This is
also converted for you from 32 bpc linear, but be sure to
choose the Working Space as the output profile and that in
Cineon Settings, you use the Standard preset.
The great thing about Cineon/DPX with a standard 10-bit
profile is that it is a universal standard. Facilities around
the world know what to do with it even if they’ve never
encountered a file with an embedded color profile. As was
detailed earlier in the chapter, it is capable of taking full
advantage of the dynamic range of film, which is to this day
the most dynamic display medium widely available.

Conclusion
This chapter concludes Section II, which focused on the
most fundamental techniques of effects compositing. In
the next and final section, you’ll apply those techniques.
You’ll also learn about the importance of observation, as
well as some specialized tips and tricks for specific effects
compositing situations that re-create particular environ-
ments, settings, conditions, and natural phenomena.

378

III
Creative Explorations

Chapter 12 Light 381
Chapter 13 Climate and the Environment 405
Chapter 14 Pyrotechnics: Heat, Fire, Explosions 429

There are two kinds of light: the glow that illuminates
and the glare that obscures.
—James Thurber

Light

T here’s more to understanding how light works in the
world than knowing the science of it, although that is cer-
tainly essential. The work of a compositor is much like that
of a painter or cinematographer, in that a combination of
technical knowledge, interpretation, and even intuition are
all part of getting a scene “right.”
Other areas of digital production rely on elaborate models
to simulate the way light works in the physical world. Like
a painter, the compositor observes the play of light in the
three-dimensional world to re-create it two-dimensionally.
Like a cinematographer, you succeed with a feeling for
how lighting and color decisions affect the beauty and
drama of a scene and how the camera gathers them.
Several chapters in this book touch upon principles of
the behavior of light. Chapter 5 was about the bread and
butter work of the compositor—matching brightness and
color of a foreground and background. Chapter 9 was all
about how the world looks through a lens. Chapter 11
explored less straightforward ways in which After Effects
can re-create the way color and light values behave.
This chapter is dedicated to practical situations involving
light that you as a compositor must re-create. It’s important
to distinguish lighting conditions you can easily emulate
and those that are essentially out of bounds—although, for
a compositor with a good eye and patience, the seemingly
“impossible” becomes a welcome challenge and the source
of a favorite war story.

Source and Direction
In many scenes, there is clearly more involved with light
than matching brightness and contrast channel by channel.

382

III: Creative Explorations

Light direction is fundamental, especially where the quality
of the light is hard (direct) rather than soft (diffuse).
Such a huge variety of light situations are possible in a
shot, and in an infinite array of combinations, that it
becomes difficult to make any broad statements stand up
about lighting. This section, however, tries to pin down
some general guidelines and workflows for manipulating
the light situation of your scene.

Location and Quality
You may have specific information about the lighting
conditions that existed when source footage was shot. On a
set, you can easily identify the placement and type of each
light, and away from set, this information may be found in a
camera report or on-set photos. For a naturally lit shot, it’s
mostly a question of the position of the sun relative to the
camera and the reflectivity of the surrounding environment.
Sometimes the location and direction of light is readily
apparent, but not as often as you might think. Hard, direct
light casts clear shadows and raises contrast, and soft, dif-
fuse light lowers contrast and casts soft shadows (if any).
That much seems clear.
These, however, are broad stereotypes, which do not always
behave as expected in the real world. Hard light aimed
directly at a subject from the same direction as the camera
actually flattens out detail, effectively decreasing contrast.
And artificial lighting is usually multiple sources in a single
scene, which work against one another to diffuse hard
shadows (Figure 12.1).

Figure 12.1 Interior sets, like interior
environments, are typically lit by more
than one source, creating multiple soft
highlights and shadows.

383

Chapter 12 Light

Neutralize Direction and Hotspots
When the direction or diffusion of light on a foreground
element doesn’t match the target background environ-
ment, that’s potentially a big problem. The solution is
generally to neutralize the mismatch by isolating and mini-
mizing it, rather than actually trying to ﬁx the discrepancy
by attempting to simulate relighting the element in 2D.
Every shot in the world has unique light characteristics, but
a couple of overall strategies apply. Once you’ve exhausted
simple solutions such as ﬂopping the shot (where lighting
is on the wrong side), you can
. Isolate and remove directional clues around the ele-
ment, such as cast shadows (typically by matting or
rotoscoping them out).
. Isolate and reduce contrast of highlights and shad-
ows in the element itself, typically with a Levels or
Curves adjustment (potentially aided by a luma matte,
described later in this chapter).
. Invert the highlights and shadows with a counter-
gradient.
The simple way to undo evidence of too strong a key light
in a scene is to create a counter-gradient as a track matte
for an adjustment layer; a Levels or Curves effect on this
layer affects the image proportionally to this gradient. The
Ramp effect can be set and even animated to the position
of a key light hotspot (Figure 12.2).

Figure 12.2 Counter-gradients (this one created with the Ramp effect) can serve an adjustment layer used to lower
the brightness and contrast in the hotspot region.

384


A radial ramp is merely linear, which is not the correct
model for light falloff. Light’s intensity diminishes propor-
tionally to its distance from the source squared, according
to the inverse square law. An object positioned twice as far
from a single light source is illuminated by one-quarter the
amount of light. To mimic this with a gradient, precomp it,
duplicate the radial gradient layer, and set the upper of the
two layers to a Multiply blending mode (Figure 12.3).

Figure 12.3 A standard Ramp
gradient (top left) is linear, as can be
seen in the Histogram, but light falls
off in a logarithmic, inverse-square
pattern, so the matte used in Figure
12.2 multiplies together two linear
gradients (bottom left) with Linear
blending enabled (bottom right) in
Project Settings even though it’s not a
32 bpc linear HDR project. Again, light
works in linear.

Of course, you don’t want to ﬁght the fundamental source
lighting in this way unless you absolutely have to do so;
hopefully you will rarely have to “ﬁx” lighting and will most
often want to work with what you’ve got to make it even
stronger and more dramatic.

385

Chapter 12 Light

Color Looks
Have you ever seen unadjusted source clips or behind-
the-scenes footage from a favorite movie? It’s a striking
reminder about the bold and deliberate use of color in
modern films. Look at the work prints or on-set making-of
video—the magic more or less disappears.
In older films color looks had to be accomplished optically
and photochemically. Strong color filters would be used in
the camera’s matte box to reduce or enhance the effect of
the source lighting and color. The well-known bleach-bypass
method would be used to strip certain colors out in the film
lab. Nowadays, a digital production pipeline has made the
photochemical approach rarer, although optical filters still
play a large role in shooting. Meanwhile, it’s becoming
more and more common for an entire feature-length pro-
duction to be graded through a digital intermediate, or D.I.
After Effects has an advantage over D.I. software such as a
DaVinci system in that it is a true compositing system, with
fine controls over image selection. After Effects was not
created principally with the colorist in mind, so its primary
color tools (as described in Chapter 5) are simpler and
less interactive. Third-party solutions such as Colorista and
Magic Bullet Looks, both from Red Giant, have appeared
to help bridge this gap.
Keeping in mind that your job as a compositor is to emu-
late the world as it looks when viewed with a camera, it can
be effective to begin by emulating physical lens elements.

The Virtual Lens Filter
Suppose a shot (or some portion of it) should simply be
“warmer” or “cooler.” With only a camera and some film,
you might accomplish this transformation by adding a lens
filter. It could be a solid color (blue for cooler, amber to
warm things up) or a gradient (amber to white to change
only the color of a sky above the horizon).
Add a colored solid and set its blending mode to Color.
Choose a color that is pleasing to your eye, with brightness
and saturation well above 50%. Use blue or green for a
cooler look, red or yellow for a warmer one (Figure 12.4).

386


Figure 12.4 Here, the four color filters are applied as a test with a Color blending mode, and with the Linear mode, so that
they behave a lot like lens filters of an equivalent color.

At 100%, this is the equivalent of a full-color tint of the image,
which is too much. Dial Opacity down between 10% and
50%, seeking the threshold where the source colors remain
discernable, filtered by the added color to set the look.
To re-create a graded filter, typically used to affect only the
sky, apply the Ramp effect to the solid and change the Start
Color to your tint color; an amber filter adds the look of an
extremely smoggy urban day. This is best applied with an
Add or Screen mode instead of Color because the default
End Color, white, desaturates the lower part of the image.

Black and White
When removing color from an element entirely, there is a
huge difference between using the Hue/Saturation effect
and the likely alternatives. Counterintuitively, Hue/Satura-
The flag of Mars is a red, green, and
tion is typically not the best choice to create a black-and- blue tricolor selected by the Mars
white image because it maintains luminance proportions, Society and flown into orbit by the
and as was mentioned in a sidebar back in Chapter 6, Space Shuttle Discovery. It was
that’s not how the eye sees color. Figure 12.5 illustrates the not used by Marvin the Martian to
claim Planet X.
difference.

Figures 12.5 This is the flag of Mars (left): it shows three fields of pure red, green, and blue (no earthly location was logical
enough to choose this). Tint (center) compensates for the perceptual differences in human color vision when desaturating,
Hue/Saturation (right) does not.

387

Chapter 12 Light

Even if converting an image to black and white is only an
intermediate step, you’re best off doing so either using
the Tint effect at the default settings or a fully desaturated
solid (black, white, or gray, it doesn’t matter) with a Color
blending mode. To really get the conversion right may
involve adjusting or shifting color channels prior to the
color-to-black-and-white conversion (Figure 12.6).

Figure 12.6 A real color to grayscale conversion may involve carefully rebalancing color, contrast, or saturation. Here, the
face and lamp are important and get individual adjustments in color prior to conversion. (Images courtesy of 4charros.)

Day for Night
Stronger optical effects are even possible, such as making a
daytime scene appear as if it were shot on a moonlit night.
Many images benefit from a subtle
reduction in overall Saturation Known in French as la nuit américaine (and immortalized in
using the Hue/Saturation tool. This Francois Truffaut’s ode to filmmaking of the same name),
moves red, green, and blue closer this involves a simple trick. Shoot an exterior scene under
together and can reduce the “juicy”
ordinary daylight with a dark blue lens filter to compen-
quality that makes bright colored
images hard to look at. sate for the difficulty of successful low light night shoots. If
there is direct sunlight, it’s meant to read as moonlight.
Lighting techniques and film itself have improved since
this was a common convention of films, particularly

388


westerns, but digital cameras tend to become noisy and
muddy under low light.

Figure 12.7 An ordinary twilight shot of a house at dusk becomes a spooky Halloween mansion. Converting day for night
avoids the problems associated with low-light shooting. (Images courtesy of Mars Productions.)

Figure 12.7 shows the difference between a source image
that is blue and desaturated and an actual night look; if
instead you’re starting with a daylight image, look at the
images on the book’s disc, which take the image more
in that direction. Overall, remember that the eye cannot
see color without light, so only areas that are perceived to
be well illuminated should have a hue outside the range
between deep blue and black.

Color Timing Effects
Digital tools can of course go far beyond what is possible
with lens ﬁlters. The industry standard tools rely on a
three-way color corrector, which allows you to tint the image
in three basic luminance ranges—highlights, midtones,
and shadows—adjusting each separately via wheels which
control hue and brightness.
Colorista from Red Giant is a three-way color corrector
that opens the door to coloring footage the way it would
be done with a feature ﬁlm D.I. grade. Third-party plug-ins
are generally considered beyond the scope of this book,
and colorist work based on three-way color correction has
merited whole books and courses of its own, but this plug-
in opens the door to grading primary and secondary colors
like a pro.
A primary grade might involve injecting unexpected hues
into each range: blue shadows, pink highlights, green

389

Chapter 12 Light

midtones. The secondary grade then isolates an area of the
scene—often faces or other dramatically key elements—
and determines how their color and contrast complements
the scene as a whole.
The grade is then applied across several shots in what is
traditionally called the color timing process—literally, mak-
ing color consistent across time, which typically involves
If you don’t like the idea of adding
a third-party plug-in to your work-
much more than simply applying the same adjustment to
flow, there is even a free alternative every shot. You can use the techniques described here and
wired to a Color Balance (HLS) in Section II to ﬁrst balance a shot, then add its color look,
effect included in The DV Rebel’s ﬁnally bringing out any key exceptional details; then voila,
Guide by Stu Maschwitz (Peachpit
you are a colorist.
Press) as RebelCC.

Source, Reflection, and Shadow
Sometimes you work with source footage that contains
strong lighting and thus offers a clear target. Other times,
it’s up to you to add a strong look. Either way, reference is
your friend. You will be surprised how much bolder and
more fascinating nature’s choices are than your own.
Unexpected surprises that simply “work” can be the kiss
of love for a scene—that something extra that nobody
requested but everyone who is paying attention appreci-
ates. Details of light and shadow are one area where this
extra effort can really pay off.
Big, bold, daring choices about light can and should
become almost invisible if they are appropriate to a scene,
adding to the dramatic quality of the shot instead of merely
showing off what you as an artist can do.

Backlighting and Light Wrap
The conditions of a backlit scene are a classic example
New in this edition of the book, the where the compositor often does not go far enough to
light wrap formula outlined below match what actually happens in the real world.
has been converted to a script cre-
ated by Jeff Almasol. You can find it This technique is designed for scenes that contain back-
on the book’s disc as rd_Lightwrap. lighting conditions and a foreground that, although it
Select the matted source layer and
let this script do the work.
may be lit to match those conditions, lacks light wrapping
around the edges (Figure 12.8).

390


Figure 12.8 The silhouetted figure is
color corrected to match, but lacks
any of the light wrap clearly visible
around the figures seated on the
beach.

Set up a light wrap effect as follows:
1. Create a new composition that contains the back-
ground and foreground layers, exactly as they are
positioned and animated in the master composition.
You can do this simply by duplicating the master comp
and renaming it something intuitive, such as Light
Wrap. If the foreground or background consists of
several layers, it will probably be simpler to precompose
them into two layers, one each for the foreground and
background.
2. Set Silhouette Alpha blending mode for the fore-
ground layer, punching a hole in the background.
3. Add an adjustment layer at the top, and apply Fast Blur.

4. In Fast Blur, check the Repeat Edge Pixels toggle on
and crank up the blurriness.
5. Duplicate the foreground layer, move the copy to the
top, and set its blending mode to Stencil Alpha, leaving
a halo of background color that matches the shape of
the foreground (Figure 12.9, top). If the light source is
not directly behind the subject, you can offset this layer
to match, producing more light on the matching side.
6. Place the resulting comp in the master comp and adjust
opacity (and optionally switch the blending mode to
Figure 12.9 The background is blurred into the
Add, Screen, or Lighten) until you have what you’re
matte area (top) in a precomp and added back
after. You may need to go back to the Light Wrap comp into the scene to better integrate the silhouette
to further adjust the blur (Figure 12.9, bottom). (bottom).

391

Chapter 12 Light

When there is no fill light on the foreground subject
whatsoever, most cameras are incapable of picking up as
much detail in the foreground as your eye might see. In
your reference photo, an unlit foreground subject might
What Causes a Lens Flare?
appear completely silhouetted. Because the foreground
Unlike your eye, which has only one very flexible subjects are often the stars of the scene, you might have to
lens, camera lenses are typically made up of a series
of inflexible lens elements. These elements are compensate, allowing enough light and detail in the fore-
coated to prevent light reflecting off of them under ground that the viewer can see facial expressions and other
normal circumstances. Extreme amounts of light, important dramatic detail.
however, are reflected somewhat by each element.
In other words, this might be a case where your reference
conflicts with what is needed for the story. Try to strike a
Zoom lenses contain many focusing elements and
tend to generate a complex-looking flare with lots
balance, but remember, when the story loses, nobody wins.
of individual reflections. Prime lenses generate
fewer. Flares
For our purposes a “flare” is any direct light source that
Many factors besides the lens elements contribute appears in shot, not just a cheesy 17-element lens flare
to the look of a flare. Aperture blades within the whenever the sun pokes around the moon in some science
lens cause reflective corners that often result in fiction television show from the early 1990s (or in several
streaks; the number of streaks corresponds to
the number of blades. The shape of the flares
shots from the more recent movie Hancock). These don’t
sometimes corresponds to the shape of the aperture come for free in After Effects; 3D lights don’t even create a
(a pentagon for a five-sided aperture, a hexagon for visible source if placed in shot until you add the Trapcode
six). Dust and scratches on the lens also reflect light. Lux effect (included on this book’s disc).
Real lens flares are never cheesy: Our eyes accept them as
FInally, lens flares look very different depending
natural, even beautiful artifacts without necessarily under-
on whether they were shot on film or video; excess
light bleeds out in different directions and patterns. standing anything about what actually causes them (Figure
12.10). Transformers, composited at ILM, is a recent exam-
ple of a visual effects movie full of great looking light optics.

Figure 12.10 You might think of a lens
flare as one of those element rings
caused by a zoom lens pointing at the
sun, but flares occur anytime a bright
light source appears in or near frame.
Here, the green traffic light causes a
large flare without even appearing
in frame, and numerous other lights
bloom and flare. (Image from Quality
of Life, courtesy of Benjamin Morgan.)

392


Therefore, to get lens flares or even simple glints right
(not cheesy), good reference is often key. Only a tiny per-
centage of your viewers may know the difference between
lens flares from a 50 mm prime and a 120 mm zoom lens, Prior to the 1970s-era of Easy
yet somehow, if you get it wrong, it reads as phony. Odd. Rider and moon shots, flares were
regarded as errors on the part of the
Here are some things you should know about lens flares: cinematographer, and shots con-
taining them were carefully noted
. They are consistent for a given lens. Their angles vary
on the camera report and retaken.
according to the position of the light, but not the shape
or arrangement of the component flares.
. The big complex flares with lots of components are
created by long zoom lenses with many internal lens
elements. Wider prime lenses create simpler flares.
. Because they are caused within the lens, flares beyond
the source appear superimposed over the image, even
over objects in the foreground that partially block the
source flare.
Moreover, not every bright light source that appears in
frame will cause a lens flare—not even the sun.
The Lens Flare effect included with After Effects is rather
useless as it contains only three basic presets. Knoll Light
Factory, available from Red Giant Software, is much more
helpful both because the presets correspond to real lenses
and because the components can be fully customized in
a modular fashion. The lens flare plug-in offered by The
Foundry as part of Tinderbox also makes realistic-looking
flares possible, although the included defaults are not so
convincing.

Reflected Light
Reflected light is another “kiss of love” opportunity for a
scene. You might not notice that it’s missing, but a glim-
mer, glint, or full reflection can add not only realism but
pizzazz to a shot.
Glints are specular flares that occur when light is reflected Figure 12.11 This sequence shows
toward the camera from shiny parts of an element in scene, the glint that plays off the chrome
areas of the taxi as it passes a spot in
such as the chrome of the taxi in Figure 12.11, taken from the frame where the sun is reflected
the Chapter 5 color matching example. directly into the camera lens.

393

Chapter 12 Light

Figure 12.12 There’s not a whole lot to a glint when you look at it closely, but it
helps sell the plane.

The same plug-ins used for flares (Trapcode Lux, Knoll
Light Factory, and Tinderbox) can be used to create glints,
with more modest settings that don’t create all the lens
reflections. Figure 12.12 shows that there’s not necessar-
ily a whole lot to a single glint. An Add mode makes these
work even in 32 bpc HDR (in which case they create over-
range values, just as they would on film).

Light Scattering and Volume
Light scatters as it encounters particles in the air, most
dramatically causing the phenomena of volumetric light or
God rays. Our atmosphere does not permit light to travel
directly to the camera, uninterrupted, as it does in outer
space. Instead, the light ricochets off tiny particles in the
air, revealing its path.
The effect can be subtle. Lights that appear in the scene,
casting their beams at the camera, tend to have a glowing
halo around them. If the light traveled directly to the
camera, the outline of the source light would be clear.
Instead, light rays hit particles on their way to the camera
and head off in slightly new directions, causing a halo
(Figure 12.13).

