SlideShare a Scribd company logo
germany austria switzerland
The
great
2017
mobile
network
test
he network operators in Germany, Aus-
tria and Switzerland wait for the results
of our annual mobile network benchmark
with the highest tension imaginable. As
in the years before, we have conducted this
benchmark as part of our well-proven co­
operation with the Aachen-based network
testing specialist P3 communications.
However, we have never before experien­
ced such harsh disputes with some of the test
candidates in the forefront of our test. They
argued about questions like which smartphone
models were to be used for the measure-
ments, how the test routes should be put to-
gether or how we should balance the indivi-
dual voice and data results. Eventually, all
their reasoning could be ­traced back to one
pur­pose: Some of the candidates hoped to
gain advantages in those areas where they
believed to be particularly strong and having
the edge over their competitors.
We did show good sportsmanship inter­
preting the pressure built up by some of the
candidates in the run-up of our test as proof
for the high relevance and acceptance of our
benchmarks within the whole industry.
At the end of 2016, we conducted our annual
mobile network benchmark for the 23rd time.
Together with our renowned benchmarking
partner P3 communications, once again we have
investigated, which mobile operators in Germany,
Austria and Switzerland are a cut above the rest
– with utmost effort and our objective,
customer-oriented testing methodology.
60%
DATA
Youtube Quality
Success RaƟo
Web Page Download
File Up- and Download
40%
VOICE
Call Success RaƟo
Call Setup Time
Speech Quality
Quality benchmark for many years
Traditionally, fairness and transparency play
a very big role in our network test (also see
page 16). In any case, with all decisions con-
cerning methodology and scoring, our high-
est priority is always to ensure the signifi-
cance of our benchmark for you, our readers.
As a matter of principle, connect and P3
value objective and authoritative results
about the actual quality and performance of
the cellular networks second to none. Eve-
rybody interested can find an exact descrip-
tion of our methodology on pages 14 and 15.
So, on the following pages you can read
the answers to many thrilling questions: Has
Deutsche Telekom been able to defend its
top rank in Germany? How does O2 rank,
after its owner Telefónica has started to com-
bine this network‘s former radio cells with
those of E-Plus, which they bought in 2014
(also see page 15). Which operators come
out on top inAustria and Switzerland, where
the contest is traditionally conducted on an
especially high performance level?
Not all contestants will like all of the
answers and results. This again proves to us
that we have done a good job. Because this is
the only way for us to guarantee that truly
everybody can fully rely on our test results.	
			Hannes Rügheimer
Practical
relevance
in mind
In the scoring of our
test results, we ac­
count for the steadily
growing importance
of data communica­
tions.
the benchmark
Köln
Frankfurt
am MainWiesbaden
München
Berlin
Düsseldorf Wuppertal
Bielefeld
Oldenburg
Duisburg
Essen
Kassel
Halle
Dresden
Magdeburg
StuƩgart
Augsburg
Hamburg
Drivetest
Walktest
Roads
Train
Unna
Ascheberg
Greven
Bramsche
Vechta
Seevetal
Wedemark
Bad Oeynhausen
Schwerte
Soest
Brandenburg
Dessau-Rosslau
Wiedemar
Grimma
Plauen
Hof
Bayreuth
Lauf an
der Pegnitz
Pfaffenhofen
an der Ilm
Dachau
Guenzburg
Kirchheim
unter Teck
Idstein
Limburg
Troisdorf
Leonberg
gs
ruƩga
BaB
I
WW
e
T
Bludenz
Imst
Zirl
Worgl
Zell am See
Sankt Johann
im Pongau
Bischofshofen
Hallein
Seewalchen
am AƩersee
Laarkirchen
Kremsmunster
Enns
Melk Brunn
am Gebirge
Koƫngbrunn
Gleisdorf
Völkermarkt
Velden
am Wörthersee
SpiƩal an
der Drau
Kufstein
Linz
Traun
St. Pölten
Baden
Salzburg
Feldkirch
Villach
Wien
Graz
Klagenfurt
InnsbruckZürich
Luzern
Uster
Schaĭausen
Thun
Lugano
Lausanne
Yverdon-les-Bains
Winthertur
Bern
Geneve
Basel
Chur
Cham
Baden
Dubendorf
Volketswil
EffreƟkon
DieƟkon
Sarnen
Bellinzona
Spiez
Bulle
Vevey
