Ue4 siggraph-nick-whiting

#UE4
Best Practices and Lessons
for Unreal Engine 4 and VR in the
Classroom
Nick Whiting
Lead Engineer, VR and Visual Scripting
Epic Games

#UE4
Epic has had many academic collaborations, and
over the past year, but this talk will focus on
lessons learned from VR-centric courses:
Stanford University
CS210: Software Project Experience
CS211: Content Creation in Virtual Reality

#UE4
Stanford University
CS210:
Software Project Experience
Senior projects course, where students
pick a problem from a set of corporate
partners, and work with them to deliver
a product at the end of the course.

#UE4
Stanford University
CS211:
Content Creation in Virtual
Reality
Students build up a small VR
project, which culminates with a
final assignment that involves
making a multiplayer, shared virtual
experience.

#UE4
My Educational Qualifications:

#UE4
● Went to university

#UE4
● …

#UE4
● …
● ...

#UE4
● (Completely unqualified to teach anybody much of
anything)

#UE4
● (Completely unqualified to teach anybody much of
anything)
Time for some on the job training! Along the way, I
picked up a few lessons that might be helpful to others.

#UE4
Lesson 1:
Students can explore problems that aren’t
immediate business needs

#UE4
Lesson 1: Exploration
Virtual Reality has been an area of focus at Epic, but
keeping up is (several) full time jobs.
● Platform support
● Making the technical challenges of VR invisible
from the user, so they can focus on the experience

#UE4
One thing we’ve learned is that VR development is all
about trying…
...and repeatedly failing.

#UE4
We need a way to rapidly try out diverse ideas, and
identifying the successful concepts.
Ideal problem for investigation has multiple possible
solutions that may pan out, but we expect some will be
more successful than others.

#UE4
Problem: How do we adapt the editor for VR?

#UE4
The editor is fertile ground for ideas! But, there are a
lot of unexplored problems for us:
● How do you interact with objects in VR?
● What do controls look like?
● How do you translate UI to VR?

#UE4
Define and constrain the problem:
Investigate ways to “graybox” levels without taking off
the VR headset.
Workflows:
● Placing of new shapes in the editor
● Manipulating those shapes
● Moving around the world

#UE4
Two teams rose to the challenge:
Karaage Jarvis

#UE4
Team Karaage’s Approach:
● Focus on novel hardware devices
● Refine the interactions with those specific devices
● Haptic feedback

#UE4
Team Karaage:

#UE4
Key Findings:
● The importance of haptic feedback
○ Made interactions much more natural
● Movement with a 3D mouse
● Experimentation with new UI layouts
○ Ring of options

#UE4
Team Jarvis’ Approach:
● Focus on motion controllers
● Manipulation of pre-placed meshes, and limited
content browser support
● Study the methods of interaction with those objects

#UE4
Team Jarvis:

#UE4
Key Findings:
● Interactions with the world through motion controls
○ Lots of similarities to smartphone gestures
● Voice is a viable option!
● Importance of context in control scheme

#UE4
Conclusions:
These students took unexpectedly different
approaches, based on their interest and experiences.
They tried things that we didn’t think of, and couldn’t
afford the time to try.
They were quick to give back to the community!

#UE4
Conclusions:
Need to be sure to frame the questions very
specifically, and limit the scope of the problem.
Guide early, guide often!
Make sure the students don’t fear failure. We want
bold, risky ideas!

#UE4
Lesson 2:
Innovation comes from both sides of the
partnership

#UE4
Lesson 2: Innovation
What if instead of teaching computer science students
to use UE4, UE4 taught people computer science?
Programming is a hard skill to learn:
● Lots of syntax to learn
● Typos can cause failure
● Specifics of the language often get in the way of
understanding the core concepts

#UE4
The problems learning programming are very similar to
problems we face in game development.
In order to democratize game development, we had to
teach non-programmers some fundamentals of
programming. Our goal was to help people help
themselves...

#UE4
Enter Blueprints...