394


Three-Way Blur
After Effects offers quite a few blur effects, but
three are most common for general usage:
Gaussian Blur, Fast Blur (which at best quality is no
different), and Box Blur, which can match the other
two but offers more flexibility.

At the default Iterations setting (1), a Box Blur can
Figure 12.13 You’re so used to light halation that it looks wrong to lower the seem crude and, well, boxy, but it can approximate
exposure so that it disappears. the look of a defocused lens without all the more
complex polygons of Lens Blur; you can also hold it
Add more particles in the air (in the form of smoke, fog, out to the horizontal or vertical axis to create a nice
or mist), and you get more of a halo, as well as the condi- motion blur approximation (where Directional Blur
is actually too smooth and Gaussian).
tions under which volumetric light occurs. God rays are
the result of the fact that light from an omnidirectional
source, such as the sun, travels outward in a continuous arc Raising the Box Blur Iterations setting above 3 not
(Figure 12.14). only amplifies the blur but refines the blur kernel
beyond anything the other two effects are capable
of producing. What actually occurs is that the blur
goes from a square appearance (suiting the name
box blur) to a softer, rounder look. You’re more
likely to notice the difference working with over-
range bright values in 32 bit HDR.

Fast Blur and Box Blur also each include a Repeat
Edge Pixels checkbox; enable this to avoid dark
borders when blurring a full frame image. The
same setting with these two effects will not, alas,
produce the same amount of blur even if Box Blur
is set to 3 iterations (to match Fast Blur).

Figure 12.14 The Devil Rays are a baseball team; these are God rays.

The CC Light Rays effect is probably most helpful among
those included with After Effects to re-create volumetric
light effects, and even God rays. It not only boosts and

395

Chapter 12 Light

causes halation around the source light, but it also adds
rays coming straight at camera. These rays can be made
more prominent by boosting radius and intensity, but in
order to create a God rays effect and not overwhelm an
entire image with rays, it’s usually best to make a target
source and apply the effect to that. For example, you can
1. Add a solid of your preferred color.

2. Apply Fractal Noise (default settings are acceptable
to begin).
3. Mask the solid around the target God rays source area.
Feather it heavily.
4. Apply CC Light Rays. Place the Center at the God rays
target. Boost Intensity and Radius settings until the
rays are prominent.
5. For rays only (no fractal noise) set Transfer Mode
to None.
Figure 12.15 The included CC Light
Rays effect is essential to creating your 6. Set a Subtract mask or Alpha Inverted track matte to
own volumetric light effects in After
Effects. Masks or mattes can be used create occluded areas for the rays to wrap around, as
to occlude rays. in Figure 12.15.
You can further hold out and mask out the rays as needed,
even precomping and moving the source outside of frame
if necessary. To make the rays animate, keyframe the Evolu-
tion property in Fractal Noise or add an expression such as
A more straightforward plug-in for time*60 causing it to undulate over time. Different Fractal
volumetric light is Trapcode Lux, Type and Noise Type settings will also yield unique rays.
which can derive volume and a
flare from any After Effects 3D light
simply by applying it to an adjust-
Shadows
ment layer. As there is light, so must there be shadows. Unfortunately,
they can be difﬁcult to re-create in 2D because they inter-
act with 3D space and volume, none of which 2D layers
have. The behavior of shadows can be unpredictable, but
luckily, your audience typically doesn’t know how they
should look in your scene either.
You can certainly cast a shadow from a matted layer onto a
plane by positioning each of them in 3D space and prop-
erly positioning a light. Be sure that you ﬁrst change Casts
Shadows for the matted layer from its default Off setting to

396


On or Only (the latter option making it possible to create a
precomp containing only the shadow).
You can instead corner pin the matte to the angle at which
the shadow should fall and avoid the 3D setup altogether.
In either case, the problem is that the illusion breaks if
the light source is more than 10 degrees off-axis from the
camera. The more you light a 2D element from the side,
the more it just breaks (Figure 12.16).

Figure 12.16 Compare the fake 3D shadow with the real thing and you instantly grasp the problem with this approach. You
can cast a good shadow head-on, but not at this steep an angle.

There’s also the possibility of cheating: if it’s easy to add
ground surface that would obscure a shadow (for example,
grass instead of dirt), do so, and no one will even expect to
see a shadow because it no longer belongs there.

Contact Shadows and Indirect Light
For the most part, successful shading in a 2D scene relies on
getting what you can practically, and getting creative when
there are no practical shadows. There are plenty of cases
where a full cast shadow would be correct and no shadow at
all clearly looks wrong, but a simple contact shadow will at
least remove glaring contrast.
A contact shadow is a lot like a drop shadow, basically just
an offset, soft, dark copy directly behind the foreground.
A drop shadow, however, is good only for casting a shadow
onto an imaginary wall behind a logo, whereas a contact

397

Chapter 12 Light

shadow is held out to only the areas of the foreground that
have contact with the ground plane.
Figure 12.17 shows the foreground layer duplicated and
placed behind the source. A mask is drawn around the
base, and it is then offset downward. A blur is applied to
soften the transparency channel. That gives you the matte.

Figure 12.17 A simple contact shadow can make the difference between an
object that appears to sit on a surface and an object that appears to float in space.

A shadow is not just a black overlay; it’s an area of reduced
light and therefore, color. Instead of darkening down the
matte itself to create the shadow, create an adjustment
layer just below the contact shadow layer and set an Alpha
track matte. Add a Levels (or if you prefer, Curves) effect
Figure 12.18 The obvious way to cre-
and adjust brightness and gamma downward to create your
ate a shadow is just to overlay a black shadow. Treat it like a color correction, working on sepa-
layer and adjust opacity (top) but that rate channels if necessary; the result is more interesting
approach looks bland, fails to match
and accurate than a pool of blackness (Figure 12.18).
the shadows in the scene and flattens
out detail. By applying the shadow Reﬂected light is another type of contact lighting that
layer as a track matte to an adjustment
layer you can treat the shadow as a plays a role in how things look in the real world, and thus
color correction and match the color in your composited scenes. Most objects in the world have
and contrast of the scene’s existing diffuse surfaces that reﬂect light; when the object is promi-
shadows (bottom).
nent or colorful enough, your eye expects it to affect other
more neutral neighboring surfaces (Figure 12.19).

398


Figure 12.19 The color influence of indirect light is not always so evident as with a bright saturated object, but it is always
in play in a natural setting.

In 3D computer graphics this fact has been realized via
global illumination, which considers light and surface
interactions in a render. If you’re working with a computer-
generated scene, aim for a pass that includes these types
of light interactions. If you have to create the effect from
scratch, the method is similar to that of shadows or color
matching: Create a selection of the area reflecting light,
including any needed falloff (that may be the hard part),
and use color correction tools to color match the adjacent
object reflecting color.

Multipass 3D Compositing
Some artists, including a majority of those who work
predominantly in 3D, labor under the delusion that you
should finalize the look of a computer-generated element
in one pass. Certainly, as it becomes more and more pos-
sible to adjust the look of a 3D model in real time (via the
GPU, OpenGL) this becomes tempting.
However, it’s possible to do better by dividing the render of
a single element into multiple passes. This is different from
rendering in layers, which while also useful for composit-
ing is really only about separating foreground elements
from the background. Multipass rendering is the technique
of isolating individual surface qualities and creating a sepa-
rate render for each. By surface qualities I mean things like
specularity and wear and tear, also known as grunge. In his

399

Chapter 12 Light

excellent book Digital Lighting & Rendering, Second Edition
(Peachpit Press, 2006), Jeremy Birn calls out multiple ben-
efits yielded by rendering a model on multiple passes:
. Changes can be made with little or no re-rendering. If
a shadow is too dark or a glow is the wrong color, the
adjustment can be made right in After Effects.
. Integration often requires multiple passes where the
model interacts with the scene, casting a shadow on the
ground or being reflected in water. If the cast shadow is
simply part of a single render you lose all control over
its appearance and cannot apply it as recommended in
the previous section.
. Reflections, which often consume massive amounts of
time to process, can be rendered at lower quality and
blurred in After Effects.
. Bump Maps can be applied more selectively (held out
by another pass such as a highlight or reflection pass).
Got UV Maps? After Effects has the . Glows can be created easily in 2D by simply blurring
means to use them with the RE:Map and boosting exposure of a specular pass.
plug-in from RE:Vision Effects. This . Depth of Field can be controlled entirely in 2D by
allows you to map a texture to an
object, using the UV map for coor- using a Z pass as a matte for a blur adjustment layer.
dinates, without returning to the . Less render power and time is required to render any
3D app that generated it or waiting one pass than the entire shaded model, so a lower
for Live Photoshop 3D to support
mapping in After Effects (which it powered computer can do more, and redoing any one
doesn’t in CS4). element takes far less time than redoing the entire
finished model.
Putting multiple passes to use is also surprisingly simple;
the artistry is in all of the minute decisions about which
combination of adjustments will bring the element to life.
Table 12.1 (on the next page) describes some common
render passes and how they are typically used.
Other passes might include: a Fresnel (or Incidence) pass
showing the sheen of indirect light and applied to an
adjustment layer with a Luma Matte (raise Output Black in
Levels to re-create sheen); a Grunge or Dirt map, applied as
a Luma Inverted Matte, allowing you to dial in areas of wear
and tear with Levels on an adjustment layer; a Light pass for
any self-illuminated details; and a Normal pass showing the

400


TABLE 12.1 Ten Typical Multipass Render Layer Types
COLOR/ TYPICAL
TYPE GRAYSCALE BLENDING MODE DESCRIPTION USE
Diffuse Color Normal Full color render; includes Color basis for the element;
diffuse illumination, color main target for primary color
correction and texture, excludes
reflections, highlights and
shadows
Specular Color Add or Screen Isolated specular highlights Control how highlights are
rendered; can be reused to
create a glow pass by simply
blurring and raising exposure
Reflection Color Add or Screen Self-reflections, other objects, Control the prominence and
environment color of reflections
Shadow Grayscale Luma Inverted Isolated translucent shadows Control appearance, color and
Matte in scene softness of shadows; applied
as a track matte to an adjust-
ment layer with a Levels or
Curves effect
Ambient Color Color Color and texure maps without Color reference, can be used
diffuse shading, specular to make the color/texture
highlights, shadows, or of an object more pure and
reflections visible
Occlusion Grayscale Luma Inverted Shadows that result from soft Adds natural soft shadows to
Matte illumination, simulating light an object; these can be tinted
from an overcast sky or well-lit to reflect the color of reflected
room light
Beauty Color Normal A render of all passes Reference: this is how the
object or scene would appear
if rendered in a single pass
Global Color Add or Screen Indirect light added to the Control intensity of indirect
Illumination scene by global illumination, lighting in scene
potentially including raytraced
reflections and refractions
Matte/ Mask/ Grayscale Luma Matte Can be used to contain multiple
Alpha transparency masks for portions
of the object or scene, one each
on the red, green, blue, and
alpha channels
Depth/Z-depth/ Grayscale or Luma Matte Describes the distance of Can be used to control depth
Depth Map non-image surface areas from the camera effects such as fog and lens
floating point blur, as well as light fall-off

401

Chapter 12 Light

direction of surface normals for relighting purposes. Many,
many more are possible—really anything that you can
isolate in a 3D animation program. Figure 12.20a through i
show a robot set up for multipass rendering and a few of its
component render layers.

A

B C D E

F G H I
Figure 12.20a through i A basic multipass setup (a, above) with the beauty pass (b) as reference, and made up of the fol-
lowing color passes: diffuse (c), specular (d) and reflection (e) as well as grayscale passes applied as luma mattes to adjust-
ment layers, each containing a Levels effect: grunge (f ), incidence (g), and occlusion (h). A depth matte (i) can be applied
in various ways; here it is used as reference for a lens blur on an adjustment layer (to give the appearance of shallow depth
of field).

402


Note that none of these passes necessarily requires a
transparency (alpha) channel, and at the biggest old-
school effects houses it is customary not to render them,
since multiple passes of edge transparency can lead to
RPF and EXR
image multiplication headaches.
RPF files are an Autodesk update to RLA. After
The general rules for multipass compositing are simple: Effects offers limited native support for these files
(via the effects in the 3D Channel menu) but more
. Use the Diffuse layer as the base. robust support for some of the finer features of RPF
such as Normal maps is only available via third-
. Apply color layers meant to illuminate the base layer, party plug-ins. Commercially available plug-ins
such as specular and reflection, via Add or Screen that can translate normal maps for use in After
blending modes. Effects include ZBornToy (which also does amazing
things with depth maps; a demo is available on
. Apply color layers meant to darken the base layer, if this book’s disc) from Frischluft and WalkerFX
any, via Multiply or Darken blending modes. Channel Lighting, part of the Walker Effects collec-
. Apply grayscale maps as luma mattes for adjustment tion. There is a free option for Windows only called
Normality (www.minning.de/software/normality).
layers. Apply Levels, Curves, and Hue/Saturation to
allow these mattes to influence the shading of the
object or scene. As mentioned in Chapter 8, After Effects can also
extract camera data from RPF files (typically gener-
. Control the strength of any layer using that layer’s ated in 3DS Max or Flame); place the sequence
Opacity. containing the 3D camera data in a comp and
choose Animation > Keyframe Assistant > RPF
Note that multipass renders present an excellent case to
Camera Import.
enable Blend Colors Using 1.0 Gamma in Project Settings,
whether or not you assign a Working Space (and whether
or not that working space is linearized). The most popular way to get a multipass render
into After Effects these days is via the EXR format,
Multipass rendering is only partially scientific and accurate; which can store all of the various passes and
successful use of multiple passes is a highly individualized related data in a single file. EXR is now supported
directly in After Effects CS4 thanks to the inclusion
and creative sport. With the correct basic lighting setup of EXtractoR and IDentifier, two plug-ins from
you can use multipass renders to place a given 3D element fnordware.
in a variety of environments without the need for a com-
plete re-render.
Varied environments are themselves the subject of the fol-
lowing chapter.

403

CHAPTER

13
Climate and the Environment

Conversation about the weather is the last refuge
of the unimaginative.
—Oscar Wilde

Yes, yes, let’s talk about the weather.
—W. S. Gilbert, The Pirates of Penzance,
or the Slave of Duty

Climate and the Environment

E ven if you’re not called upon to re-create extreme
climate conditions (as someone once pointed out seems
to be a theme in my work), even a casual glance out the
window demonstrates that the meteorological phenomena
are always in play: a breeze is blowing in the trees, or water
and particulate in the air are changing the appearance of
buildings and the land closest to the horizon.
This chapter offers methods to create natural elements such
as particulate and wind effects, as well as to replace a sky, or
add mist, fog, or smoke, or other various forms of precipita-
tion. All of these are more easily captured with a camera
than re-created in After Effects, but sometimes the required
conditions aren’t available on the day of a shoot. Mother
nature is, after all, notoriously ﬁckle, and shooting just to get
a particular environment can be extraordinarily expensive.
It’s rare indeed that weather conditions cooperate on
location, and even rarer that a shoot can wait for perfect
weather or can be set against the perfect backdrop. Trans-
forming the appearance of a scene using natural ele-
ments is among the more satisfying things you can do as a
compositor.

Particulate Matter
Particulate matter in the air inﬂuences how objects appear
at different depths. What is this matter? Fundamentally, it

406


is water and other gas, dust, or visible particulate usually
known as pollution. Even in an ideal, pristine, pollution-
free environment there is moisture in the air—even in the
driest desert, where there also might be heavier forms of Particulate matter does not occur
particulate like dust and sand. The amount of haze in the in outer space, save perhaps when
air offers clues as to the occasional cloud of interstellar
dust drifts through the shot. Look
. The distance to the horizon and of objects in relation at photos of the moon landscape,
to it and you’ll see that the blacks in the
distance look just as dark as those
. The basic type of climate; the aridness or heaviness of in the foreground.
the weather
. The time of year and the day’s conditions
. The air’s stagnancy (think Blade Runner)
. The sun’s location (when it’s not visible in shot)
The color of the particulate matter offers clues to how
much pollution is present and what it is, even how it feels:
dust, smog, dark smoke from a ﬁre, and so on (Figure 13.1).

Figure 13.1 Same location, different conditions. Watch for subtleties: backlighting emphasizes even low levels of haze and
reduces overall saturation; more diffuse conditions desaturate and obscure the horizon while emphasizing foreground color.

Essentially, particulate matter in the air lowers the appar-
ent contrast of visible objects; secondarily, objects take on
the color of the atmosphere around them and become
slightly diffuse. This is a subtle yet omnipresent depth cue:
With any particulate matter in the air, objects lose contrast
further from camera; the apparent color can change quite
a bit, and detail is softened. As a compositor, you use this to
your advantage, not only to re-create reality, but to provide
dramatic information.

407

Chapter 13 Climate and the Environment

Match an Environment
Figure 13.2 shows how the same object at the same size
indicates its depth by its color. The background has great
foreground and background reference for black and
white levels; although the rear plane looks icy blue against
gray, it matches the look of gray objects in the distance of
the image.

Figure 13.2 One plane (white circle) is composited as if it is a toy model in the foreground, the
other (black circle) as if it is crossing the sky further in the distance; a look at just the planes shows
them to be identical but for their color (left). The difference between a toy model airplane flying
close, a real airplane flying nearby, and the same plane in the distant sky is conveyed with the use
of Scale, but just as importantly, with Levels that show the influence of atmospheric haze.

The technique used here is the same as the one outlined
in Chapter 5 with the additional twist that now you under-
stand how atmospheric haze inﬂuences the color of the
scene. Knowing how this works from studying a scene like
this one helps you create something similar from scratch
even without such good reference.
The plane as a foreground element seems to make life
easier by containing a full range of monochrome colors.
When matching a more colorful or monochrome element,
you can always create a small solid and add the default
Ramp effect. With such a reference element, it is simple to
add the proper depth cueing with Levels and then apply
the setting to the ﬁnal element (Figure 13.3).

Figures 13.3 If your foreground layer lacks the full range of values,
match a grayscale gradient, then swap in the element. These gradient
squares have the same settings as the planes in 13.2.

408


Create an Environment
To re-create depth cues in a shot from scratch, you must
somehow separate the shot into planes of distance. If the
source imagery is computer-generated, the 3D program
that created it can also generate a depth map for you to
use. If not, you can slice the image into planes of distance,
or you can make your own depth map to weight the dis-
tance of the objects in frame (Figure 13.4).

Figure 13.4 This grayscale depth map is made up of individual roto shapes, each shaded for relative distance (further eluci-
dated in Advanced Rotoscoping Techniques for Adobe After Effects by Pete O’Connell, Creative Cow).

There are several ways in which a depth map can be used,
but the simplest approach is probably to apply it to an
adjustment layer as a Luma (or Luma Inverted) Matte,
and then add Levels or other color correction adjustment
Getting reference is easy for anyone
to the adjustment layer. Depth cueing would be applied
with an Internet connection these
to the furthest elements, so applying this matte as a Luma days, thanks to sites and services like
Inverted Matte and then ﬂashing the blacks (raising the flickr.com and Google image search.
Output Black level in Levels) adds the effect of atmosphere
on the scene.

409


Figure 13.5 An RPF image close-up
can be clearly seen to have jagged
edge pixels, but a depth map does
not have to be perfectly pristine. An
EXR does not have this limitation.
(Created by Fred Lewis; used with per-
mission from Inhance Digital, Boeing,
and the Navy UCAV program.)