Renens
Morges
Gland
Meyrin
Carouge
Worb
Solothurn
PraƩeln
MuƩenz
T
SINCE
network test
3
600 of 1000 Points400 of 1000 Points
VOICE DATA
of180
of270
of80
of120
of50
of75
of60
of90
92%
88%
49%
Telekom Vodafone Telefónica
95%
91%
78%
95%
90%
76%
94%
92%
71%
89%
82%
62%
of30
of45
57%
54%
33%
61%
46%
30%
87%
81%
65%
87%
84%
56%
87%
68%
90%
CiƟes
Drivetest
CiƟes
Walktest
Towns
Drivetest
Roads
Drivetest
Train
Walktest
600 of 1000 Points400 of 1000 Points
VOICE DATA
of180
of270
of80
of120
of50
of75
of60
of90
92%
88%
49%
Telekom Vodafone Telefónica
95%
91%
78%
95%
90%
76%
94%
92%
71%
89%
82%
62%
of30
of45
57%
54%
33%
61%
46%
30%
87%
81%
65%
87%
84%
56%
87%
68%
90%
CiƟes
Drivetest
CiƟes
Walktest
Towns
Drivetest
Roads
Drivetest
Train
Walktest
The quality and reliability of voice connections represent
40 per cent of the final score. Which operator offers the
best network in this respect?
Data tests account for 60 per cent of
the final score.Who delivers the best
performance in this category?Voice Data
In some users‘ communications
habits, voice telephony only
plays a minor role. But conven­
tional phone calls are far from
being outdated. Otherwise, the
three German network opera­
tors would probably not have
gone to the lengths of imple­
menting VoLTE – telephony
based on sending data packets
over the LTE network.
Therefore, quality and perfor­
mance of voice telephony still
played an important part in the
drive tests and walk tests con­
ducted by P3: For this purpose,
both cars that were driving
through 17 large and many
smaller German ­­cities carried
six Samsung smartphones
each.They permanently called
their counterparts in the other
vehicle. In order to simulate the
everyday smartphone use, the
phones would constantly trans­
fer data in the background
­during the ­telephony tests.
An identical device configu­ra­
tion was used in the backpacks
Data communication is the
most prestigious category in
connect‘s network test. Firstly,
the results of these tests repre­
sent 60 per cent of the final
score.And secondly, the test
parcours to be completed by
our candidates incorporates a
large number of practice-orien­
ted applications. For instance,
the smartphones frequently
­access the most popular web
sites according to the renowned
Alexa ranking, as well as the
static „ETSI reference web ­page“
also known as „Kepler page“.
Measuring the speed and relia­
bilty of data transfers is the aim
of our upload and download tests.
We monitor uploads with test files
sized 3 MB and downloads with
1 MB files.Additionally, we verify
which amount of data travels over
the network within ten seconds.
Another scope of our testing
are Youtube videos.The popular
­video platform does not distin­
guish between standard defini­
tion (SD) and high definition (HD)
carried by the test staff who
walked around in city centres
and public buildings conducting
the walk tests.
The devices had been confi­
gured to make sure that part of
the connections would be esta­
blished via VoLTE and another
part would be transmitted via
conventional circuit-switched
telephony.
Distinct ranking order
The tests in the city centres
­already showed a clear picture:
Both in the drive tests and in
the walk tests, Deutsche Tele­
kom is ahead.Vodafone follows
at a distance of a few points.
O2 is clearly defeated what
can be seen in the table below
by means of lower success
­ratios, longer call setup times
and also a lower average
speech quality.
As a result of their test drives
through smaller cities and on
connecting roads, the P3 tes­
ting team found basically the
same ranking order: Deutsche
Telekom leads,Vodafone fol­
lows at a comparatively small
distance, and O2 comes in last
with a considerably larger gap.
While Telefónicas network
keeps up quite well in smaller
towns, its distance to the lea­
ding two contenders grows larger
on the connecting roads. On the
whole, compared to last year‘s
test, O2 improved in the voice
category. For the voice calls
­examined by the testing staff
in trains, even Telekom and
­Vodafone showed some weak-
nesses. But O2 scores again
far behind them.