#UE4
Design Goals of Blueprints:
● Nobody should be stuck waiting for a programmer
● Allow non-programmers to do things programmers
would normally do
○ Create gameplay, tweak systems, prototype
● Blueprints should be contextual, to help users learn
● Forgiving! No typos...
“If you can make a flowchart, you can make a Blueprint”

#UE4
Design Goals of Blueprints:
The Ultimate Goal:
Blueprints should be a language that both
programmers and non-programmers can understand.

#UE4
“When I touch the trigger, play a sound at the Horn,
and then open the door.”

#UE4
From these small, simple building blocks of logic, you
can evolve very complex behaviours.
“When I press A, I want to jump.”
“When I press B, I want to attack.”
“When the sword touches me, I want to lose health”
None of these parts are complex, but when you put
them all together, you can make a full game...

#UE4
What does this help student projects groups do?
● Multidisciplinary groups. Not everyone has to be a
programmer! Designers, artists, and others can
work together
● Fail faster!
● Share their work with a much broader audience

#UE4
Take a step back…
How else might visual programming be useful?

#UE4
Fundamentals of Computer Science / Engineering:
● Breaking down problems into manageable parts
● Recognizing Patterns
● Creating solutions for those components
● Putting those simple solutions back together

#UE4
Traditional Programming Education:
● Typos hurt!
● Large toolchain, with a slow turnaround
● Inconsistent documentation and toolset make it
hard to discover
● Obtuse debugging tools
● Translation is an issue (algorithm to code)

#UE4
Could visual programming be a friendlier intro?
● You can make logic errors, but not “typos”
● Highly contextual searching helps you stumble your
way to what you want
● Great for visual learners
○ You don’t have to know exactly what to type to
get it to do something
● Instant (and often fun!) feedback
● Focus on the learning of the fundamentals, not the
specifics of the language

#UE4
Lesson 3:
Different Worlds

#UE4
Lesson 3: Different Worlds
Business and academics have very different sets of
priorities, and we’ve seen how we can use that to our
advantage.
In other areas though, the metrics we use and the
metrics the students use might not be so aligned...

#UE4
Metrics of Success:
Scoping
For us, we scope conservatively, and are very focused
on a well-defined problem
For students, it’s tempting to solve all the problems at
once!

#UE4
Metrics of Success:
The Allure of Sexy Solutions
For us, we’re often skeptical of sexy solutions
For students, playing with new toys is a lot of fun!

#UE4
Metrics of Success:
Failure
For us, failure is a constant on the path to discovery,
especially with emerging technologies like VR.
For students, “fail” is a scary word!

#UE4
Documentation:
With games, we want the fastest path from download
to making games possible!
● Focus is on samples, and explanations of how to
adapt that content
● Documentation focuses on the “how” questions
○ How do I set up a player?
○ How do I make a sign flash?

#UE4
Documentation:
With education, we want to understand what we’re
doing as deeply as possible.
● Focus is on understanding the complex systems
● Usually want to tear apart or replace systems,
instead of adapting them
● Often want documentation on things game
developers rarely modify

#UE4
Documentation:
Fortunately, students are often willing to contribute
back!

#UE4
Extra Credit:
Invest in Ambassadors!

#UE4
Extra Credit: Ambassadors
As much as we’d like to be everywhere and help
everyone, there are only 24 hours in a day.
At Stanford for CS210 alone, there were five groups
using UE4, including the two we were directly
mentoring.
We needed help!

#UE4
Extra Credit: Ambassadors!
Enter Noam Ben-Avi and Jack Reidy!
In addition to helping develop the CS211 course, they
also helped every UE4 team at the fair.
The hours we spent with them helping them out (it
wasn’t a lot!) was a huge force multiplier. We couldn’t
have done it without them!

#UE4
Extra Credit: Ambassadors!
Their course is now becoming available online:
http://www.layoutvr.com

#UE4
Thanks!
Special thanks to:
Luis Cataldi (Epic Games)
Jay Borenstein (Stanford / Facebook)
Noam Ben-Avi (Stanford)
Jack Reidy (Stanford)
Kelvin Sung (University of Washington)
Contact:
nick.whiting@epicgames.com

Ue4 siggraph-nick-whiting

More Related Content

Viewers also liked (12)

Similar to Ue4 siggraph-nick-whiting (20)

Recently uploaded (20)

Ue4 siggraph-nick-whiting