Depth data may also be rendered and stored in an RPF file,
as in Figure 13.5 (which is taken from an example included
on the disc as part of 13_multipass.aep). RPF files are in
One good reason to use a basic 3D
render of a scene as the basis for a some ways crude, lacking even thresholding in the edges
matte painting is that the perspec- (let alone higher bit depths), but they can contain several
tive and lens angle can be used not types of 3D data, as are listed in the 3D Channel menu.
only for reference, but to generate a This data can be used directly by a few effects to simulate
depth map, even if the entire scene
is painted over.
3D, including Particle Playground, which accepts RPF data
as an influence map.
More extreme conditions may demand actual 3D particles,
which are saved for later in the chapter.

Sky Replacement
Sky replacement is among the most inexpensive and fun-
damental extensions that can be made to a shot. It opens
up various possibilities to shoot faster and more cheaply.
Not only do you not have to wait for ideal climate condi-
tions, you can swap in a different sky or an extended physi-
cal skyline.
Skies are, after all, part of the story—often a subliminal
one, but occasionally a starring element. An interior with
a window could be anywhere, but show a recognizable sky-
line outside the window and locals will automatically gauge

410


the exact neighborhood and city block of that location,
along with the time of day, time of year, weather, outside
temperature, and so on, possibly without ever really paying
conscious attention to it.
You could spend tens (even hundreds) of thousands of
dollars for that view apartment on Central Park East to
ﬁlm a scene at golden hour (the beautiful “hour” of sunset
that typically lasts about 20 minutes and is missing on an
overcast day). The guerilla method would be to use your
friend’s apartment, light it orange, shoot all day, and add
the sunset view in post. In many cases, the real story is
elsewhere, and the sky is a subliminal (even if beautiful)
backdrop that must serve that story (Figures 13.6).

Figure 13.6 For an independent film with no budget set in San Francisco, the
director had the clever idea of shooting it in a building lobby across the bay in
lower-rent Oakland (top left), pulling a matte from the blue sky (top center), and
match moving a still shot of the San Francisco skyline (from street level, top right)
for a result that anyone familiar with that infamous pyramid-shaped building
would assume was taken in downtown San Francisco (bottom). (Images courtesy
of The Orphanage.)

411


The Sky Is Not (Quite) a Blue Screen
Study the actual sky (there may be one nearby as you read
this) or even better, study reference images, and you may
notice that the blue color desaturates near the horizon,
cloudless skies are not always so easy to come by, and even
clear blue skies are not as saturated as they might some-
times seem.
Some combination of a color keyer, such as Keylight, and
a hi-con luminance matte pass or a garbage matte, as
needed, can remove the existing sky in your shot, leaving
nice edges around the foreground. Chapter 6 focuses on
strategies for employing these, and Chapter 7 describes
supporting strategies when keys and garbage mattes fail.
The ﬁrst step of sky replacement is to remove the existing
“sky” (which may include other items at inﬁnite distance,
such as buildings and clouds) by developing a matte
for it. As you do this, place the replacement sky in the
background; a sky matte typically does not have to be as
exacting as a blue-screen key if the replacement sky is also
fundamentally blue, or if the whole image is radically color
corrected as in Figure 13.7.

Figure 13.7 A very challenging matte for several reasons,
not the least of which is the uneven desaturated quality of
the sky. Placing the pigeon in a radically different environ-
ment requires color and grain matching to compensate for
a less than stellar source.

412


Infinite Depth
A locked-off shot can be completed with the creation of
the matte and a color match to the new sky. If, however,
there is camera movement in the shot, you might assume
that a 3D track is needed to properly add a new sky
element.
Typically, that’s overkill. Instead, consider
. When matching motion from the original shot, if
anything in the source sky can be tracked, by all means
track the source.
. If only your foreground can be tracked, follow the
suggestions in Chapter 8 for applying a track to a
3D camera: Move the replacement sky to the distant
background (via a Z Position value well into four or
ﬁve digits, depending on camera settings). Scale up to
compensate for the distance; this is all done by eye.
. A push or zoom shot (Chapter 9 describes the differ-
ence), may be more easily re-created using a tracked 3D
camera (but look at Chapter 8 for tips on getting away
with a 2D track).
The basic phenomenon to re-create is the scenery at
inﬁnite distance that moves less than objects in the fore-
ground. This is the parallax effect, which is less pro-
nounced with a long, telephoto lens, and much more
obvious with a wide angle. For the match in Figure 13.6, a
still shot (no perspective) was skewed to match the correct
angle and tracked in 2D; the lens angle was long enough
and the shot brief enough that they got away with it. A
simpler example is included on the disc.

Fog, Smoke, and Mist
An animated layer of translucent clouds is easy to re-create
in After Effects. The basic element can be fabricated by
applying the Turbulent Noise effect to a solid, then adding
a blending mode such as Add or Screen with the appropri-
ate Opacity setting. On the book’s disc, 13_smokyFlyover.
aep contains a simple example of layers of smoke laid out
as if on a three-dimensional plane.

413


Figure 13.8 The new Turbulent Noise effect (shown at the default setting) is a decent stand-in for organic-looking ambient
smoke and fog. Animate the Evolution if you want any billowing of the element. The same alternate Fractal Types as found
in the older Fractal Noise effect are available, as seen in the pop-up menu.

Turbulent Noise at its default settings already looks smoky
(Figure 13.8); switching the Noise Type setting from the
default, Soft Linear, to Spline improves it. The main thing
to add is motion, which I like to do with a simple expres-
sion applied to the Evolution property: time*60 (I find
60 an appropriate rate in many situations, your taste may
vary). The Transform properties within Fractal Noise can
be animated, causing the overall layer to move as if being
blown by wind.
Brightness, Contrast, and Scale settings influence the
apparent scale and density of the noise layer. Complexity
and Sub settings also affect apparent scale and density,
Turbulent Noise is a faster and but with all kinds of undesirable side effects that make the
more natural looking alternative to smoke look artificial. The look is greatly improved by layer-
Fractal Noise. ing at least two separate passes via a blending mode (as in
the example project).

Masking and Adjusting
When covering the entire foreground evenly with smoke
or mist, a more realistic look is achieved using two or

414


three separate overlapping layers with offset positions
(Figure 13.9). The unexpected byproduct of layering 2D
particle layers in this manner is that they take on the illu-
sion of depth and volume.

Figure 13.9 Overlay layers of fabri-
cated smoke to add dimensionality
and depth.

The eye perceives changes in parallax between the fore-
ground and background and automatically assumes these
to be a byproduct of full three-dimensionality, yet you save
Fractal Noise texture maps maintain
the time and trouble of a 3D volumetric particle render. Of the advantage over Turbulent Noise
course, you’re limited to instances in which particles don’t in that they can loop seamlessly
interact substantially with movement from objects in the (allowing reuse on shots of varying
scene; otherwise, you instantly graduate to some very tricky length). In Evolution Options,
enable Cycle Evolution, and animate
3D effects (although I have seen it done in After Effects Evolution in whole revolutions (say,
with warps). from 0° 2 × 0.0°). Set the Cycle
(in Revolutions) parameter to the
Particle layers can be combined with the background via number of total revolutions (2). The
blending modes, or they can be applied as a Luma Matte first and last keyframes now match,
to a colored solid (allowing you to specify the color of the and a loopOut(“cycle”)
particles without having a blending mode change it). expression continues this loop
infinitely.
To add smoke to a generalized area of the frame, a big
elliptical mask with a high feather setting (in the triple
digits even for video resolution) will do the trick; if the
borders of the smoke area are apparent, increase the mask
feather even further (Figure 13.10).

415


Figure 13.10 This mask of a single
smoke element from the shot in
Figure 13.8 has a feather value in the
hundreds. The softness of the mask
helps to sell the element as smoke
and works well overlaid with other,
similarly feathered masked elements.

Moving Through the Mist
The same effect you get when you layer several instances of
Turbulent or Fractal Noise can aid the illusion of moving
forward through a misty cloud. That’s done simply enough
(for an example of ﬂying through a synthetic cloud, see
13_smokyLayers.aep), but how often does your shot consist
of just moving through a misty cloud? Most of the time,
clouds of smoke or mist are added to an existing shot.
You can use the technique for emulating 3D tracking (see
Chapter 8) to make the smoke hold its place in a particular
area of the scene as the camera moves through (or above)
Selling the Effect with Diffraction
it. To make this work, keep a few points in mind:
There is more to adding a cloud to a realistic shot
than a simple A over B comp; water elements in . Each instance of Fractal Noise should have a soft ellipti-
the air, whether in spray, mist, or clouds, not only cal mask around it.
occlude light but diffract it. This diffraction effect
can be simulated by applying Compound Blur to an . The mask should be large enough to overlap with
adjustment layer between the fog and the back- another masked instance, but small enough that it
ground and using a precomposed (or prerendered) does not slide its position as the angle of the camera
version of the fog element as its Blur layer.
changes.
. A small amount of Evolution animation goes a long
This usage of Compound Blur is is detailed further
in the next chapter, where it is used to enhance the
way, and too much will blow the gag. Let the movement
effect of smoky haze. of the camera create the interest of motion.
. Depending on the length and distance covered in the
shot, be willing to create at least a half-dozen individual
masked layers of Fractal Noise.
13_smokyFlyover.aep features just such an effect of moving
forward through clouds. It combines the tracking of each
shot carefully into place with the phenomenon of parallax,
whereby overlapping layers swirl across one another in a

416


believable manner. Mist and smoke seem to be a volume
but they actually often behave more like overlapping, trans-
lucent planes—individual clouds of mist and smoke.

Billowing Smoke
Fractal Noise works ﬁne to create and animate thin wispy
smoke and mist. It will not, however, be much help if you
need to fabricate thick, billowing clouds. Instead of a plug-
in effect, all you need is a good still cloud element and you
can animate it in After Effects. And all you need to create
the element is a high-resolution reference photo—or even
a bag of cotton puffs, as were used to create the images in
Figure 13.11.

Figure 13.11 A good static image, even cotton puffs arranged on black posterboard, photographed in daylight, can be
used as the foundation of billowing smoke.

To give clouds shape and contour, open the image in
Photoshop, and use the Clone Stamp tool to create a cloud
with the shape you want. You can do it directly in After
Effects, but this is the kind of job for which Photoshop
was designed. Clone in contour layers of highlights (using
Linear Dodge, Screen, or Lighten blending modes) and
shadows (with Blending set to Multiply or Darken) until
the cloud has the look you’re after (Figure 13.12).

417


Figure 13.12 The elements from
Figure 13.11 are incorporated into this
matte painting, and the final shot con-
tains a mixture of real and composited
smoke.

So now you have a good-looking cloud, but it’s a still. How
do you put it in motion? This is where After Effects’ excel-
lent distortion tools come into play, in particular Mesh
Warp and Liquify. A project containing just such a cloud
animation appears on the disc as 13_smokeCloud.aep.

Mesh Warp
Mesh Warp lays a grid of Bézier handles over the frame; to
animate distortion, set a keyframe for the Distortion Mesh
property at frame 0, then move the points of the grid and
realign the Bézier handles associated with each point to
bend to the vertices between points. The image to which
this effect is applied follows the shape of the grid.
By default, Mesh Warp begins with a seven-by-seven grid.
Before you do anything else, make sure that the size of
the grid makes sense for your image; you might want to
Mesh Warp, like many distortion
tools, renders rather slowly. As you
increase its size for a high-resolution project, and you can
rough in the motion, feel free to reduce the number of rows to ﬁt the aspect ratio of your
work at quarter-resolution. When shot, for a grid of squares (Figure 13.13).
you’ve finalized your animation,
you can save a lot of time by pre- You can’t typically get away with dragging a point more
rendering it (see Chapter 4). than about halfway toward any other point; watch care-
fully for artifacts of stretching and tearing as you work, and
preview often. If you see stretching, realign adjacent points
and handles to compensate. There is no better way to learn
about this than to experiment.

418


Figure 13.13 The Mesh Warp controls
are simple, just a grid of points and
vectors. You can preset the number
and quality; more is not necessarily
better (just more to control). Points can
be multiselected and dragged, and
each point contains Bézier handles for
warping the adjacent vectors.

I have found that the best results with Mesh Warp use mini-
mal animation of the mesh, animating instead the element
that moves underneath it.

Liquify
Mesh Warp is appropriate for gross distortions of an entire
element. The Liquify effect is a brush-based system for fine
distortions. 13_smokeCloud.aep includes a composition
that employs Liquify to swirl a cloud. Following is a brief
orientation to this toolset, but as with most brush-based
painterly tools, there is no substitute for trying it hands-on.
The principle behind Liquify is actually similar to that of
Mesh Warp; enable View Mesh under View Options and
you’ll see that you’re still just manipulating a grid, albeit
a finer one that would be cumbersome to adjust point by
point—hence the brush interface.
Of the brushes included with Liquify, the first two along
the top row, Warp and Turbulence, are most often used
(Figure 13.14). Warp has a similar effect to moving a point
in Mesh Warp; it simply pushes pixels in the direction you
drag the brush. Turbulence scrambles pixels in the path of Figure 13.14 Liquify is also a mesh
the brush. distortion tool, only the mesh is
much finer than Mesh Warp’s and it is
The Reconstruction brush (rightmost on the bottom row) controlled via brushes, allowing more
is like a selective undo, reversing distortions at the default specific distortions.

419


setting; other options for this brush are contained in the
Reconstruction Mode menu (which appears only when the
brush is selected).
Liquify has the advantage of allowing hold-out areas. Draw
a mask around the area you want to leave untouched by
Liquify brushes, but set the Mask mode to None, disabling
it. Under Warp Tool Options, select the mask name in the
Freeze Area Masked menu.
Liquify was a key addition to the “super cell” element (that
huge swirling mass of weather) for the freezing of the New
York City sequence in The Day After Tomorrow. Artists at The
Orphanage were able to animate matte paintings of the
cloud bank, broken down into over a dozen component
parts to give the effect the appropriate organic complexity
and dimension.

Smoke Trails and Plumes
Many effects, including smoke trails, don’t require particle
generation in order to be re-created faithfully. This sec-
tion is included less because the need comes up often and
more to show how, with a little creativity, you can combine
techniques in After Effects to create effects that you might
think require extra tools.
Initial setup of such an effect is simply a matter of starting
with a clean plate, painting the smoke trails in a separate
still layer, and revealing them over time (presumably
behind the aircraft that is creating them). The quickest
and easiest way to reveal such an element over time is
often by animating a mask, as in Figure 13.15, or you could
use techniques described in Chapter 8 to apply a motion
tracker to a brush.
The second stage of this effect is dissipation of the trail;
depending on how much wind is present, the trail might
drift, spread, and thin out over time. That might mean that
in a wide shot, the back of the trail would be more dis-
sipated than the front, or it might mean the whole smoke
trail was blown around.

420


Figure 13.15 No procedural effect is
needed; animating out masks is quick,
simple, and gives full control over
the result.

One method is to displace with a black-to-white gradient
(created with Ramp) and Compound Blur. The gradient
is white at the dissipated end of the trail and black at the
source (Figure 13.16a); each point can be animated or
tracked in. Compound Blur uses this gradient as its Blur
Layer, creating more blur as the ramp becomes more
white. Another method, shown in Figure 13.16b, uses
a different displacement effect, Turbulent displace, to
create the same type of organic noise as in the preceding
cloud layers.

Figures 13.16a and b To dissipate a smoke trail the way the wind would, you can do so using a gradient and Compound
Blur, so that the smoke dissipates more over time, or using the Turbulent Displace effect that, like Turbulent Noise, adds frac-
tal noise, using it to displace the straight trails from Figure 13.15.

421


Wind
What is wind doing in this chapter? You can’t see it. Never-
theless, a static environment is rarely believable if it’s not
the surface of the moon, and because your job is to help
the viewer suspend disbelief, you may have to consider add-
ing the inﬂuence of wind to your scene.
The fact is that most still scenes in the real world contain
ambient motion of some kind. Not only objects directly in
the scene, but reﬂected light and shadow might be chang-
ing all the time in a scene we perceive to be motionless.
As a compositor, you always look for opportunities to add
to them in ways that contribute to the realism of the scene
without stealing focus. Obviously, the kinds of dynamics
involved with making the leaves and branches of a tree sway
are mostly beyond the realm of 2D compositing, but there
are often other elements that are easily articulated and
animated ever so slightly. Successful examples of ambient
animation should not be noticeable, and they often will not
have been explicitly requested, so it’s an exercise in subtlety.

Adding and Articulating Elements
To make it easier on yourself, look for elements that can be
readily isolated and articulated; you should be able to mask
the element out with a simple roto or a hi-con matte if it’s
Primary and Secondary
not separated to begin with. Look for the point where the
Primary animation is the gross movement of the
object, the movement of the object as a whole. object would bend or pivot, place your anchor point there,
Secondary animation is the movement of individual then animate a gentle rotation. 13_ambientAnim.aep (on
parts of the object as a result of inertia. So, for the disc) offers a simple animation of the arm of a street-
example, a helicopter crashes to the ground: That’s light, held out from the background. A little warp on the
the primary animation. Its rotors and tail bend and
shudder at impact: That’s the secondary animation.
clouds behind it, and this still image could convincingly be
For the most part, in 2D compositing, your work is a brief moving shot (Figure 13.17).
isolated to primary animation.
You also have the option of acquiring and adding elements
that indicate or add to the effect of wind motion. Figure
13.18 is an element of blowing autumn leaves shot against
a black background for easy removal and matting; granted,
you could add an element this turbulent only to a scene
that either already had signs of gusts in it or that contained
only elements that would show no secondary motion from
wind whatsoever.

422


Figure 13.17 It’s impossible to do a
full wind dynamics simulation in After
Effects, but if there should be a little
ambient motion to go with some
moving clouds, look for simple ways
to add it, such as this wiggle expres-
sion tied to the top of the light pole.

Figure 13.18 Sometimes you can find elements that will enliven your scene when comped in, such as this clip of blowing
leaves shot over black. (Footage courtesy of Artbeats.)

Precipitation
You might want to create a “dry for wet” effect, taking foot-
age that was shot under clear, dry conditions and adding
the effects of inclement weather. Not only can you not
time a shoot so that you’re guaranteed shooting in a storm
(not in most parts of the world, anyhow), but wet, stormy
conditions limit shooting possibilities and cut down on
available light. Re-creating a storm by having actual water
fall in a scene is expensive, complicated, and not necessar-
ily convincing.
I like to use the Trapcode Particular effect (demo on the
book’s disc) for the particles of accumulating rain or snow.
This effect outdoes After Effects’ own Particle Playground
for features, ﬂexibility, and fast renders. As the following
example shows, Particular is good for more than just fall-
ing particles, as well.

423


The Wet Look
Study reference photographs of stormy conditions and
you’ll notice some things that they all have in common, as
well as others that depend on variables. Here are the steps
taken to make a sunny day gloomy (Figure 13.19):

Figure 13.19 An ordinary exterior where “it never rains” becomes a deluge.

1. Replace the sky: placid for stormy (Figure 13.20a).

2. Adjust Hue/Saturation—LA for Dublin—to bring out
the green mossiness of those dry hills, I’ve knocked out
the blues and pulled the reds down and around toward
green (Figure 13.20b).
3. Exchange Tint—balmy for frigid—a bluish cast is com-
mon to rainy scenes (Figure 13.20c).
4. Fine-tune Curves—low light for daylight—aggressively
dropping the gamma while holding the highlights
makes things even moodier (Figure 13.20d).
That’s dark, but it looks as dry as a lunar surface. How do
you make the background look soaked? It seems like an
impossible problem, again, until you study reference. Then
it becomes apparent that all of that moisture in the air
causes distant highlights to bloom. This is a win-win adjust-
ment (did I really just type that?) because it also makes the
scene lovelier to behold.
You can simply add a Glow effect, but it doesn’t offer as
much control as the approach I recommend.

424


Follow these steps:
1. Bring in the background layer again (you can dupli-
cate it and delete the applied effects, Ctrl+Shift+E/
Command+Shift+E).
2. Add an Adjustment Layer below it and set the dupli-
cated background as a Luma Matte.
3. Use Levels to make the matte layer a hi-con matte that
isolates the highlights to be bloomed.
4. Fast Blur the result to soften the bloom area.