Regarding some measurement
values like call setup times and
speech quality,Vodafone is
­narrowly ahead. But while the
Düsseldorf-based operator could
claim a stage win in the voice
category last year, in the 2017
test this title goes to Telekom.
resolutions any more. It rather
­dynamically adapts the video
resolution to the bandwidth that
is currently available. In order to
respect this new strategy in our
tests, we examined the success
ratio of video playbacks, the
start times, the percentage of
playouts that took place with­-
out interruptions as well as the
videos‘ average resolution or
number of lines respectively.
Both P3 test cars checked
these indicators as part of their
drivetest, and also the walk test
teams had the same agenda.
All data measurements in Ger­
many were executed with the
LTE Cat.6 smartphone
Samsung Galaxy Note 4.
Strong Deutsche Telekom
In big cities, the results were
similar to the voice category:
Again, a very strong Telekom
takes the lead,Vodafone follows
with good results, and O2 is
clearly defeated. Separate ana­
lyses show that Vodafone >>
OPERATOR Telekom Vodafone Telefónica
DATA (CiƟes; Drivetest)
Web-Page Download (Live/StaƟc)
Success RaƟo (%/%) 99.4/99.7 98.8/99.2 94.9/96.3
Ø Session Time (s/s) 2.9/1.4 3.0/1.4 3.4/2.1
File Download (3 MB)
Success RaƟo/Ø Session Time (%/s) 99.7/1.3 99.5/1.9 99.1/6.0
90%/10% faster than (kbit/s) 12884/50934 7219/52516 1784/27666
File Upload (1 MB)
Success RaƟo/Ø Session Time (%/s) 99.7/1.3 99.2/1.4 96.7/2.7
90%/10% faster than (kbit/s) 5690/11586 3567/12864 1453/8719
File Download (10 Seconds)
Success RaƟo (%) 99.8 99.2 98.3
Ø Throughput (kbit/s) 55738 39675 13591
90%/10% faster than (kbit/s) 16703/107234 7880/94372 1993/32908
File Upload (10 Seconds)
Success RaƟo (%) 99.7 99.1 98.5
Ø Throughput (kbit/s) 27892 15841 8492
90%/10% faster than (kbit/s) 7885/42458 3735/29749 1382/17214
Youtube Videos
Success RaƟo/Start Time (%/s) 99.8/1.7 99.8/1.8 98.1/2.1
Playouts without InterrupƟons (%) 100.0 99.9 99.3
Ø Video ResoluƟon (p) 578 614 457
DATA (CiƟes; Walktest)
Web-Page Download (Live/StaƟc)
Success RaƟo (%/%) 99.3/99.3 98.5/98.9 91.3/92.1
Ø Session Time (s/s) 2.9/1.4 3.1/1.6 3.4/2.2
File Download (3 MB)
Success RaƟo/Ø Session Time (%/s) 99.6/1.4 99.1/1.9 95.5/6.3
90%/10% faster than (kbit/s) 13578/51513 6801/56497 1569/37915
File Upload (1 MB)
Success RaƟo/Ø Session Time (%/s) 98.6/1.5 97.7/2.0 91.6/3.3
90%/10% faster than (kbit/s) 3127/11409 1951/13029 1026/9050
File Download (10 Seconds)
Success RaƟo (%) 99.4 99.2 96.2
Ø Throughput (kbit/s) 57742 48961 17254
90%/10% faster than (kbit/s) 15848/111341 7252/108922 1471/47759
File Upload (10 Seconds)
Success RaƟo (%) 99.2 98.9 96.5
Ø Throughput (kbit/s) 25097 16423 8773
90%/10% faster than (kbit/s) 3455/42305 2141/39359 654/28672
Youtube Videos
Success RaƟo/Start Time (%/s) 99.4/1.7 99.0/1.9 98.2/2.3
Playouts without InterrupƟons (%) 100.0 100.0 99.0
Ø Video ResoluƟon (p) 572 609 465
DATA (Towns; Drivetest)
Web-Page Download (Live/StaƟc)
Success RaƟo (%/%) 99.0/99.6 98.2/98.5 94.3/95.5
Ø Session Time (s/s) 2.9/1.4 3.2/1.6 3.5/2.2
File Download (3 MB)
Success RaƟo/Ø Session Time (%/s) 99.8/1.6 99.0/2.3 96.7/4.4
90%/10% faster than (kbit/s) 8710/48387 6378/36364 2908/29376
File Upload (1 MB)
Success RaƟo/Ø Session Time (%/s) 99.2/1.7 97.7/1.8 94.4/3.6
90%/10% faster than (kbit/s) 3230/11227 2298/11594 1045/7779
File Download (10 Seconds)
Success RaƟo (%) 99.8 99.4 96.2