5. On the Adjustment Layer, add Exposure and Fast Blur
to bloom the background within the matte selection.
Figure 13.20e shows the result; it now looks like a wet, cold
day, but where’s the rain?

Create Precipitation
Particular contains all the controls needed to generate
custom precipitation; it contains a lot of controls, period.
A primer is helpful, to get past the default animation of
little white squares emanating out in all directions (click
under Preview in Effect Controls to see it). To get started
making rain, create a comp-sized 2D solid layer, apply
Particular, and
1. Twirl down Emitter and set an Emitter Type. For rain I
like Box so that I can easily set its width and depth, but
anything besides the default Point and Grid will work.
2. Set Emitter Size to at least the comp width in X, to ﬁll
the frame.
3. Set Direction to Directional.

4. Set X Rotation to –90 so that the particles fall downward.

5. Boost velocity to go from gently falling snow speed to
pelting rain.
Figures 13.20a through e The pro-
You might think it more correct to boost gravity than veloc- gression to a heavy, wet day (top to
ity, but gravity increases velocity over time (as Galileo dis- bottom).
covered) and rain begins falling thousands of feet above;
however, this is not something you need to replicate, as no
one will know the difference.

425


You do, however, need to do the following:
1. Move the Emitter Y Position to 0 or less so that it sits
above frame.
2. Increase the Emitter Size Y to get more depth among
those falling particles.
3. Crank up the Particles/sec and Physics Time (under
Physics) to get enough particles, full blast from the first
frame.
4. If the particles are coming up short at the bottom of
frame, increase the Life setting under Particle.
5. Enable motion blur for the layer and comp to get some
nice streaky rain (Figure 13.21).
From here, you can add Wind and Air Resistance under
Physics. If you’re creating snow instead of rain, you might
want to customize the Particle Type, even referring to your
own Custom layer if necessary for snowflakes.

Figure 13.21 In the “good enough”
category, two passes of rain, one very
near, one far, act together to create a
watery deluge. Particular could gener-
ate this level of depth with a single
field, and the mid-ground rain is miss-
ing, but planes are much faster to set
up for a fast-moving shot like this. The
rain falls mainly on two planes.

Composite Precipitation
What is the color of falling rain? The correct answer to this
zen koan-like question is that raindrops and snowflakes are
translucent. Their appearance is heavily influenced by the

426


background environment, because they behave like tiny
falling lenses. They diffract light, defocusing and lowering
the contrast of whatever is behind them, but they them-
selves also pick up the ambient light.
Therefore, on The Day After Tomorrow our crew found suc-
cess with using the rain or snow element as a track matte
for an adjustment layer containing a Fast Blur and a Levels
(or Exposure) effect, like a reflective, defocused lens.
This type of precipitation changes brightness according to
its backlighting, as it should. You may see fit to hold out
specific areas and brighten them more, if you see an area
where a volumetric light effect might occur.
The final result in Figure 13.19b benefits from a couple
of extra touches. The rain is divided into multiple layers,
near and far, and motion tracked to match the motion of
the car from which we’re watching this scene. Particular
has the ability to generate parallax without using multiple
layers, but I sometimes find this approach gives me more
control over the perspective.
Because we’re looking out a car window, if we want to call Figure 13.22 It looks jaggy because
attention to the point of view because the next shot reverses you don’t want particles to be any
higher resolution than they need to
to an actor looking out this window, it’s only appropriate be, or they take up massive amounts
that the rain bead up. This is also done with Particular, with of render time. There are two keys
Velocity turned off and Custom particles for the droplets. to creating this particle: It uses the
adjusted background, inverted, with
And because your audience always can tell when you have the CC Lens effect to create the look.
Look at raindrops on a window some-
the details wrong, even if they don’t know exactly what’s
time and notice that, as little lenses,
wrong, check out Figure 13.22 for how the droplet is they invert their fisheye view of the
designed. scene behind them.

Once again, it is attention to detail and creative license
that allow you to simulate the complexities of nature. It
can be fun and satisfying to transform a scene using the
techniques from this chapter, and it can be even more fun
and satisfying to design your own based on the same prin-
ciples: Study how it really works, and notice details others
would miss. Your audience will appreciate the difference
every time.
The next chapter heats things up with fire, explosions, and
other combustibles.

427

CHAPTER

14
Pyrotechnics:
Heat, Fire, Explosions

My nature is to be on set, blowing things up.
—Ken Ralston (winner of five Academy
Awards for visual effects)

Pyrotechnics:
Heat, Fire, Explosions

I t may not be a majority, but many people first become
interested in a visual effects career as borderline pyroma-
niacs or even gun nuts. You have to follow your passion in
life, I suppose. Creating a conflagration on the computer
isn’t quite as much fun as simply blowing stuff up, but it
keeps these people off of our streets.
The serious truth is that many types of explosions are still
best done through a combination of practical and virtual
simulations. There are, however, many cases in which
compositing can save a lot of time, expense, and hazard.
Blowing stuff up on set is fun, but it involves extensive
setup and a not insubstantial amount of danger to the cast
and crew. Second chances don’t come cheap.
On the other hand, there’s often no substitute for the
physics of live-action mayhem. I hope it doesn’t come as a
disappointment to learn that not everything pyrotechnical
can be accomplished start to finish in After Effects. Some
effects require actual footage of physical or fabricated
elements being shot at or blown up, and good reference
of such events is immensely beneficial. Practical elements
might rely on After Effects to work, but pyrotechnical shots
are equally reliant, if not more so, on practical elements.

Firearms
Blanks are dangerous, and real guns deadly. To create a
shot with realistic gunfire safely requires
. A realistic-looking gun prop in the scene
. Some method to mime or generate firing action on set

430


. The addition of a muzzle flash, smoke, cartridge, or
shell discharge (where appropriate)
. The matching shot showing the result of gunfire:
debris, bullet hits, even blood Stu Maschwitz’s book, The DV
Rebel’s Guide (Peachpit Press) is
After Effects can help with all of these to some extent, and
definitive on the subject of creating
some of them completely, relieving you of the need for an action movie, perhaps on a low
more expensive or dangerous alternatives. budget, and likely with the help
of After Effects. Included with the
The Shoot cover price, you get a couple of nifty
After Effects tools for muzzle flashes
For the purposes of this discussion let’s assume that you and spent shells, and some serious
begin with a plate shot of an actor re-creating the action of expertise on the subject of making
firing a gun, and that the gun that was used on set produces explosive action exciting and real.
nothing: no muzzle flash, no smoke, no shell. All that’s
required is some miming by the actor of the recoil, or kick,
which is relatively minor with small handguns, and a much
bigger deal with a shotgun or fully automatic weapon.
Happily, there’s no shortage of reference, as nowhere out-
side of the NRA is the Second Amendment more cherished
than in movies and television. Granted, most such reference
is itself staged, but remember, we’re going for cinematic
reality here, so if it looks right to you, use it as reference.
Figure 14.1 shows something like the minimum amount
that needs to be composited to create a realistic shot of
a gun being fired (albeit artfully executed in this case).
Depending on the gun, smoke or a spent cartridge might
also discharge. As important as the look of the frame is the
timing; check your favorite reference carefully and you’ll
find that not much, and certainly not the flash, lingers Figure 14.1 Much of the good
longer than a single frame. reference for movie gunfire is other
movies; you typically want the most
The actual travel of the bullet out of the barrel is not gen- dramatic and cinematic look, which
erally anything to worry about; at roughly one kilometer is a single frame of muzzle flash and
contact lighting on surrounding
per second, it moves too fast to be seen amid all the other elements. (Image courtesy of Mars
effects, particularly the blinding muzzle flash. Productions.)

Muzzle Flash and Smoke
The clearest indication that a gun has gone off is the flash
of light around the muzzle, at the end of the barrel. This
small, bright explosion of gunpowder actually lasts about
1
⁄48 second, short enough that when shot live it can fall

431

Chapter 14 Pyrotechnics: Heat, Fire, Explosions

between frames of film (in which case you might need to
restore it in order for the action of the scene to be clear).
Real guns don’t discharge a muzzle flash, as a rule, but
movie guns certainly do.
A flash can be painted by hand, cloned in from a practical
image, or composited from stock reference. It’s not too
significant how you generate it, although muzzle flashes
have in common with lens flares that they are specific to
the device that created them. Someone in your audience is
bound to know something about how the muzzle flash of
your gun might look, so get reference: certain guns emit
a characteristic shape such as a teardrop, cross, or star
(Figure 14.2).

Figure 14.2 The angle of the shot
and the type of gun affect the muzzle
flash effect. The first image is from an
M16 rifle; the other is from a handgun.
(Images courtesy of Artbeats.)

Any such explosion travels in two directions from the end
of the barrel: arrayed outward from the firing point and in
a straight line out from the barrel. If you don’t have source
that makes this shape at the correct angle, it’s probably
simplest to paint it.
The key is to make it look right on one frame; this is a rare
case where that’s virtually all the audience should see, and
where that one frame can be almost completely different
from those surrounding it. If it looks blah or only part of
the way there, it’s too well matched to the surrounding
frames. Focus on the one frame until you believe it for
explosiveness and dramatic flourish.
Technically speaking some guns, for example rifles, may
cause quite a bit of smoke, but most emit little or none at
all. If you do make a smoke puff with Turbulent Noise held
out by a soft mask, which you certainly could, my advice is
to make it evaporate relatively quickly so you don’t blow
the gag.

432


Shells and Interactive Light
If the gun in your scene calls for it, that extra little bit of
realism can be added with a secondary animation of a shell
popping off the top of a semi-automatic. Figure 14.3 shows
how such an element looks being emitted from a real gun
and shot with a high-speed shutter.
It’s definitely cool to have a detailed-looking shell pop off
of the gun, although the truth is that with a lower camera
shutter speed, the element will become an unrecognizable
two-frame blur anyhow, in which case all you need is a four-
point mask of a white (or brass colored) solid.
The bright flash of the muzzle may also cause a brief
reflected flash on objects near the gun as well as the
subject firing it. Chapter 12 offers the basic methodology: Figure 14.3 A shell pops off of the
fired gun, but it could just as well be
Softly mask a highlight area, or matte the element with
a Shape Layer with motion blur (or
its own highlights, then flash it using an adjustment layer check The DV Rebel’s Guide for a Par-
containing a Levels effect or a colored solid with a suitable ticle Playground-based setup to create
blending mode. them automatically). (Images courtesy
of Artbeats.)
As a general rule, the lower the ambient light and the
larger the weapon, the greater the likelihood of interac-
tive lighting. A literal “shot in the dark” would fully illumi-
nate the face of whomever (or whatever) fired it, just for
a single frame. It’s a great dramatic effect, but one that is
very difficult to re-create in post. Firing blanks on set or
any other means of getting contact lighting of a flash on
set would be invaluable here.
By contrast, or rather by reduced contrast, a daylit scene
will heavily dampen the level of interactivity of the light.
Instead of a white hot flash, you might more accurately
have saturation of orange and yellow in the full muzzle
flash element, and the interactive lighting might be
minimal. This is where understanding your camera and
recording medium can help you gauge the effect of a small
aperture hit by a lot of light.

Hits and Squibs
Bullets that ricochet on set are known as squib hits because
they typically make use of squibs, small explosives with the
approximate power of a firecracker that go off during the

433


take. Sometimes squibs are actual firecrackers. It is pos-
sible to add bullet hits without using explosives on set, but
frenetic gunplay will typically demand a mixture of on-set
action and post-production enhancement.
Figure 14.4 shows a before-and-after addition of a bullet
hit purely in After Effects. Here the bullet does not rico-
chet but is embedded directly into the solid metal of the
truck. In such a case, all you need to do is add the results
of the damage on a separate layer at the frame where the
bullet hits; you can paint this (it’s a few sparks). The ele-
ment can then be motion tracked to marry it solidly to the
background.
At the frame of impact, and continuing a frame or two
Figure 14.4 This sequence of frames
shows a second bullet hitting the cab
thereafter, a shooting spark and possibly a bit of smoke (if
of the truck, using two elements: the the target is combustible—not in the case of a steel vehi-
painted bullet hit and the spark ele- cle) will convey the full violence of the bullets. As with the
ment, whose source was shot on black
muzzle flash, this can vary from a single frame to a more
and added via Screen mode. (Images
courtesy of markandmatty.com.) fireworks-like shower of sparks tracked in over a few frames
(Figure 14.5).
A bullet hit explosion can be created via a little miniature
effects shoot, using a fire-retardant black background (a
flat, black card might do it) and some firecrackers (assum-
ing you can get them). The resulting flash, sparks, and
smoke stand out against the black, allowing the element
to be composited via a blending mode (such as Add or
Screen), a hi-con matte (Chapter 6), or a plug-in such as

Figure 14.5 A source spark element shot against black can be composited using Add or Screen blending mode—no matte
needed. (Images courtesy of markandmatty.com.)

434


Knoll UnMult. If dangerous explosives aren’t your thing,
even in a controlled situation, stock footage is available. If
debris is also part of the shot, however, the more that can
be done practically on set, the better (Figure 14.6).
So to recap, a good bullet hit should include
. Smoke or sparks at the frame of impact, typically lasting
between one and five frames
. The physical result of the bullet damage (if any)
painted and tracked into the scene
. Debris in cases where the target is shatterable or
scatterable
Later in this chapter, you’ll see how larger explosions have
much in common with bullet hits, which are essentially
just miniature explosions. In both cases, a bit of practical
debris can be crucial to sell the shot.

Energy Effects
There is a whole realm of pyrotechnical effects that are
made up of pure energy. At one end of the very bright
spectrum is lightning, which occurs in the atmosphere of Figure 14.6 Animating debris is
tedious and unrewarding when
our own planet daily, on the other end are science fiction compared with shooting a BB gun at
weapons that exist only in the mind (not that the U.S. mili- breakaway objects and hurling debris
tary under Ronald Reagan didn’t try). at the talent. (Images courtesy of The
Orphanage.)
A lightning simulation and a light saber composite have
quite a bit in common, in that they rely on fooling the
eye into seeing an element that appears white hot. The
funny thing about human vision is that it actually looks
for the decay—the not-quite-white-hot areas around the
hot core—for indications that an element is brighter than
white and hotter than hot.

Core and Decay
In previous editions of this book I half-joked that the
recipe for creating a filmic light saber blur was top secret.
This time around I’m motivated to spill the (pork and)
beans instead of going the quick and easy route, thanks
to the Internet superstars of the low-budget light saber,

435


Ryan and Dorkman, who have provided an entire light
saber battle on the disc (Figure 14.7).
You may find the light saber to be somewhat played out
after three decades, but the techniques you need to make
a good light saber battle apply to any other energy-driven
effect.
The Beam effect (Effect > Generate > Beam) automatically
gives you the bare minimum, a core and surrounding glow,
and it is 32 bpc and can be built up, but like so many auto-
mated solutions it’s a compromise; you don’t really think
they busted out Beam on Phantom Menace, do you?
The real thing is created by hand, and it’s not really all that
much more trouble considering how much more possibil-
ity lies in the result. You get more control over the motion
and thresholds.
Figures 14.8a through e show the basics for a single light
saber effect:
1. In the first comp, make the background plate (Figure
14.8a) a guide layer (because this is not the final comp)
and create a masked white solid. In this case, the posi-
tion and arcs of the light sabers are all hand rotoscoped
(Figure 14.8b), as detailed in Chapter 7.
Figure 14.7 Spectacular dueling
action from Ryan versus Dorkman. 2. Drop (or precomp) this comp into a new comp, apply
(Sequence courtesy of Michael Scott.) Fast Blur to the resulting layer (turn on Repeat Edge
Pixels), and set the blending mode to Add (or Screen).
3. Duplicate this layer several times, and adjust Fast Blur
so that each layer has approximately double the blur
of the one above it. With six or seven layers you might
have a Blur Radius ranging from 5 on the top level (the
core) down to 400 or so.
To automate setup you could even apply this expression
text to the duplicate layer’s Blurriness setting:
thisComp.layer(index-1).effect("Fast Blur")
➥("Blurriness")*2

This takes the Blurriness value from the layer above
and doubles it so that as you duplicate the layer, each
one below the top is twice as soft (Figure 14.8c).

436


Figures 14.8a through e The initial roto comp is set up with
generous padding (a, top left) so that masks can move out
of frame without being cut off. The roto itself is shaped to
frame the full area of motion blur, where applicable, from
the source (b, top right). The glow effect (c, middle left)
comes from layering together several copies of the roto,
each with different amounts of Feather on the mask. This
is then tinted as a single element (d, middle right) and
tweaked in Levels (e, bottom) for the proper glow intensity.

4. Drop this comp into your main composition to com-
bine it with footage and give it color (Figure 14.8d).
The Ryan versus Dorkman approach uses Color Bal-
ance and is composited in 16 bpc; one 32 bpc alterna-
tive (becauase Color Balance doesn’t work in HDR) is
simply to use Levels, adjusting Input White and Gamma
on individual red, green, and blue channels; you could
also apply Tint and Map White To values brighter than
white (Figure 14.8e).

437


That’s the fundamental setup; here are some other obser-
vations to really sell a scene like this:
. Motion blur; notice how by rotoscoping the arc of
movement and adding the edge threshold you get this
for free in the preceding ﬁgures.
. Contact/interactive lighting/glow (Figure 14.9).

Figure 14.9 You get a few things for
free: contact lighting occurs on the
face from the blue glow; it could and
should be boosted in low light. Layer
order of the sabers doesn’t matter
when they cross; either way their
values are added together.

. Physical damage/interaction with the environment;
the same types of interactions described for bullet hits
apply, so add sparks, ﬂares, and other damage to the
surrounding environment.
. Flashes/over-range values (Figure 14.10).

Figure 14.10 Flashes occur dozens of times throughout the battle; each one appears to have a unique shape, but they all
use the same four-frame flare, and its unique shape comes from being composited with the rest of the scene.

I don’t even need to tell you that these techniques are
good for more than light sabers; suppose you intend to
generate a more natural effect such as lightning. Reference
shows this to possess similar qualities (Figure 14.11) and
the same techniques will sell the effect.

438


There are a couple of built-in effects that will create light-
ning in After Effects. With either Lightning or Advanced
Lightning, you’re not stuck with the rather mediocre look
of the effect itself; you can use the same method as with the
light saber. Turn off the glow and use the effect to generate a
hard white core, and follow the same steps as just described.
It’s worth the trouble to get beyond the canned look, and it
opens all of the possibilities shown here and more.
In some cases you might go beyond these examples and
create an element that throws off so much heat and energy
that it distorts the environment around it. Figure 14.11 Actual reference images
contain energy effects with realistic
thresholding and interaction with the
Heat Distortion surrounding environment. (Image
courtesy of Kevin Miller via Creative
Heat distortion, that strange rippling in the air that occurs Commons license.)
when hot air meets cooler air, is another one of those
effects compositors love. Like a lens flare, it’s a highly vis-
ible effect that, if properly motivated and adjusted, adds
instant realism even if your viewers don’t know that hot gas
bends light.
Figures 14.12a and b show the fabricated results of heat
distortion in a close-up of a scene that will also incorporate
fire. When your eye sees heat distortion, it understands
that something intense is happening, just like with the
decay/threshold of bright lights, as described earlier. The
mind is drawn to contrast.

What Is Actually Happening
Stare into a swimming pool, and you can see displacement
caused by the bending of light as it travels through the
water. Rippled waves in the water cause rippled bending
of light. There are cases in which our atmosphere behaves
like this as well, when ripples are caused in it by the colli-
sion of warmer and cooler air, a medium that is not quite
as transparent as it seems.
As you know from basic physics, hot air rises and hot par-
ticles move faster than cool ones. Air is not a perfectly clear Figures 14.12a and b Heat haze by
itself can look a little odd (a) but it
medium but a translucent gas that can act as a lens. This
adds significantly to the realism of a
“lens” is typically static and appears flat, but the applica- scene containing a prominent heat
tion of heat causes an abrupt mixture of fast-moving hot source (b).

439


air particles rising into cooler ambient air. This creates
ripples that have the effect of displacing and distorting
what is behind the moving air, just like ripples in the pool
or ripples in the windows of an old house.
Because this behavior resembles a lens effect, and because
the role of air isn’t typically taken into account in a 3D ren-
der, it can be adequately modeled as a distortion overlaid
on whatever sits behind the area of hot air.

How to Re-create It
The basic steps for re-creating heat distortion from an
invisible source in After Effects are
1. Create a basic particle animation that simulates the move-
ment and dissipation of hot air particles in the scene.
2. Make two similar but unique passes of this particle anima-
tion—one to displace the background vertically, the other
to displace it horizontally—and precompose them.
3. Add an adjustment layer containing the Displacement
Map effect, which should be set to use the particle ani-
mation comp to create the distortion effect. Apply it to
It can be useful to generate the par-
ticles for the displacement map itself the background.
in 3D animation software, when Particle Playground is practically ideal for this purpose
the distortion needs to be attached
to a 3D animated object, such as a because its default settings come close to generat-
jet engine or rocket exhaust. The ing exactly what you need, with the following minor
distortion is still best created in After adjustments:
Effects using that map.
a. Under Cannon, move Position to the source in the
frame where the heat haze originates (in this case,
the bottom center as the entire layer will be reposi-
tioned and reused).
b. Open up Barrel Radius from the default of 0.0 to
the width, in pixels, of the source. Higher numbers
lead to slower renders.
c. Boost Particles Per Second to something like 200.
The larger the Barrel Radius, the more particles are
needed.
d. Under Gravity, set Force to 0.0 to prevent the
default fountain effect.

440


The default color and scale of the particles is ﬁne for
this video resolution example, but you might have to
adjust them as well according to your shot. A larger for-
mat (in pixels) or a bigger heat source might require
bigger, softer particles.
4. Now duplicate the particles layer and set the color of
the duplicated layer to pure green. As you’ll see below,
the Displacement Map effect by default uses the red
and green channels for horizontal and vertical displace-
ment. The idea is to vary it so that the particles don’t
overlap by changing Direction Random Spread and
Velocity Random Spread from their defaults.
5. The heat animation is almost complete; it only
needs some softening. Add a moderate Fast Blur
(Figure 14.13).

Figure 14.13 This displacement
layer, matted against gray merely
for clarity, was created with the
included steps and used with the
Displacement Map effect to create
the effect shown in Figure 14.12.

Now to put the animation to use: Drag it into the main
comp, and turn off its visibility. The actual Displacement
Map effect is applied either directly to the background
plate or preferably to an adjustment layer sitting above all
the layers that should be affected by the heat haze. Dis-
placement Map is set by default to use the red channel for
horizontal displacement and the green channel for vertical
displacement; all you need to do is select the layer contain-
ing the red and green particles under the Displacement
Map Layer menu.
Heat displacement often dissipates before it reaches the
top of the frame. Making particles behave so that their

441


lifespan ends before they reach the top of the frame is
accurate, but painstaking. A simpler solution is to add a
solid with a black-to-white gradient (created with the Ramp
effect) as a luma matte to hold out the adjustment layer
containing the displacement effect. You can also use a big,
soft mask.

Fire
Within After Effects, fire synthesis (from scratch) is way
too hot to handle; there’s no tool, built-in or plug-in, to
make convincing looking flames. If fire is at all prominent
in a shot, it will require elements that come from some-
where else—most likely, shot with a camera, although 3D
animators have become increasingly talented at fabricating
alternatives here and there.

Creating and Using Fire Elements
Figure 14.14 shows effects plates of fire elements. The big
challenge when compositing fire is that it doesn’t scale
very realistically—a fireplace fire will look like it belongs
in the hearth, no matter how you may attempt to scale or
retime it.
Fire elements are ideally shot in negative space—against
a black background, or at least, at night—so that they can
be composited with blending modes and a minimum of
rotoscoping. Fire illuminates its surroundings—just some-
thing to keep in mind when shooting.

Figure 14.14 Fire elements are typically shot in negative (black) space or occasionally in a natural setting requiring more
careful matting. By adjusting Input Black in Levels, you can control the amount of glow coming off the fire as it is blended
via Add mode, lending the scene interactive lighting for free. (Images courtesy of Artbeats.)

442


This, then, is a case where it can be worth investing in
proper elements shot by trained pyrotechnicians (unless
that sounds like no fun, but there’s more involved with a
good fire shoot than a camera rental and a blow torch).
In many cases, stock footage companies, such as Artbeats
(examples on the book’s disc), anticipate your needs. The
scale and intensity may be more correct than what you
can easily shoot on your own unless you’re pals with Mark
Pauline.

All Fired Up
Blending modes and linear blending, not mattes, are the
key to good looking fire composites. Given a fire element
shot against black (for example, the Artbeats_RF001H_
fireExcerpt.mov included on the disc and used for the
depicted example), the common newbie mistake is to try
to key out the black with an Extract effect, which will lead
to a fight between black edges and thin fire.
A first step is to simply lay the fire layer over the back-
ground and apply Add mode. To firm up a fire (or flare,
or other bright) element you can
. Ascertain that Blend Colors Using 1.0 Gamma is
enabled in Project Settings.
. Apply the Knoll Unmult (this free plug-in, included
Figures 14.15 The effect of steam (or
on your disc, makes all black areas of the image in other cases, fog) can be re-created
transparent). with a subtle Compound Blur effect.
. Fine-tune the result with a Levels effect, pushing in
on Input White and Black (as well as color matching
overall).
. Add an Exposure effect (with a boosted Exposure set-
ting) to create a raging inferno.
Compound Blur simply varies the
. Add interactive lighting for low-lit scenes (next amount of blur according to the
section). brightness of a given pixel in the
Blur Layer, up to a given maximum.
. Create displacement above the open flames (as detailed It’s the right thing to use not only
in the previous section). for fire and smoke but for fog and
mist; heavy particulate in the air
. Add an adjustment layer over the background with a
acts like little tiny defocused lenses,
Compound Blur effect, using transparency of the fire causing this effect in nature.
and smoke as a blur layer (Figure 14.15).

443


Where there’s fire there is, of course, smoke, which can
at a modest level be created with a Fractal Noise effect as
described in the previous chapter, bringing this shot home
(Figure 14.16).

Figure 14.16 All of the techniques
described here build to a result that
gives the furniture motivation to jump
out the windows.

Light Interacts
Provided that your camera does not rotate too much, a
2D fire layer, or a set of them, offset in 3D space, can read
as sufficiently three dimensional. The key to making it
interact dimensionally with a scene, particularly a relatively
dark one, is often interactive light. As was stated earlier,
fire tends to illuminate everything around it with a warm,
flickering glow.
As shown in Figure 14.17, a fire element may include a cer-
tain amount of usable glow. Input White and Input Black
in Levels control the extent to which glow is enhanced or
suppressed, respectively; you can use these controls to dial
it in and out.
Note, however, that this glow isn’t anything particularly
unique or special; you can re-create it either via a heav-
ily blurred duplicate of the source fire or using a masked
Figure 14.17 Input White and Black
and heavily feathered orange solid, with perhaps a slight
on the RGB and Red channels of
the Levels effect offer control of the amount of wiggle added to the glow opacity to cause a bit
natural glow around the element. of interactive flickering.
The better the dynamic range of the
source image, the harder you can
push this—another case for higher bit
Dimensionality
depth source. You can pull off the illusion of fully three-dimensional fire,
especially if the camera is moving around in 3D space,

444


Figure 14.18 Before-and-after sequential stills of a flyover shot. Because of the angle of the aerial camera, the shot required
3D motion tracking, in this case with 2D3’s Boujou. (Images courtesy of ABC-TV.)

directly in After Effects. I was frankly surprised at how
well this worked back when I created the shot featured in
Figure 14.18.
As shown, the background plate is an aerial ﬂyby of a for- For a shot featuring a character or
est. Because of the change in altitude and perspective, this object that reflects firelight, there’s
no need to go crazy projecting fire
shot clearly required 3D tracking (touched upon at the
onto the subject. In many cases, it
end of Chapter 8). The keys to making this shot look fully is enough to create some flickering
dimensional were to break up the source ﬁre elements into in the character’s own luminance
discrete chunks and to stagger those in 3D space so that as values, for example by wiggling the
Input White value at a low frequency
the plane rose above them, their relationship and parallax
in Levels (Individual Controls).
changed (Figure 14.19).

Figure 14.19 A top view of the 3D
motion-tracked camera from Figure
14.18 panning past one set of fires (of
which the final composition had half
a dozen). The pink layers contain fire
elements, the gray layers smoke.

445


It is easy to get away with any individual fire element being
2D in this case. Because fire changes its shape constantly,
there is nothing to give away its two dimensionality. Bor-
ders of individual fire elements can freely overlap without
being distracting, so it doesn’t look cut out. The eye sees
evidence of parallax between a couple dozen fire elements
and does not think to question that any individual one of
them looks too flat. The smoke elements were handled
in a similar way, organized along overlapping planes. As
mentioned in the previous chapter, smoke’s translucency
aids the illusion that overlapping smoke layers have dimen-
sional depth.

Explosions
The example forest fire shot also contains a large explo-
sion in a clearing. There is not a huge fundamental dif-
ference between the methods to composite an explosion
and mere fire, except that a convincing explosion might
be built up out of more individual elements. It is largely a
question of what is exploding.
All explosions are caused by rapidly expanding combus-
tible gases; implosions are caused by rapid contraction.
Just by looking at an explosion, viewers can gauge its size
and get an idea of what blew up, so you need to design the
right explosion for your situation or your result will be too
cheesy even for 1980s television sci-fi. How do you do it?

Light and Chunky
Each explosion you will see is unique, but to narrow the
discussion, I’ll organize all explosions into two basic cat-
egories. The easier one to deal with is the gaseous explo-
sion—one made up only of gas and heat. These explosions
behave just like fire; in fact, in the shot in Figure 14.20a
the explosion is fire, a huge ball of it, where something
very combustible evidently went up very quickly.
Some shots end up looking fake because they use a gaseous
explosion when some chunks of debris are needed. This
is a prime reason that exploding miniatures are still in
use, shot at high speed (or even, when possible, full-scale

446


explosions, which can be shot at standard speed). The
slower moving and bigger the amount of debris, the bigger
the apparent explosion.
If your shot calls for a chunky explosion, full of physical
debris, and the source lacks them, you need an alternate
source. Many 3D programs these days include effective
dynamics simulations; if you go that route, be sure to
generate a depth map as well because each chunk will be
revealed only as it emerges from the fireball. Many other
concerns associated with this are beyond the scope of this
discussion because they must be solved in other software.
One effect that seems to come close in After Effects is Shat-
ter, but it’s hard to recommend this unless it is specifically
a pane of glass or other plane that breaks. Shatter isn’t bad
for a decade-old dynamics simulator, but its primary limita-
tion is a huge one: It can employ only extruded flat poly-
gons to model the chunks. A pane of glass is one of the few
physical objects that would shatter into irregular but flat
polygons, and Shatter contains built-in controls for specify-
ing the size of the shards in the point of impact. Shatter
was also developed prior to the introduction of 3D in After
Effects; you can place your imaginary window in perspective
space, but not with the help of a camera or 3D controls.
A wide selection of pyrotechnic explosions is also available
as stock footage from such companies as Artbeats. In many Figure 14.20 Pyrotechnics footage is just the thing
cases, there is no substitute for footage of a real, physical when you need a big explosion filled with debris.
object being blown to bits (Figure 14.20b). (Images courtesy of Artbeats.)

In a Blaze of Glory
With good reference and a willingness to take the extra
step to marry your shot and effect together, you can create
believable footage that would require danger or destruc-
tion if taken with a camera. Even when your project has the
budget to actually re-create some of the mayhem described
in this chapter, you can almost always use After Effects to
enhance and build upon what the camera captures. Boom.
You might as well go out with a bang.

447

Index

; key, 46 24 fps footage, 20, 21, 305 All Panels workspace, 5
// (comment designator), 343 24 Pa pulldown, 20 Allow Keyframes Between
-- (decrement operator), 322 29.97 fps footage, 305 Frames toggle, 54, 58
% (modulo operator), 324 32-bit images, 373, 374–378 Alpha Add mode, 98
~ (Tilde key), 7 Alpha Bias control, 195, 196
[ ] (brackets), 45 A Alpha Channel view, 194
* key, 43 absolute time, 68–69 alpha channels. See also
(backslash) key, 46 Adapt Feature on Every Frame channels
1.0 Gamma toggle, 349, 362, option, 243, 245 adjusting settings, 18
377–378 Adaptive Sample Limit setting, cleaning up, 229
2.5D tracking, 251–252 65 considerations, xii–xxiii
2D compositions, 123 Add Keyframe option, 47 density, 96
2D layers, 122, 123, 331–333 Add mode, 95, 101–103 described, 80
2D motion blur, 65 Add transfer mode, 372 frame rate settings, 18–21
2K film plates, 24 Adjust Tension pointer, 216 fringing and, 88
3:2 pulldown, 20, 21 adjustment layers, 119–121, 159 grayscale, 107
3D calculations, 122 Adobe Encore submenu, 11 importing and, 18, 89
3D camera Adobe Media Encoder, misinterpreted, 89
mixing 2D effects, 280–281 128–129 painting, 229–232
scaling 2D layers, 287–288 Adobe Photoshop preferences, 18
smoothing moving, 252–254 brushes, 229 premultiplication and,
tracking, 250–251, 252, 264 file format, 23 87–90
3D Channel Extract effect, 410 layers, 22, 23 touching up mattes as, 230
3D compositing, 399–403 PSD files, 23 unlabeled, 89
3D frames, 10 Advanced tab, 24 viewing matte detail in, 178
3D layers .aep extension, xxiii, 15 vs. track mattes, 107
applying 2D layer as decal, .aepx extension, xxiii, 131 Alpha Cleaner, 206–207
331–333 aerender application, 126 Alpha Inverted Matte option,
fading out, 339–341 After Effects 107
importing 3D models as, coordinate systems, 328–330 Alpha Matte option, 107
282–284 Help menu, 314 alpha track mattes, 99
rendering and, 122 new features, xxiii–xxiv Alt key, 9, 26, 47
3D models, 282–284 opening multiple versions Always Preview This View
3D motion blur, 65 of, 127–128 toggle, 114
3D parallax, 251 responsiveness of, 26–29 amplitude, 314–315, 332,
3D renders, 10 time manipulation in, 68–75 343–344
3D tracking, 263–268 After Effects pipeline, 8 Anchor Point Path option, 61
8-bit color mode, 150 AfterFX.exe, 128 Anchor Point shortcut, 46
8-bit images, 349, 363 Aligned toggle, 232 anchor points, 23, 46, 61–62
8-bit pixels, 363 alignment Angle Control, 332
13.53 film, 365–366 Bézier handles, 418 Angle of View data, 275
16:9 format, 306 layers, 104, 233 angle variable, 341
16-bit color mode, 150 pixels, 232 animated vector masks, 81
16-bit pixels, 363 points, 257, 260
16-bpc projects, 148, 349

448

Index

animation Audio Settings option, 16 black levels, 141–142, 165
adding decay to, 54 audio tracks, 314–316 black pixels, 102
anchor point, 61 audio triggers effect, 341–344 Black Point control, 180
basic, 48–50 "Auto" correctors, 147 Black Softness control, 180
beat detector, 341–344 Auto Levels effect, 147 black-and-white images,
Bézier handles and, 215 Auto Select Graph Type, 50 387–388
camera, 285–288 Auto setting, 24 blacks
copying/pasting, 57 Automatic Bezier option, 56 crushed, 147–148
deformation, 226–227 auto-orientation, 285–286 current displays and, 148
following exact path, 56 Auto-Save command, 15 ﬂashing, 409
keyframe, 226 auto-tracing masks, 81 low levels of, 148
masks, 99, 212, 215, 217 AV Features tool, 41 blending colors, 377–378
paths, 52, 56 blending frames, 69–70, 118
presets, 33 B blending modes
primary vs. secondary, 422 background renders, 126–129 Add, 101–103
in real time, 226 background swatches, 232 Alpha Add, 98
reversing, 54, 61 backgrounds compositing with, 100–106
scaling, 54, 57 blue, 176, 184–193, 412 described, 82, 100–101
speed, 50 blurring, 205–206 Difference, 104
stop motion effect, 322–324 color, 30, 176, 203 Hard Light, 104
time delays, 316–318 gradients, 30–31 Hue, Saturation, and
triggering at markers, 322 green, 176, 184–193 Brightness, 105
animation path, 52 holdout mattes for, 187, 197 Light, 104
anti-aliasing, 282 matching/creating Linear, 104
aperture, 303, 370, 375 environments, 408–410 Linear Dodge, 101
application window, 4–5 multiplying into edge Luminescent Premultiply,
applications pixels, 87–89 106
memory management and, previews, 30–32 Multiply, 103
132 using with mattes, 177 Overlay, 104
QuickTime and, 358 backlighting, 156, 390–392 Screen, 101–103
text editors, 130, 131, 257 backslash () key, 46 Silhouette Alpha, 105
arguments, 254 backups, 12 Stencil Alpha, 105, 190, 391
arithmetic operators, 316 banding, 150, 349, 360, 377 Vivid Light, 104
arrays, 326–328, 335, 337–338, Beam effect, 336–337 Block Size setting, 74
340 Beam layer, 336–337 blown-out whites, 148
arrow keys, 214 beamPos variable, 336 blue channel, 180
Artbeats, xxv beat detector, 341–344 blue screens, 184–193, 412
articulated mattes, 213–216 Best quality, 70, 126 blue-screen keys, 184–193
artifacts, 20, 21, 301, 302–305 Bézier handles, 51, 93, blueScrn_mcu_
Ask User dialog, 89 213–216, 418 HD.movfootage, 187
aspect ratio, 302 Bézier masks, 93 blur. See also motion blur
asterisk (*), 43 Bézier shapes, 56, 215 backgrounds, 205–206
attributes, expressions, 315, Bézier splines, 256 boke, 275, 290–292
318 BG Renderer, 126 Box Blur, 395
audio, 16, 314–316 Bias settings, 195, 196 Compound Blur, 416, 421,
audio levels, 314, 325–327, bit-depth identiﬁer, 150 443
341–344 bitmap selection channel, 84 Fast Blur, 371–372, 391, 395
Audio Levels expression, 325, bitmapped edges, 84 foregrounds, 205–206
326, 328 black and white levels, 138 Gaussian Blur, 395