Ø Throughput (kbit/s) 36873 20028 17021
90%/10% faster than (kbit/s) 10172/73053 6410/39497 3904/35924
File Upload (10 Seconds)
Success RaƟo (%) 99.4 99.0 98.2
Ø Throughput (kbit/s) 20759 11329 6804
90%/10% faster than (kbit/s) 4144/41682 2670/19807 988/16864
Youtube Videos
Success RaƟo/Start Time (%/s) 99.6/1.8 99.4/1.8 98.2/2.2
Playouts without InterrupƟons (%) 100.0 100.0 99.8
Ø Video ResoluƟon (p) 556 580 484
OPERATOR Telekom Vodafone Telefónica
VOICE (CiƟes; Drivetest)
Call Success RaƟo (%) 99.4 98.8 95.9
Call Setup Time (s) 4.0 3.9 5.5
Speech Quality (MOS-LQO) 3.8 3.9 3.7
VOICE (CiƟes; Walktest)
Call Success RaƟo (%) 99.4 98.5 96.7
Call Setup Time (s) 4.0 4.0 5.3
Speech Quality (MOS-LQO) 3.9 3.9 3.7
VOICE (Towns; Drivetest)
Call Success RaƟo (%) 99.5 98.7 97.2
Call Setup Time (s) 4.0 3.9 5.7
Speech Quality (MOS-LQO) 3.8 3.9 3.6
VOICE (Roads; Drivetest)
Call Success RaƟo (%) 98.6 97.4 88.3
Call Setup Time (s) 4.3 4.2 6.4
Speech Quality (MOS-LQO) 3.8 3.8 3.3
VOICE (Train; Walktest)
Call Success RaƟo (%) 84.2 83.1 76.5
Call Setup Time (s) 5.1 5.4 6.9
Speech Quality (MOS-LQO) 3.6 3.6 3.1
Germany
600 of 1000 Points400 of 1000 Points
VOICE DATA
of180
of270
of80
of120
of50
of75
of60
of90
92%
88%
49%
Telekom Vodafone Telefónica
95%
91%
78%
95%
90%
76%
94%
92%
71%
89%
82%
62%
of30
of45
57%
54%
33%
61%
46%
30%
87%
81%
65%
87%
84%
56%
87%
68%
90%
CiƟes
Drivetest
CiƟes
Walktest
Towns
Drivetest
Roads
Drivetest
Train
Walktest
network test
54 1/2017
Modern cars rely heavily on connectivity. How does
this actually work out on German roads?
Railways used to be the blind spot of German mobile
network operators. Is this still true this year?
Connecting Roads
It was approximately 6600 kilo­
metres that P3‘s two test vehic­
les covered this year on German
connecting roads – on top of the
5500 kilometres that each car
covered driving through large and
smaller cities.The point of this
exercise: Gaining closer insights
about the quality and reliability of
the mobile networks on this parti­
cular type of roads.
Distinct ranking order
on the roads
While Telekom and Vodafone
­were almost at level in the voice
tests conducted on connecting
roads, their offset becomes more
obvious in the data cate­gory.
­Especially regarding the success
ratios of web page access as well
The testing staff spent about 33
hours on 15 different ICE trains
during their railway tests con­
ducted in 2016.Yet, the measu­
rements did not only take place
in these flagship trains of Deut­
sche Bahn, but also considered
regional railway connections.
The test results from German
trains should not surprise any­
body who has read the outcome
of the other categories: In the
trains, Deutsche Telekom once
again leads the pack,Vodafone
scores second with viable re­
sults, and O2 brings up the rear.
Especially when comparing
the partial results obtained on
the railways, looking over the
borders – specifically at the re­
sults of the respective tests in
Austria and Switzerland – may
make German railway custo­
mers quite envious. Both alpine
countries are considerably
ahead in this respect.And even
Telekom, which scores best in
the measurements taken in
German railways, cannot mea­
sure up with the results of the
Austrian and Swiss operators
– by far. Deutsche Bahn has
as downloads and uploads,Tele­
kom clearly ranks first and keeps
its competitor from Düsseldorf at
a distance. Still, this match takes
place at a very high level when
looking at the distance of O2.