449

Index

blur (continued) Liquify effect, 419–420 Camera Settings dialog, 271
Lens Blur effect, 294–295 preset, 230 camera shots
simulating camera blur, renaming, 232 reviewing, 37
290–292 scaling, 231 stabilizing moving shots,
three-way blur, 395 Build From One Image setting, 252–254
Blur Before Difference setting, 74 studying, 37
183 bullet hits, 431, 434–435, 438 camera shutter, 64, 65
boke blur, 275, 290–292 Caps Lock, 26, 27, 222
book, organization of, xii–xxii C CC Light Burst Effect, 331
Both Channels slider, 315 cache. See also memory CC Light Rays effect, 395–396
Boujou, 42, 264 disk, 29–30 CC Particle Systems II, 330
Box Blur, 395 RAM, 29–30, 213 central processor unit (CPU),
brackets [ ], 45 camera, 271–308 28
break statement, 343 3D. See 3D camera Channel Combiner, 199–200
Bridge, 9 animating, 285–288 channels
brightness from comp space to layer alpha. See alpha channels
adjusting with Add mode, surface, 336–337 bitmap selection, 84
101–103 coordinate system, 329 blue, 180
adjusting with Brightness expression controls and, color, 32, 144–145
control, 139–140 329 green, 166, 180, 208
adjusting with Curves, fading as moving away from, matte, 73
150–157 336 red, 180
adjusting with Levels, 146 fading as turning away transparency, 80
adjusting with Screen mode, from, 339–341 Checkbox Control, 332
101–103 flat representation of, 329 child layers, 61, 62–63, 329–331
color, 105 lens blur, 290–295 Choke Matte setting, 190–191
foregrounds, 101 lens settings, 271–273 choking, 190–192, 205
gamma adjustments, mixing 2D/3D effects, chroma sampling, 198
142–143 280–281 chromatic abberation, 304–305
HSB mode, 105 projection, 288–290 CIN format, 22
Brightness & Contrast effect, push, 274, 286–288 Cinema 4D software, 266
139–140 real camera emulation, Cineon files, 360–362, 365–366,
Brightness control, 139–140 274–280 378
Brightness mode, 105 real camera settings, Cineon log space, 364–367
Brush Dynamics settings, 230 273–274 clean plate, 137, 182
Brush Tips panel, 229, 230, 232 reducing saturation away Clip Black setting, 190, 191,
Brush tool, 229 from, 333–334 193
brush-based tools, 229, 230–232 simulating camera blur, Clip controls, 197
brushes 290–295 Clip Rollback, 297
adding, 232 single-node/targeted, 266 Clip values, 197
adjusting on Brush Tips stabilizing moving camera, Clip White setting, 190, 191,
panel, 229 252–254 193
adjusting size, 231 storytelling with, 284–290 clipping, 138, 148
After Effects vs. Photoshop, as tracking tool, 249–251 clips
229 Unified Camera Tool, 285 animating timing of, 72
clone, 229, 231 virtual compared to real, cloning from, 233
defining on the fly, 231 271 duration, 69
deleting, 232 working with grain, 295–301 source, 69, 386
hardness, 231 camera report, 275–276 speed, 72

450

Index

clone brushes, 229, 231 Color Finesse, 159–160 direction, 172–173
Clone Stamp, 229 color matching, 144–145 dramatic lighting, 167–171
clones/cloning, 229, 231, gamma, 142–143 gamma slamming, 173–174
232–235 overview, 136–137 ordinary lighting, 162–166
clouds, 413–418, 423 three-way color corrector, overview, 161
Collapse Transformations 160 position, 172–173
toggle, 117, 118, 281 tools/techniques, 159–160 scenes with no clear
collapsed layers, 118, 281 color depth, 16, 150, 349 reference, 170–171
collapsed transforms, 117, 118 Color Finesse, 159–160 color mismatch, 167
Collect Files command, 13–14 Color Key, 179 Color modes, 105
Collect Source Files menu, 13 color keying, 175–209 Color Profile Converter effect,
color best practices, 176–178 357
8-bit, 150, 349 blue-screen, 184–193 color profiles, 351, 352, 357,
16-bit, 148, 150, 349 chroma subsampling, 198 378
32-bit, 374–378 color spills, 207–208 color quantization, 349
background, 30, 203 difference mattes, 182–183 color range, 100, 208
balancing, 388, 390 edge selection, 205–206 color sampling, 337–338
blending, 377–378 Extract tool for, 179–180 Color Settings option, 16
brightness, 105 green-screen, 184–193 color spaces
calibration, 350–351 Linear Color Key effect, defining, 350
color-coding layers, 42–43 180–182 encoding vs. compositing,
converting to black and matte holes, 191, 206–207 367, 370
white, 387–388 on-set effects, 202–204 HSLA, 337, 338
desaturated, 208 optimizing Keylight, linear, 370, 378
effect on cinematography, 193–201 log, 361–362, 366, 367
302, 307, 386 overview, 176 monitor, 361, 373
expressed as percentages, steps for, 184–188 RGBA, 337, 338
167 tips for creating mattes, toggling between, 374
in expressions, 337–338 176–178 YUV, 198, 199–200
foreground, 203 color management, 347–359 color spills, 178, 200, 207–208
hue, 105, 157–159, 387 bypassing, 359 color timing, 161, 389–390
light, 398–399 disabled by default, 348, 351 color wheels, 160
masks, 95, 96, 214 display management, Colorista plug-in, xxiv, 160, 389
matching. See color 354–357 comment designator (//), 343
matching enabling, 350–351, 352 comments in Timeline, 42–43
removing color matting, 90 features, 347–350 Comp Marker Bin, 46
RGB, 179–180, 196, 349, input profile, 354–357 comp markers, 43, 46
353 monitor calibration, Comp Markers option, 43
saturation, 105, 157–159, 350–351 comp space, 329
387–388 output profile, 357–358 Compander effect, 376–377
sRGB, 352 output simulation, 354–357 compositing
user interface, 129 overview, 346, 347–350 with blending modes,
color channels, 32, 144–145 project working space, 100–106
Color Control, 332 351–353 color spaces, 367
color conversion, 337–338 QuickTime, 9, 37, 358 key steps for, xii
color correction, 135–174 Color Management tab, 36 multipass, 399–403
applying via adjustment color matching, 161–174 optical, 100
layers, 159 basic technique, 162–173 overview, 83
channels, 144–145 channels for, 144–145 compositing formula, 83

451

Index

Composition panel, 6 context-clicking, 11 uses for, 150
Composition section, 34 continue loop, 262–263 vs. Levels, 150–151
Composition settings, 119 Continue Tracking option, 244 curves, logarithmic, 366–367
Composition viewer, 11, 46, contrast Custom Output Simulation
55, 114 adjusting with Curves, dialog, 355–356
compositions. See comps 152–157 Cycle Mask Colors option, 95,
Compound Blur, 416, 421, 443 adjusting with Input/ 96, 214
Compress-Expand Dynamic Output levels, 138–142 cyclorama, 202
Range preset, 377 adjusting with Levels, 146,
compression 148 D
lossless, 22, 36 input/output levels, Darken mode, 96, 97
lossy, 22 138–142 day-for-night effect, 388–389
LZW, 22, 36 Contrast value, 139–140 daylight scene, 162–166
QuickTime, 36 Convert Vertex tool, 93 Deactivate Live Update toggle,
comps coordinate systems, 328–330 26
2D, 123, 422 coplaner 3D layers, 288 decay, 54
3D, 399–403 Copy Image, xxiv, 469 decay variable, 344
analyzing with Miniflow copying decaying oscillation, 343–344
tool, 59 animations, 57 decimal values, 370
color-coding, 42–43 expressions, 313 decrement operator (--), 322
creating, 24 keyframes, 57 decrement/increment method,
embedded, 119 masks, 99 44
finding in Project panel, properties, 57 Default Spatial Interpolation to
114–115 core mattes, 192, 206 Linear option, 48
importing PSD files as, 23 Corner Pin effect, 331–333 deformation animations,
length, 45 Corner Pin tracking, 242, 226–227
locking, 114 255–261 delay variable, 317, 318
master, 112, 113, 123 Correct Luminance Changes deleting
multiple, 110–119 setting, 74 brushes, 232
naming, 113 Correspondence Points expressions, 313
nested, 59, 70–71, 116–119 (Reshape tool), 222 items from projects, 14, 15
numbering, 111 CPU (central processor unit), keyframes, 60
order, 113 28 points, 93, 217
parenthesized, 118 CRT monitor, 148 delta, 197
placing in Render Queue, crushed blacks, 147–148 demos, on DVD, xxiv
33 CRW format, 22 density, 96–97
settings, 14 Current Time Indicator, 46, depth, infinite, 413
subcompositions, 119, 123 57, 60 depth cues, 407, 408, 409
switching to layer surface, Curves, 150–157 depth map, 410
336–337 adjusting brightness, depth of field, 274, 290–292,
timing, 117 150–157 294
trimming duration to work adjusting contrast, 152–157 Despill Bias, 189, 196
area, 45 adjusting gamma, 143, 150, Despot cleanup tools, 199–200
concatenation, 119 152–157 difference mattes, 182–183
Confidence graphs, 355 adjusting gradients, 153 Difference mode, 95, 104, 233
Confidence settings, 244–245 drawbacks, 150–151 diffraction effect, 416
Consolidate All Footage gamma slamming, 173–174 digital film, 365–367
command, 14–15 S-curve adjustment, 156 Directional Blur setting, 67
Constant mode, 231 single point adjustments in, disk cache, 29–30
154, 156–157

452

Index

displacement map, 440–441 E Effects & Presets panel, 33
display. See also monitors; Effects Controls panel, 6
ease(), 334, 336, 341
screen effects plug-ins, 33
easeIn(), 335
black detail and, 148 elements
easeOut(), 335
multiple, 7 context-clicking, 11
Easy Ease functions, 50–52
resolution, 25 duplicating, 45
Edge Feather control, 179
settings, 25–26, 129 grouping, 111
edge mattes
setup, 7 renaming, 41
creating for edge selection,
Display Style option, 16 rotating, 422–423
205–206
dist variable, 334 showing in layers, 54
protecting edge detail with,
distortion else clause, 323
177
heat, 439–442 Enable Disk Cache option,
edge multiplication, 87–88, 89
lens, 275, 276–280 29–30
Edge Thin control, 179
Liquify effect, 419–420 Enable Motion Blur toggle,
edges
Mesh Warp effect, 418–419 67, 72
bitmapped, 84
with Reshape, 219–224 Enable OpenGL toggle, 27
feathered, 84, 179
dividers, 5 encoding color spaces, 366–367
finding, 205
docking zones, 6–7 End key, 43, 44
fringing in, 90
dot(), 340 energy effects, 435–439
refining, 187
dot notation, 315 environment
Edit Original command, 12
dot product, 340 creating, 408–410
editing
double-matte method, 197 matching, 408
anchor points, 61
downsampling, 24 environment reference, 204
Boot.ini file, 132
DPX format, 22, 360–362, 378 Eraser tool, 229
expressions, 313
Draft quality, 24, 70, 125 Erodilation plug-in, xxiv, 205,
masks, 214, 217
drawing 469
preferences, 131
masks, 91–92, 99, 214, 216 Error Threshold setting, 74
temporal edits, 233
points, 216 errors
E/EE keyboard shortcuts, 58
drop zones, 6 expressions, 11, 313
effects. See also filters; specific
duration premultiplication, 90
effects
layers, 69 rendering, 122
accessing in Effects &
source clips, 69 rounding, 288
Presets panel, 33
trimming to Work Area, 45 write-errors, 37
animation presets, 33
Duration setting, 231 Excel, 57
applying to layers, 33
dust busting, 235 Excel spreadsheets, 57
audio triggers, 341–344
dust removal, 235 Expansion property, 227
described, 33, 83
DV format, 348, 354 explosions, 430, 431, 446–447
energy, 435–439
DVCPRO HD format, 198 Exposé feature, 50
explosions, 430, 431,
DVCPRO50 format, 198 exposure, 275, 375–376
446–447
DVD, included with book, Exposure control, 370, 375–376
firearms, 430–435
xxiv–xxv Expression Controls, 332
Glow, 121
dynamic range, 362–364 Expression menu, 312, 313
heat distortion, 439–442
Cineon files, 360, 365 expressions, 309–344
incompatible, 376–377
Compress-Expand Dynamic attributes, 315, 318
in layers, 123
Range preset, 377 audio triggers effect,
linking to properties,
low, 370, 373 341–344
314–316
on monitor, 347 color in, 337–338
resetting, 140
overview, 362–363 coordinate systems, 328–330
smoke, 103
copying, 313
study of, 37
creating, 312–313

453

Index

expressions (continued) Extract tool, 179–180 missing items, 12
default, 312 Extrapolate Motion option, presets, 33
deleting, 313 244–245 in Project panel, 114–115
disabling, 313 eyedropper tool, 188 text strings, 12
editing, 313 unused source, 12
effects tracks parented F used source, 12
layers, 330–331 fades, 336, 339–341 fire, 442–446. See also smoke
enable/disable toggle, 312, Fast Blur, 371–372, 391, 395 firearms, 430–435
313 feathered mattes, 98 First Vertex, 93, 100, 214, 224
errors, 11, 313 feathering flag of Mars, 387
exposing, 313 alpha selections, 84–86 Flame, 70
fades, 336, 339–341 edges, 84, 179 flares, 275, 392–394
Graph Editor toggle, 312 masks, 92 Fleischer, Max, 212
inserting text, 312 feathering effect, 217 floating panels, 7
jittery motion, 324 .ffx file suffix, xxiv floating point files, 363, 373,
keyframes and, 311, 314 Field Of View (FOV), 314–315 376
layer index, 316–318 fields, interlacing, 19–20 floating point TIFF format, 376
layer space transforms, file extensions, 12 Flowchart view, 59, 113
328–337 files fog, 413–417
limitations, 311 Cineon, 360–362, 365–366, folders
linking effects to properties, 378 grouping elements with,
314–316 collecting for footage, 111
links, 314–316 13–14 organizing projects with, 14,
looping keyframes, 318–320 floating point, 363, 373, 376 111–112
markers in, 320–324 missing, 12 source, 112
methods, 316 naming conventions, 36 watch, 128
overview, 310–311 PSD, 23 footage. See also camera shots
pasting, 313 QuickTime, 37 24 fps vs. 29.97 fps, 305
pasting text, 312 RPF, 403, 410 collecting files for, 13–14
pickwhip, 312, 315, 332 source, 11–15 consolidating, 13–15
properties, 315 fills, 94 HD. See HD footage
reducing layer saturation, film HDTV, 353, 368
333–335 Cineon and, 360 importing into After Effects,
resources for, 344 digital, 365–367 8–11
Rotation, 318 exposure to light, 367 importing still sequences
text layer formatting and, Film Fix, xxiv for, 19
311 film formats, 306–307 importing with
time in, 327–328 Filtering setting, 74 premultiplied images,
time remapping, 324–328 filters, 386–389 87–88
using tracking with, final output, 36, 112 information about, 19
261–263 final renders, 46 interpretation of, 17–21,
values, 311 Find Edges, 205 87–90
vectors, 340 Find Missing Footage option, missing, 11–12
wiggle effect, 325–327 12 plate, 24
EXR format, 22, 374, 376 finding items raw, 33–36
EXR images, 403 comps, 114–115 reloading, 11–12
Extensible Markup Language effects, 33 removing unused, 14, 15
(XML), 131–132 footage, 12 rendering, 33–36
extensions, 12 keyframes, 58 replacing, 11

454

Index

searching for, 12 keyframes between, 54 glints, 393, 394
sequences for, 19 missing, 10 Global Smoothness setting, 74
source, 11, 14, 17, 22–24, from multiple sequences, 9 Glow effect, 121
23, 112 navigation, 43–44, 46 gradients
Force Alphabetical Order overlays, 70 adjusting with Curves, 152,
option, 10 overwriting, 19 153, 155
foreground swatches, 232 playing only frames with backgrounds, 30–31
foregrounds markers, 322–324 banding in, 150
blurring, 205–206 selecting multiple, 9 black to white, 138–139
brightness, 101 skipping, 24–25 Ramp, 144
color, 203 snapping to, 54 grain, 295–301
matching/creating subsets, 9 management of, 295–301
environments, 408–410 Frames Per Second setting, 19 noise as, 299
mattes with, 177 framesToTime(), 343 removal of, 299–300
rotoscoping features in, 187 Free Transform mode, 92 graininess, 183, 189
formats. See also specific formats Freeze Frame option, 69, 71, Graph Editor, 47–56
film, 306–307 183 animation in, 48–57
non-square pixel, 19, 20–21 freq variable, 344 Confidence graphs, 244
raster image, 22 frequency parameter, 325–326 controlling keyframes in,
source, 18–19 fringing, 18, 88, 90, 200 47–56
source footage, 22–24 From Current Time option, 30 controls, 47–48
video, 306–307 fromCompToSurface(), 336, 337 copying/pasting
virtual cinematography, 305 Full Screen option, 30 animations, 57
FOV (Field Of View), 314–315 full-screen mode, 7, 30 Easy Ease functions, 50–52
FOV parameter, 314–315 Furnace plug-ins, 70, 72 enabling, 47–48
Fractal Noise effect, 104, fxphd.com site, xxv expressions, 312
414–416 Graph Options menu, 50
fractal noise patterns, 104 G Graph View, 48–57
frame blending, 69–70, 72, 118 gamma Hold keyframes, 55
Frame Mix mode, 70 1.0 Gamma toggle, 349, 362, Show Properties menu, 48
frame rates 377–378 Transform Box, 54
adjusting, 18–21, 68–69 adjusting with Curves, 143, U/UU shortcuts, 58–59
maintaining, 119 150, 152–157 vs. Layer view, 57–58
misinterpreted, 18–19 adjusting with Levels, Graph Options menu, 50
QuickTime, 19 142–143, 146 Graph View, 48–57
speed and, 305–306 described, 143 gray card, 369
FrameCycler, 37 monitors, 368–369 gray matching, 162
frames QuickTime, 348–349, 358 grayscale alpha channels, 107
adding eases to, 50–52 spiking and, 148 grayscale gradient, 138
averaging, 69–70 video gamma space, grayscale ramp, 369
bad, 8, 19 368–369 grayscale transparency channel,
cloned, 232, 233 gamma boost, 143 146
creating new images from, Gamma control, 142–143 green channel, 166, 180, 208
69–70 gamma slamming, 173–174 green screen keys, 184–193
display of, 16, 17 gamut, 353 green screens, 179, 184–193
effect of time measurement garbage mattes, 81, 185, 190, Gridiron Flow, 14
on, 68–69 191, 206 grids, 30, 31, 265
filling with layers, 46 Gaussian Blur, 395 Grids & Guides preferences, 31
freezing, 69, 71 gaussRandom(), 328 grouping zones, 7

455

Index

Grow Bounds effect, 118 raised gamma, 153 virtual, 83
Guess button, 18, 87, 89 histograms Import & Replace Usage
guide layers, 30–31, 119, Extract tool, 179–180 option, 124
121–122 gamma adjustments and, Import dialog, 9, 12
guides, 31 143 Import option, 124
guns, 430–435 Levels, 138, 145–150 importing
problem solving with, 3D models, 282–284
H 148–150 3D tracking data, 263–268
hack shortcuts, 130–131 Hold keyframes, 55, 71 with alpha channels, 87–88,
halation, 275 Hold option, 56 89
halos, 98, 391, 394, 395 holdout masks, 201, 259 alpha channels and, 18
Hand tool, 25 holdout mattes, 186, 187, 201 footage into After Effects,
handles, Bézier, 51, 93, Home key, 43, 44 8–11
213–216, 418 HSB mode, 105 footage with premultiplied
Hard Light mode, 104 HSLA array, 338 images, 87–88
HD display, 353 HSLA (hue, saturation, images, 90
HD footage lightness, and alpha) color Maya scenes for 3D
examples, xxv space, 337, 338 tracking, 265–268
multiprocessing and, 24 hue, 105, 157–159, 387 preferences for, 89
pixel aspect ratio and, 21 hue, saturation, lightness, and PSD files as compositions,
resolution, 24 alpha (HSLA) color space, 23
HD format, 276, 306–307 337, 338 sequences, 8–10, 119
HDR (high dynamic range) Hue mode, 105 source files, 8, 9–10
described, 346 Hue/Saturation control, still sequences for moving
linear floating-point HDR, 157–159, 206–208, 387, 388 footage, 19
369–378 hue-shift, 207, 208 Increment and Save command,
linearized working space, 15
373, 377 I increment/decrement method,
Merge to HDR function, idx variable, 323–324 44
204, 364 ifstatement, 321, 323 index attribute, 316–318, 320

source, 373–383 Image Lounge, xxiv infinite depth, 413
HDR Compander, 376–377 image pipeline, 14 Info palette, 37, 164
HDR images, 204, 363–364 image planes, 291–292 Info panel, 188
HDR pipeline, 374, 376, 377 image sequences, 8–9 In/Out points, 44, 45, 69, 335
HDTV footage, 353, 368 images Input Black controls, 138–142
HDTV profile, 353 8-bit, 349, 363 input profile, 354–357
HDTV standard, 306–307 32-bit, 373, 374–378 Input White controls, 138–142
HDV format, 198, 273 black-and-white, 287–388 Instant HD tool, xxiv
headroom, 148 downsampling, 24 interlaced fields, 19–20
heat distortion, 439–442 EXR, 403 interpolation, 48, 222, 334–335
Help menu, 314 HDR, 204, 363–364 Interpret Footage dialog,
hero shots, 161 importing, 89 17–21, 87–90
Hide Layer Controls command, layering over opaque interpretation rules, 355
32 background, 86–87 Intersect mode, 95
high-contrast mattes, 79, morphing, 219–225 inverse square law, 385
179–184, 212 panning around, 61 Invert effect, 140
highlights plate, 137 Invert Selection, 45
color matching, 164–166 RGB, 179–180 inverted layers, 45, 96–98
gamma slamming and, sharpening, 299, 301 Inverted toggle, 95
173–174 still, 19

456

Index

J effect of time measurement Knoll Light Factory Unmult,
on, 68–69 469
J key, 60
expressions and, 311, 314 Kodak 5218 profile, 355
JavaScript
finding, 58 Kronos tool, 72, 235
arithmetic operators, 316
between frames, 54
arrays, 326–328, 335,
337–338, 340
hiding/showing, 32, 60 L
history of, 213 Label Defaults command, 42
comments, 343
Hold, 55, 71 laptop computers, 43
decrement operator, 322
linear, 55 Layer Controls, 32, 332
described, 314
looping, 318–320 layer markers, 43
if statement, 321, 323
mask, 98–99, 217, 218 Layer Markers option, 43
resources, xxiv
navigation, 45, 60 Layer menu, 11
variables, 317
nudging forward/backward, layer space transforms,
jitter, 324
45, 60 328–337
JPEG format, 22
offsetting values of, 60–61 Layer Switches tool, 41
K Opacity, 222 Layer view, 57–58
roving, 56 layers
K key, 60 selecting, 60 2D, 122, 123, 331–333
Keep Colors option, 182 selecting multiple, 60 3D. See 3D layers
Key Colors setting, 182 setting, 60 adding at specific time, 44
Key Correct Pro, xxiv shortcuts, 60 adding tags to, 42–43
Key Correct tools, 206–207 Time Remap, 71, 72 adding to composition, 45
keyboard shortcuts. See also keying. See color keying adjustment, 119–121, 159
shortcuts Keylight, 188–201 aligning, 104, 233
brush tools, 231, 232 basic steps, 184–188 applying motion tracks to,
context-clicking, 11 Bias settings, 195, 196 218
display resolution/size, 25 blue/green screens, applying presets to, 33
E/EE, 58 184–193 Beam, 336–337
J key, 60 choking, 190–192, 200, 205 centering, 46
K key, 60 clean-up tools, 197–201 child, 61, 62–63, 329–331
keyframe navigation, 60 decision-making collapsed, 118, 281
layers, 45–46 comparisons, 194 color-coding, 42–43
maximizing screen, 7 fringing, 200 context-clicking, 11
RAM previews, 30 noise suppression, 198–200 coplaner 3D, 288
Timeline, 43–46 optimizing, 193–201 duplicating, 45
Timeline navigation, 43–44 Preblur option, 198–200 duration, 69
transforms, 46–47 Screen Balance, 189, 195, effects within, 123
U/UU, 58–59 196 filling frames with, 46
keyed layers, 189 Screen Gain, 189, 195 guide, 30–31, 119, 121–122
KeyEd Up script, 131 screen matte generation, hiding/showing, 42
Keyframe Assistant, 61, 314 194–201 index attribute, 316–318
keyframe transitions, 56 spill suppression in, 200–201 inverted, 45, 96–98
keyframes three-pass method, 190–193 keyboard shortcuts for,
adding, 47 tips for using, 193–201 45–46
animating, 226 uses for, 184 keyed, 189
controlling in Graph Editor, vs. Primatte Keyer, 209 length, 45
47–56 Keys tool, 41 Live 3D, 23
copying/pasting, 57 Knoll Light Factory, xxiv, 303, locked, 42
deleting, 60 393 mattes, 42, 98
deselecting, 60
dropped, 319

457

Index

layers (continued) histograms, 145–150 linear blending, 349, 373,
moving up/down in stack, overview, 137 377–378
45 resetting, 140 Linear Color Key, 179, 180–182
nudging forward/backward vs. Curves, 150–151 linear floating-point HDR,
in time, 45 Levels (Individual Controls) 369–378
null, 42 control, 144 linear keyers, 179–184
offset, 63 light, 381–403 linear keyframes, 55
opacity, 86–87, 325–328 adding full-color tint for linear keys, 179–184
order of, 44, 104, 108 film, 386–388 Linear mode, 104
parent, 62–63, 329–331 adjusting color with no Linear option, 56
parenting to single layer, 61 clear reference, 170–171 linear pixels, 370
Photoshop, 22, 23 backlighting, 156, 390–392 linearized working space,
pixel, 120 CC Light Rays effect, 101–102, 103, 373, 377
precomping, 113, 116–117 395–396 links, 314–316
properties, 122 color, 398–399 Liquify effect, 419–420
reducing saturation, color keying and, 203 Live 3D layers, 23, 28
333–335 color timing effects, Live Update, 26, 27
replacing, 45 389–390 LME (Local Motion
selecting, 45 day-for-night effect, 388–389 Estimation) technology, 73
Shape, 91, 94, 121 daylight, 162–166 Local Motion Estimation
showing items in, 54 diffuse, 203 (LME) technology, 73
shy, 42 dramatic, 167–171 Local Smoothness setting, 74
soloing, 42 exposure to film, 367 Lock icon, 114
speed, 48 flares, 392–394 Lock Source Time, 233
splitting, 45 glints, 393, 394 locking
target, 99 God rays, 394–396 comps, 114
timing, 117 halos, 98, 391, 394, 395 layers, 42
transform properties, 46–47 hard vs. soft, 383 masks, 96, 214
unlocking, 42 hotspots, 384–385 log color space, 361–362, 366,
using as lens filters, 386–389 indirect, 399, 400, 401 367
Layer/Source tool, 41 inverse square law, 385 logarithmic curve, 366–367
length(), 334 light wrap, 390–392 loopIn(), 318–319
lens angle, 274 location/quality of, 383 loopInDuration(), 320
lens artifacts, 301, 302–305 multipass 3D compositing, looping
Lens Blur effect, 294–295 399–403 continue loop, 262–263
lens distortion, 275, 276–280 neutralizing direction of, keyframes, 318–320
lens filters, 386–389 384–385 previews, 30, 37
Lens Flare effect, 336 ordinary, 162–166 while() loop, 343
lens flares, 302–303, 336, reflected, 393–394, 399–403 loopOut(), 318–319
392–394 scattered, 394–396 loopOutDuration(), 320
lens settings, 271–273 shadows and, 383, 384, lossless compression, 22, 36
Lenscare, xxiv, 294, 468 396–399 lossless output, 36
lesson files, installing, xxiv source/direction of, lossy compression, 22
Levels, 137–160 382–385 low-loss output, 36
adjusting gamma with, volumetric, 394–396 Luma Inverted Matte option,
142–143, 146 Light mode, 104 107
bracketing Input levels, light sources, 390–399 Luma Key, 179–180
147–148 Lighten mode, 96, 97 Luma Matte option, 107
contrast adjustment with, Lightning effect, 435, 438–439 luma mattes, 107, 180, 183–184
138–142 linear(), 334–335 luminance, 179

458

Index

Luminescent Premultiply masking motion blur, 218–219 master comps, 112, 113, 123
mode, 106 masks, 91–100 match moving, 268
LZW compression, 22, 36 adding to images, 95 Matching Softness setting,
animated, 81, 99, 212, 215, 180–181
M 217 Matching Tolerance
Mac OS–based systems arranging, 91 eyedropper, 181
F9 key caution, 50 auto-tracing, 81 Math.abs(), 326

forcing a crash, 132 Bézier, 93 Math.acos(), 341

incompatible formats, 12 color, 95, 96, 214 Math.cos(), 344

memory and, 132 combining selections, 95–98 Math.exp(), 344

rendering and, 126, 128, contracting, 92 Math.floor(), 324

130–131 copying/pasting, 99–100 Math.Log(), 326, 327

macros, xxiii. See also scripts darkening, 96, 97 Math.min(), 323

Magic Bullet Colorista, xxiv, deactivating, 95 Matte Channel control, 73
160, 389, 469 density, 96–97 matte channels, 73
Magic Bullet Looks, 159–160, described, 80, 91 Matte Choker, 205
469 determining points with Matte Layer control, 73
magnification settings, 24 First Vertex, 224 mattes
Make Template option, 36 disabling, 91, 95 alpha track, 99
markers drawing, 91–92, 99, 214, 216 articulated, 213–216
comp, 43, 46 editing, 214, 217 avoiding holes in, 191,
in expressions, 320–324 expanding, 92 206–207
layer, 43 feathering, 92 choked, 190–192, 205
playing only frames with garbage matte, 81 core, 192, 206
markers, 322–324 hiding/showing, 32, 96, 214 described, 79
replacing numbers with holdout, 201, 259 Despot cleanup tools,
names, 46 improving, 201–208 199–200
snapping to, 54 lightening, 96, 97 detecting flaws in, 177, 194
triggering animation with, limitations of, 216–219 difference, 182–183
322 locking, 96, 214 double, 197
Mask Color swatch, 96 masking motion blur, feathered, 98
Mask Expansion option, 92 218–219 garbage, 81, 185, 190, 191,
Mask Feather property, 92 moving, 99–100 206
Mask Interpolation command, multiple, 96, 213, 214 for green/blue screens,
100 names, 221 184–193
mask keyframes, 98–99, 217, opacity, 95–96 high-contrast, 79, 179–184
218 organizing, 96 holdout, 186, 187, 201
Mask Opacity property, 95–96 overlapping, 95–98, 213, 214 improving, 193
Mask Path channels, 99 overlapping transparency, layers, 42, 98
Mask Path keyframes, 99 96–98 luma, 107, 180, 183–184
Mask Path properties, 99 placing in motion, 98–100 problems with, 193–194,
Mask Path stopwatch, 218 removing from images, 95 204–208
mask paths, 92, 93, 99 roto, 187 protecting edge detail, 177
Mask Shape dialog, 92 scaling, 92 refining, 196–201
Mask Shape tool, 91–92 selections with, 80–81 screen, 194–201
mask shapes, 91–92, 216, 218 setting up to track, 217, 218 sharing, 107
mask tracking, 217, 218, shapes, 91–94, 214, 215, 218 softening, 198
247–248 target, 99, 220, 224 tips for, 176–178
mask vertices, 215, 311 transparent, 95, 96 track, 106–108
vector, 81

459

Index

maxAngle variable, 339, 341 motion tracking, 237–268 relocating, 245
maxOffset variable, 328 2.5D tracking, 251–252 showing in Timeline, 244
Maya 3D scenes, 265–268 3D camera, 250–251 mouse, 11, 25, 214
Media Encoder, 128–129 3D tracking, 263–268 movies, creating with Output
MediaCore, 348, 354 Adapt Feature option, 245 Modules, 35–36
memory AE camera as tracking tool, muliprocessing, 132
caching, 29–30, 213 249–251 multidimensional properties,
management of, 132 Confidence settings, 327
preferences, 29 244–245 multipass rendering, 399–403
Memory & Cache, 29–30 continue loop, 262–263 Multiply mode, 103, 372
Merge to HDR function, 204, Continue Tracking option, Multiprocessing option, 28
364 244 multiprocessing settings, 28–29
mesh, 226, 227 Corner Pin tracking, 242,
Mesh Warp effect, 418–419 255–261 N
metadata, 36 essentials, 239–249 naming conventions, 36
methods, expressions, 316 with expressions, 261–263 navigation
midtones, 162, 166 Extrapolate option, 244–245 frames, 43–44, 46
minAngle variable, 339, 341 feature region, 240, 241 keyframes, 45, 60
Miniflow tool, 59, 113, 114 matching motion blur, Timeline shortcuts, 43–44
Minimal workspace, 5 248–249 nearestKey(), 320
Minimax, 204 matching multiple objects, negatives, film, 360, 364–365
mist, 413–417 249–252 nesting
Mocha X-splines, 256–257, 259 MochaAE, 255–261 compositions, 59, 70–71,
MochaAE, 255–261 nulls, 245–248 116–119
modulo operator (%), 324 overview, 238–239 preserving resolution, 119
monitors. See also display point tracking, 239–249 time, 118–119
color calibration, 350–351 rotation/scale data for, network rendering, 127
CRT, 148 242–244 New Workspace command, 6
external broadcast, 351 rotoscoping and, 212 node view, 59
Mac vs. PC, 368 search regions, 240, 241 noise
maximizing screen, 7 setting up to track masks, blurring and, 294–295
multiple, 7 218 Fractal Noise effect, 104,
setup, 7 smoothing moving camera, 414–416
video space of, 368–369 252–254 as grain, 299
morphing, 219–225 stabilizing moving shots, reducing with spill
morphing tools, 217 252–254 suppression, 201
motion Stop Tracking option, 244 noise suppression, 198–200
jittery, 324 subpixel motion, 244 None mode, 95
placing masks in, 98–100 SynthEyes 3D, xxiv, 42, non-square pixel formats, 19,
motion blur, 64–65. See also blur 266–268 20–21
2D, 65 TrackerViz, 217, 218, Normality tool, 403
3D, 65 247–248 Notepad, 130
described, 64 types of tracks, 241–242 NTSC DV format, 198
edge multiplication and, 88 motion tracks NTSC format, 305
masking, 218–219 applying to layers, 218 Nuke, 70
settings, 64–67 averaging, 247 null layers, 42
tracking and, 248–249 locking, 245 nulls, 245–248
motion paths mask tracking, 247–248 numbers
controlling in viewer, 55 offsetting, 245 random, 327, 328
hiding/showing, 32 version, 37

460

Index

numeric keypad, 45 Output Modules, 34, 35–36 Pen tool, 91, 93, 216
numLayersattribute, 326 output profile, 357–358 performance, 26–27
Over Range preset, 361 Photo-JPEG, 36
O overlay grids, 31 Photoshop
offset layers, 63 Overlay mode, 104 CS4 3D version, 282–284
offsetAngle variable, 318 file format, 23
online review, 36 P footage, 235
opacity Page Up/Page Down keys, 29, importing 3D models as 3D
fades, 336, 339–341 43–44 layers, 282–284
layers, 86–87, 325–328 Paint panel, 229 layers, 22, 23
masks, 95–96 paint strokes, 212, 230–232 PSD files, 23
overlapping, 86 Paint Workspace, 229 Photoshop video, 235
settings for, 86 painting pickwhip, in expressions, 312,
Opacity expression, 326, 328, basics, 229–232 315, 332
334 guidelines for, 212 pins, 226–227
Opacity keyframes, 222 Photoshop vs. After Effects, pipelines
Opacity shortcut, 47 229 After Effects, 8
opaque pixels, 106 uses for, 212, 229 HDR, 374, 376, 377
OpenEXR format, 376 vector paint tools, 230 image, 14
OpenGL options, 27 Pan Behind tool, 47, 61–62 optimizing, 109–132
operators, JavaScript, 316 panel groups, 4–5, 7 render, 122–129
optical compositing, 100 panel tabs, 11 pivot points, 61
optical flow, 70 panels, 4–8. See also specific Pixel Aspect Ratio (PAR),
Optics Compensation effect, panels 20–21
278–280 Pangrazio, Mike, 44 Pixel Corps, xxv
optimizing panning, 61 pixel layers, 120
Keylight, 193–201 PAR (Pixel Aspect Ratio), Pixel Motion mode, 70, 72–73
output, 36 20–21 pixels
pipeline, 109–132 parent layers, 62–63, 329–331 8-bit, 363
previews, 123–126 Parent tool, 41 16-bit, 363
projects, 129–132 parent-child relationships, 61, black, 102
renders, 123–126 62–63 Cineon log encoding,
Option key, 25, 26, 47, 63 parenthesized comp, 118 366–367
Options section, 34 parenting layers, 61 dimensions, 117
orientation, 274, 285–286 Particle Playground, 410 fringing, 200
Out points. See In/Out points particles, 415, 423, 425–427, linear, 370
output 439–440 multiplying background
final, 36, 112 particulate matter, 406–410 into edge, 87–89
linear floating point and, pasting non-square formats, 19,
378 animations, 57 20–21
lossless, 36 expressions, 313 opaque, 106
low-loss, 36 keyframes, 57 preserving positions, 232
online review, 36 masks, 99, 100 semitransparent, 106
optimized, 36 properties, 57 square, 19, 21, 84
simulation of, 354–357 paths threshold, 84
via Render Queue, 33–36 animating, 52, 56 tracking motion of, 70
Output Module settings, 35–36, mask, 92, 93, 99 white, 102
88, 123–124 motion. See motion paths plate footage, 24
pen, stylus, 230 plate image, 137

461

Index

plug-ins Preserve Underlying properties. See also attributes,
effects, 33 Transparency option, expressions
on included DVD, xxiv, 105–106 copying/pasting, 57
468–469 presets layers, 122
PNG format, 22 animation, 33 linking effects to, 314–316
Point Controls, 331–332, 333 brushes, 230 multidimensional, 327
point of interest, 285–286 clones, 233 pre-expression values, 318
point tracking, 239–249 keyframe transitions, 56 selecting all keyframes for,
points settings, 33 60
adding, 93 Presets folder, 33 showing in Graph Editor, 48
adding to masks, 217 preview settings, 30 ProRes 422 format, 198
animated masks, 217 previews, 24–32 proxies, 124–126
deleting, 93, 217 backgrounds, 30–32 PSD files, 23. See also Adobe
drawing, 216 Draft quality and, 24 Photoshop
moving, 93, 214 first-pass keys, 186, 187 Pulldown, 20–21
rotating, 214 looping, 30, 37 Puppet tools, 225–228
scaling, 214 optimizing, 123–126 push, 274, 286–288
transforming, 214 preferences, 26–27, 30 PV Feather, xxiv, 217, 469
Position expression, 327 RAM. See RAM previews pyrotechnics, 430–435
Position property, 61 speeding up, 24–25
Position shortcut, 47 updating external devices Q
post-effect flag parameter, 337 for, 30 quality/resolution settings,
Posterize Time option, 70, 119 video, 30 24–26
Post-Render Actions, 36 Primatte Keyer, xxiv, 209 quantization, 119, 148–150
precipitation, 423–427 profiles, color, 351, 352, 354, QuickTime
precomping, 113, 114, 116–117 378 challenges, 19, 37
preferences Project panel, 6, 8, 11, 111, color management issues, 9,
described, 45 114–115 37, 358
display, 129 Project Settings dialog, 16–17 frame rates, 19
general, 129 projects. See also comps gamma, 348–349, 358
grids/guides, 31 analyzing, 59 low-loss output, 36
imports, 89 automatically saving, 15 write-errors, 37
memory, 29 backups of, 12 QuickTime files, 37
previews, 26–27, 30 combining, 13, 14 QuickTime Player, 37
projects, 129–132 considerations, xxiii
restoring to defaults, 129 consolidating, 13–15 R
setting, 129–132 copying, 12
rack focus shots, 290–292
text, 130–131 hack shortcuts, 130–131
Radiance format, 376
UI colors, 129 moving, 13
radiansToDegrees(), 341
undo levels, 129 naming conventions, 36
radiometrically linear color
premultiplication, 87–90 optimizing, 129–132
data, 373
Pre-render option, 126 preferences, 129–132
rain, 423–427
pre-rendering, 123, 124–126 removing items from, 14, 15
RAM cache, 29–30, 213. See also
Preserve Constant Vertex renaming, 41
cache; memory
Count option, 214 replacing, 124
RAM previews
Preserve Frame Rate toggle, reverting to previous, 15
caching, 29–30, 213
119 saving, 15
customizing, 30
Preserve Resolution When settings for, 15–17
full-screen mode, 30
Nested toggle, 119 templates, 111–112
keyboard shortcut for, 30
versions, 15, 36, 37