Similar to the large and small
­cities before,Telefónica also loses
valuable points on the connecting
roads and thus falls back further
behind the leading two German
operators.
The test results in this category
are quite obvious: Car drivers
who need robust data connec­
tions on the road – whether for
navig­ation, for communication or
mobile entertainment purposes
– currently cannot pass the
­mobile networks of Deutsche
­Telekom or Vodafone.
­recently started an initiative to
enhance connectivity especially
in its ICE trains in close coope­
ration with the German mobile
network operators. But this
does not seem to have much
­impact on this year‘s mobile
network test.
Much need for improvement
When looking at the details,
there are many similarities bet­
ween the railways and the con­
necting roads:While Telekom
and Vodafone were almost at
level regarding voice phone
calls in trains, their offset grows
in the data category. Here,
­Telekom offers the best results
– but still shows a lot of room
for improvement.Voda­fone‘s
results are mid-level, and O2
once again comes in last.
­Success ratios, like those for
web surfing, of around 85 per
cent at Telekom, about 77 per
cent at Vodafone and approxi­
mately 63 per cent at O2
­convey a clear message:When
it comes to connectivity in Ger­
man railways, there remains a
lot of work to be done.
Telekom
wins this
year‘s
­network test at a distinct dis­
tance to runner-up Vodafone.
Both in the voice and data
­categories, the Bonn-based
operator turns out to be the
strongest conveyor.
Although we have raised our
requirements, last year‘s win­
ner Telekom not only mana­
ged to defend its position, but
actually continued to improve
its score.Therefore Telekom
absolutely deserves the first
place in Germany – and this
for the sixth time in a row.
Compared to
the results of
last year, both
the networks
of O2 and E-Plus managed to
improve.
This is particularly underlined
by the results of the voice
measurements.A valid expla­
nation for O2 not scoring any
better may be the problems
caused by the ongoing inte­
gration of both networks.
So we hope in the best
­interest of O2‘s customers
that this integration will
­con­tinue to foster noticeable
­improvements.
Vodafone also
improved both
in the voice
and data cate­
gories compared to last year‘s
results. But still Telekom out­
performed their Düsseldorf-
based competitor in this
year‘s voice tests.Anyway,
the 2017 network test em­
phasizes Vodafone‘s clear
­improvements in the data
scores. In this context,
Vodafone‘s excellent Youtube
results are particularly eye-
catching.All in all, these
­results entail a second rank
with a good overall score.
OPERATOR Telekom Vodafone Telefónica
DATA (Roads; Drivetest)
Web-Page Download (Live/StaƟc)
Success RaƟo (%/%) 99.0/99.3 95.9/96.3 90.1/91.5
Ø Session Time (s/s) 3.0/1.5 3.2/1.6 3.5/2.3
File Download (3 MB)
Success RaƟo/Ø Session Time (%/s) 99.3/2.0 97.0/2.5 93.8/4.8
90%/10% faster than (kbit/s) 6409/43956 5908/41958 2359/29851
File Upload (1 MB)
Success RaƟo/Ø Session Time (%/s) 98.6/1.9 96.8/1.8 90.8/3.8
90%/10% faster than (kbit/s) 2475/10344 2632/12214 959/8032
File Download (10 Seconds)
Success RaƟo (%) 99.5 96.9 95.4
Ø Throughput (kbit/s) 30575 26531 15444
90%/10% faster than (kbit/s) 7496/59980 6780/55954 2801/36096
File Upload (10 Seconds)
Success RaƟo (%) 99.2 96.3 92.5
Ø Throughput (kbit/s) 15882 12376 6963
90%/10% faster than (kbit/s) 3060/33803 2635/20692 937/17988
Youtube Videos
Success RaƟo/Start Time (%/s) 99.6/1.8 97.4/1.9 92.7/2.2
Playouts without InterrupƟons (%) 100.0 100.0 99.9
Ø Video ResoluƟon (p) 543 593 492
OPERATOR Telekom Vodafone Telefónica
DATA (Train; Walktest)
Web-Page Download (Live/StaƟc)
Success RaƟo (%/%) 84.5/85.5 76.7/78.9 62.7/61.7
Ø Session Time (s/s) 3.4/1.9 3.7/2.1 4.5/3.7
File Download (3 MB)
Success RaƟo/Ø Session Time (%/s) 88.2/5.4 81.2/4.9 72.6/11.8
90%/10% faster than (kbit/s) 2440/29183 2491/36934 726/12771