462

Index

options for, 30 Adobe Media Encoder, rig removal, 234, 235
performance, 129 128–129 ROI (Region of Interest), 24,
resolution/quality and, in background, 126–129 26
24–25, 24–26 final renders, 46 Rosco Ultimatte colors, 195
skipping frames, 24–25 footage, 33–36 Rotate tool, 47
specifying output device guide layers and, 119, rotation
for, 30 121–122 camera, 285–286
switching to alpha channels, multipass, 399–403 elements, 422–423
194 network renders, 127 motion tracking, 242–243
Ramp controls, 146 optimizing renders, points, 214
Ramp effect, 30, 138–139, 146 123–126 Rotation expression, 318
Ramp gradient, 144 post-render options, Rotation property, 318
random(), 327, 328 123–124 Rotation shortcut, 47
raster image formats, 22 pre-renders, 123, 124–126 Rotobezier shapes, 215–216
raw footage, 33–36 proxies, 124–126 rotoscoped shapes, 251
Raw tracks, 242 raw footage, 33–36 rotoscoping, 211–236
rd_MergeProjects.jsx script, 14 render order, xxii, 108, articulated mattes and,
Record Options, 226 122–123 213–216
red, blue, green, and alpha Render Queue, 33–36 automatic, 81
(RGBA) color space, 337, 338 settings for, 34–36 cloning and, 229, 232–235
red channel, 180 showing process of, 123 color keying and, 187
redefinery, 468 Watch Folder for, 128 complex shapes, 213
Reduce Project command, 14 Repeater, 94 described, 80, 212
ReelSmart Motion Blur plug-in, Replace Footage option, 11 dust removal, 235
xxiv, 68 Reset Levels option, 140 guidelines for, 212
Reference graph, 50 Reshape tool, 219–224 history of, 212
reflected light, 393–394, resolution masking motion blur,
399–403 Auto setting, 24 218–219
Region of Interest (ROI), 24, display, 25 morphing, 219–225
26 half, 24–25, 35 Puppet tools, 225–228
Reload Footage command, 12 HD footage, 24 setting up to track masks,
remapping expressions, keyboard shortcuts, 25 218
324–328 nesting and, 119 wire removal in, 234–235
Remove Color Matting option, RAM previews, 24–25 working with Rotobezier
90 Reveal Composition in Project shapes, 215–216
Remove Unused Footage option, 115 roving keyframes, 56
command, 14, 15 Reveal Expression Errors RPF files, 403, 410
Render Multiple Frame option, 11 rulers, 31–32
Simultaneously option, 28 Reveal Layer Source in Project Rush Render Queue, 127
render pipeline, 122–129 option, 11, 115
Render Queue, 33–36, 119 Revert command, 15 S
Render Queue panel, 6 RGB color, 179–180, 196, 349, Safe Margins, 31
render settings, 34–35 353 sample point parameter, 337
Render Settings dialog, 34–36 RGB histogram, 147 sample radius parameter, 337
Render Settings section, 34–35 RGB images, 179–180 sample time parameter, 337
rendering RGB values, 349, 353, 354 sampleImage(), 337, 338
3D frames, 10 RGBA (red, blue, green, and Samples Per Frame setting, 65
adjustment layers and, alpha) color space, 337, 338 sampleSize variable, 338
119–121 rgbToHsl(), 337, 338 sanityCheck.exr file, 374

463

Index

sanityCheck.tif file, 351, 374 third-party, xxv cast, 396–397, 400
saturation scroll wheel, 25 color keying and, 203
color, 105, 157–159, S-curve adjustment, 156 color matching, 164–165
387–388 SD (Standard Definition), 21 contact, 397–398
reducing, 333–335 SDTV NTSC profile, 353, 355 techniques for creating,
Saturation mode, 105 SDTV PAL profile, 353 396–399
saving searches. See finding items Shake, xxiii, 70, 190
advanced options for, 15 seedRandom(), 327 Shape Effects, 94
Auto-Save, 15 Select Following Keyframes Shape layers, 91, 94, 121
projects, 15 option, 60 Shape Path Visibility button, 32
Scale shortcut, 47 Select Previous Keyframes shapes
scaling option, 60 Bézier, 56, 215
animation, 54, 57 Selection tool, 47, 92, 216, 232 complex, 213
brushes, 231 selections, 77–108. See mask, 91–94, 214, 215, 218
masks, 92 also blending modes; repeatable, 94
motion tracking, 241–242 compositing; masks Rotobezier, 215–216
nonuniform, 35, 46 alpha channels, 80 rotoscoped, 251
points, 214 alpha channels/ vector, 80
proportionally, 54 premultiplication, 87–90 sharpening images, 299, 301
timing, 61 bitmap selection channel, Shatter effect, 447
Transform Box, 54 84 Shift Channels, 180, 205, 299
values, 54 calculating opacity settings Shift key, 46, 47, 58
scene-referred color data, 383 for, 86–87 shortcuts, hack, 130–131. See
screen. See also display combining, 83, 95–98 also keyboard shortcuts
full-screen mode, 7, 30 edge mattes for, 205–206 Shortcuts file, 130
maximizing, 7 feathered alpha, 84–86 shots. See camera shots
Screen Balance control, 189, keyframes, 60 Show Animated Properties
195, 196 layers, 45 toggle, 48
Screen Colour eyedropper, 188 masks, 80–81 Show Channel icon, 32
Screen Colour setting, 188 mattes, 79 Show Channel pulldown, 144
Screen Gain control, 189, 195 transparency selections Show Graph Editor Set toggle,
Screen Grow/Shrink control, from effects, 84 48
200 types of, 79–83 Show Graph Tool Tips, 53
screen matte, 194–201 semicolon (;) key, 46 Show Properties menu, 48
Screen mode, 101–103 Separate Fields setting, 20 Show Reference Graph option,
Screen Preblur option, Sequence Footage Frames per 50
198–200 Second option, 119 Show Rendering in Process
Screen Shrink/Grow value, sequences option, 123
190, 191–192 advantages of, 19 Show Selected Properties
Screen Softness control, 198 Bridge and, 9 toggle, 48
Screen transfer mode, 372 importing, 8–10, 119 Show Transform Box option,
scripts missing frames, 10 54
background renderers, for moving footage, 19 shutter, camera, 64, 65
126–127 multiple, 9 shutter angle, 274
hack shortcuts, 130–132 numbering, 36, 111 Shutter Angle setting, 64–67
preferences, 129 selecting frames from, 9 Shutter Phase setting, 65, 67
rd_MergeProjects.jsx script, Set First Vertex command, 100 shutter speed, 64, 65
14 Set Proxy setting, 124 shy layers, 42
resources on DVD, xxiv–xxv shadows, 396–399 Shy toggle, 42

464

Index

Silhouette Alpha mode, 105 spill correction, 189 text reminders, 121
Silhouette blending mode, 105 spill suppression, 189, 200–201, TextEdit, 130
Simple Choker, 190–192, 204 207 TGA format, 22
sine wave, 344 splitting layers, 45 three-way blur, 395
Single Frame mode, 231, 232 spreadsheets, 57 three-way color corrector, 160,
sky replacement, 410–413 square pixels, 19, 21, 84 389
slamming gamma, 173–174 squibs, 433–434 threshold pixels, 84
Slider Control, 332 sRGB color, 352 TIFF format, 22, 36, 374
smoke, 413–421. See also fire; sRGB profile, 353 Tilde key (~), 7
pyrotechnics Stabilize tracks, 241, 242, 245 time. See also timing
from fire, 442–446 Stamp tool, 229 absolute, 68–69
fractal noise as, 103 Standard Definition (SD), 21 current, 44, 57
from gun, 431–432 Standard workspace, 5–6 display of, 16
smooth() expression, 253–254 Starch tool, 227, 228 in expressions, 327–328
smoothing Stencil Alpha mode, 105, 190, manipulating in Timeline,
curves, 254 391 68–75
global, 74 Stencil blending mode, 105 nested, 118–119
local, 74 still images, 19 random, 327–328
moving camera, 252–254 stop motion effect, 322–324 time attribute, 320
Smoothing setting, 73 Stop Tracking option, 244 time delay, 316–318
Smoothness Iterations setting, stopwatch, 60 Time Indicator, 44
74 Stretch values, 69 time navigation, 43–44
Snap button, 54 strokes Time Remap expression, 328
snapping clone, 232, 234–235 Time Remap feature, 70, 71, 72
to frames, 54 paint, 212, 230–232 Time Remap property,
to markers, 54 shapes and, 94 323–324, 325
snow, 423, 425, 426–427 stylus pen, 230 time remapping expressions,
Softness option, 183 subcompositions, 119, 123 324–328
solo layers, 42 subpixel motion, 244 Time Sampling section, 34
Solo Switches option, 42 Subtract mode, 95, 96 Time Stretch tool, 41, 69, 70
sound. See audio Switches Affect Nested Comps timecode, 16
source clips, 69, 386 option, 118 Timeline, 39–75
source files, 11–15 SynthEyes, xxiv, 42, 266–268, color-coding layers/
finding, 12 468 compositions, 42–43
importing, 8, 9–10 column views in, 40–43
organizing, 8–11 T comments added in, 42–43
replacing, 11–12 tags, adding to layers, 42–43 described, 40
unused, 12, 14, 15 target layers, 99 keyboard shortcuts, 43–46
source folders, 112 target masks, 99, 220, 224 motion blur, 64–65
source footage, 11, 14, 17, target variable, 338 navigation keyboard
22–24, 112 telecine conversions, 21 shortcuts, 43–44
source formats, 22–23 Telecine process, 360, 365 spatial offsets, 61–63
Source Frame mode, 72 templates time manipulation in, 68–75
source materials, 11–12 composition settings, 24 view options for, 46
Source Time Shift, 233 Output Modules, 36 zooming in/out, 46
Source tool, 41 projects, 111–112 Timeline panel, 6
Spacebar, 25 Render Queue, 36 Time-Reverse Keyframes
spatial offsets, 61–63 text editors, 130, 131, 257 option, 61
Speed graphs, 50 text preferences, 130–131 Time-Reverse Layer option, 69

465

Index

timeToFrames(), 323 overlapping, 96–98 version numbers, 36
Timewarp feature, 72–75, 284 pixels, 106 versions, project, 15
Timewarp setting, 67 preserving, 105–106 vertex count, 214, 217
time-wiggling effect, 324–327 track mattes and, 183–184 vertices, 100, 311
timing. See also time transparency channel, 80 video formats, 306–307
color, 389–390 Transparency Grid icon, 30 video gamma space, 368–369
comps, 117 transparency selections, 84 Video Preview settings, 30
considerations, 117 Trapcode Form, 468 video space, 368–373
dimensions, 117 Trapcode Horizon, 469 View menu, 224
layers, 117 Trapcode Lux, 396, 469 View Options command, 32
scaling, 61 Trapcode Particular, xxiv, 423, View panels, 24–32
Tint control, 386–388 468 viewers, 5, 24–32
Title/Action Safe overlays, 31 tree/node view, 59 vignettes, 303–304
Toggle Mask button, 32 Triangles, 227 virtual camera, 271
Tolerance option, 183 trimming, 45 virtual cinematography
Tools panel, 5 troubleshooting "ﬁlm/video look," 301–307
track mattes, 106–108 banding, 148–150, 349, 360, formats/aspect ratios for,
applying to other track 377 302, 305
mattes, 108, 123 with histograms, 148–150 importance of color in, 302,
considerations, 108 mattes, 193–194 307
described, 106 premultiplication errors, 90 virtual images, 83
options for, 107 Tuning section, 72, 73 Vivid Light mode, 104
render order, 108, 123 Twixtor, 74
transparency and, 183–184 W
vs. alpha channels, 107 U Warp plug-in, 469
Tracker Controls panel, 239, überkey, 58 Warp Layer control, 73
241, 244 Ultimatte plug-in, 195 warping tools
Tracker panel, 241–245 undo levels, 129 Mesh Warp effect, 418–419
TrackerViz, 217, 218, 247–248 Uniﬁed Camera Tool, 285 Reshape, 219–224
tracking. See motion tracking unit vectors, 340 Watch Folder option, 128
tracks Unlock All Layers command, water
audio, 314–316 42 precipitation, 423–427
motion. See motion tracks user interface, 129, 214 underwater shots, 303
Raw, 242 User Interface Colors control, weather conditions, 405–427
Stabilize, 241 214 fog/smoke/mist, 413–421
Transform, 241 U/UU keyboard shortcuts, lightning strikes, 435,
Transfer Controls tool, 41 58–59 438–439
Transform Box, 54, 58 UV maps, 400 particulate matter in air,
Transform controls, 46–47 406–410
Transform effect, 123 V precipitation, 423–427
Transform menu, 46 sky replacement, 410–413
value(), 318
Transform tracks, 241 snow, 423, 425, 426–427
Value Graph, 50
transforms wind, 422–423
valueAtTime(), 311, 318, 341
collapsed, 117, 118 Weighting setting, 74
Vector Detail setting, 73
keyboard shortcuts, 46–47 while() loop, 343
vector dot product, 340
layer space, 328–337 white levels, 142
vector paint tools, 230, 232
points, 214 white pixels, 102
vector shapes, 80
transparency White Point control, 180
vectors, 73–74, 340
masks, 95, 96 White Softness control, 180

466

Index

whites working space Z
blown-out, 148 Color Management,
Z axis, 288
color matching, 165–166 351–353
Z coordinate, 329
wiggle(), 254, 324–327 linearized, 101–102, 103,
Z depth, 67, 251
wiggle effect, 325–327 373, 377
Z space, 251
wind, 422–423 Workspace menu, 5–6
ZBornToy plug-in, xxiv, 403,
Window menu, 5 workspaces, 4–7
468
windows, working with, 7 world coordinates, 329
Zoom Blur setting, 67
Windows systems world space, 329
zoom lenses, 303
incompatible formats, 12 Write On option, 231–232
zoom settings, 25–26
memory and, 132
zooming in/out
rendering and, 126, 128, X methods for, 25–26
130–131 X key, 232 with mouse scroll wheel, 46
wire removal, 234–235 XML (Extensible Markup panels, 7
Work Area, 45, 46 Language), 131–132 on Timeline, 46
workﬂow XMP metadata, 36 vs. pushing, 274, 286–288
effects/presets, 33 X-splines, 256–257, 259 Zorro Layer Tagger, 42–43
features, xii–xxiii
non-destructive, 8 Y
output with Render Queue,
Y axis, 286
33–36
YUV format, 198, 199–200, 348,
streamlining, 8–15
354

467

What’s on the DVD?
A lthough this book is designed not to rely on tutorials, many of the
techniques described in the text can be further explored via the dozens of
projects and accompanying footage and stills included on the disc. Wher-
ever possible, HD (1920×1080) clips are incorporated; other examples use
NTSC footage or stills if that is all that’s required to get the point across.
Additionally, the DVD includes demos of more than a dozen plug-ins and
applications. These demos are similar to the real software for everything
but output, allowing you to experiment with your own footage.
. Custom scripts from redeﬁnery. A number of custom scripts created
by Jeff Almasol are included with this edition, including Lightwrap and
CameraProjectionSetup, which duplicate the exact steps described to set
those up (Chapters 12 and 9, respectively); the three scripts described in
the appendix on the accompanying disc; and two scripts included with
previous editions and updated for version CS4: Duplink, which creates
“instance” objects, and Merge Projects, which integrates the structure of
an imported project into the master project. More information on these
is included as comments in the scripts themselves (which can be opened
with any text editor).
. SynthEyes (Andersson Technologies) provides fully automatic, as well as
user-controlled, matchmoving for single or batchprocessed shots; it’s a
stand-alone program that exports to After Effects.
. Lenscare (Frischluft) creates a more natural and beautiful lens blur than
the built-in After Effects tools.
. ZbornToy (Frischluft) enhances what you can do with RPF format ﬁles in
After Effects, including relighting a 3D render that includes a normal
map in After Effects.
. Trapcode Particular (Red Giant Software) designs 3D particle systems that
simulate air resistance, gravity, and turbulence. It provides a real-time
preview, as well as controls that allow you to freeze time and manipulate
a camera in the scene.
. Trapcode Form (Red Giant Software) is a cousin to Trapcode Particular
and allows you to array particles across a mesh, so that they can hold a
three dimensional shape.

468

. Trapcode Horizon (Red Giant Software) creates 3D-aware gradients and
spherical maps, which is useful to create skies and multi-color gradients
for use with a 3D camera.
. Warp (Red Giant Software) is a natural companion to Mocha for After
Effects CS4 and goes far beyond the native After Effects corner pinning
tools in both features and image quality.
. Knoll Light Factory Pro (Red Giant Software) includes such pre-built
lighting effects as lens flares, sparkles, glows, and more. It also provides
individual lens components so you can create your own custom effects.
. Knoll Light Factory Unmult (Red Giant Software) is not a demo but a full
working version of a plug-in that performs one task: removing all black
matting from edges and converting it to transparency.
. Magic Bullet Colorista (Red Giant Software) adds 3-way Lift/Gamma/
Gain color control to After Effects, allowing you to individually style the
colors of shadows, mid-tones and highlights.
. Trapcode Lux (Red Giant Software) simulates light reflection by using
After Effects’ built-in lights to create visible light that corresponds to
your layers’ lighting schemes.
. Magic Bullet Looks (Red Giant Software) is an intuitive, real-world based
application that lets you dramatically change the color look of your
footage.
. ReelSmart Motion Blur (RE: Vision Effects) allows you to procedurally
generate motion blur for moving elements in a shot which lack it (or
lack enough of it). After Effects’ built-in motion blur is available only on
animated elements.
. PV Feather (RE: Vision Effects) adds features long missing from After
Effects and available in comparable packages such as Shake: the ability
to control per-vertex (or per-spline) mask feather.
. Twixtor (RE: Vision Effects) is an alternative to After Effects’ built-in
Timewarp effect for optical-flow-based retiming of footage.
. Erodilation and Copy Image (ObviousFX) are not demos but freeware.
These plug-ins allow for quick manipulation of matte data and quick
production of stills, respectively.
. Also included are dozens of After Effects CS4 project files demonstrat-
ing techniques described in the book. These range from simple demon-
strations of single concepts to completed shots.
. Live Action effects and location footage is included from Pixel Corps,
Kontent Films, and more.

469

Get free online access
to this book!
And sign up for a free trial to Safari Books
Online to get access to thousands more!
With the purchase of this book you have instant online,
searchable access to it on Safari Books Online! And while you’re
there, be sure to check out the Safari on-demand digital library
and its Free Trial Offer (a separate sign-up process)—where
you can access thousands of technical and inspirational books,
instructional videos, and articles from the world’s leading
creative professionals with a Safari Books Online subscription.

Simply visit www.peachpit.com/safarienabled
and enter code OAFYSAA to try it today.

Adobe.after.effects.cs4 visual.effects.and.compositing.studio.techniques

More Related Content

Viewers also liked (10)

Similar to Adobe.after.effects.cs4 visual.effects.and.compositing.studio.techniques (20)

Adobe.after.effects.cs4 visual.effects.and.compositing.studio.techniques