File Upload (1 MB)
Success RaƟo/Ø Session Time (%/s) 87.0/2.9 78.8/3.2 62.4/7.3
90%/10% faster than (kbit/s) 1151/8555 1058/10974 526/5915
File Download (10 Seconds)
Success RaƟo (%) 87.9 81.2 72.6
Ø Throughput (kbit/s) 21106 16802 7094
90%/10% faster than (kbit/s) 3825/40116 2488/37275 911/13041
File Upload (10 Seconds)
Success RaƟo (%) 87.1 81.5 73.4
Ø Throughput (kbit/s) 13661 9099 2894
90%/10% faster than (kbit/s) 999/29577 767/20322 216/8658
Youtube Videos
Success RaƟo/Start Time (%/s) 88.1/2.1 78.4/2.4 89.3/3.1
Playouts without InterrupƟons (%) 100.0 99.2 98.4
Ø Video ResoluƟon (p) 499 522 402
Data on Railways
Single review
887 841 643TOTAL
very good good sufficient-RaƟng
Shown voice, data and total scores are rounded.
VOICE
max. 400
max. 1000 Points
DATA
max. 600
366 352 270
521
373
489
Telekom Vodafone
Telefónica
could definitely improve over
last year‘s results in the data
­category. To some extent, this
supports the Düsseldorf operator
counterbalancing its shortfall in
the data score.Also,Vodafone
achieves especially good results
at You­tube playback – in this
­category, the Düsseldorfers
­partly draw level with Telekom.
Still, this is not sufficient to grant
Vodafone a partial victory.The
reason is that Vodafone scores
a little worse than Telekom in the
­discipline of web page access
and – slightly less distinctive –
regarding file uploads and
­downloads.
However, when we look at O2,
their deficits are even more ob­
vious.While the inner city walk
tests show a success ratio of 99.3
per cent for web page access in
the Telekom network and 98.5
percent at ­Vodafone, this value
drops to 91.3 per cent in Tele­
fónica‘s network. Statistically,
this means that ­almost one out
of ten web page views will fail
when O2 customers are walking
through city centres.
When observing indicators like
the success ratios or average
speeds of file downloads in the
big city drive and walk tests, the
results show the same trend that
looks increasingly familiar this
year:Telekom leads,Voda­fone
follows at not too big a distance,
and O2 clearly comes in last.
For example, according to our
walk tests, file downloads run at
average data rates of more than
13578 kbps in 90 per cent of the
cases in the Telekom network. In
the Vodafone network, it is still
more than 6801 kps, and O2
achieves only a minimum of
1568 kbps. So the latter candi­
date accomplishes not much
­more than a tenth of the speed
offered by test winner Telekom.
Drive tests in smaller towns
The drive tests conducted in
small towns gave equivalent re­
sults:Again,Telekom achieves
the best measurement values,
Vodafone follows at a distinct
but not huge distance – and
Telefónica once again comes
in last.
Similar to the inner city drive
tests, the success ratios of web
page access via the O2 network
are worse than in the networks of
the leading two providers, but still
better than O2‘s walk test results.
Our download and upload tests
show comparable results.
Once more,Vodafone turns out
to be a Youtube star in smaller
towns. Its top performance in this
category is at the same level as
test winner Telekom. Both opera­
tors seem to have a very perfor­
mant “peering“ to the content de­
livery network of Google‘s video
platform.
O2‘s weak overall scores can
be explained to some extent with
the ongoing integration of the
former E-Plus network and the
distortions coming along with it.
While O2 at least improved in the
voice category, the 2017 network
test indicates a stagnation for
this operator in the data category.
network test
76 1/2017
In comparison to the previous year, all Austrian operators im-
proved – on an already very high level.Who wins the race in
the alpine republic this time?
Since connect‘s network test
has been including the alpine
countries,Austrian network
operators used to have a neck-
and-neck contest on the high­
est level. Compared to the other
countries in the DACH region,
the Austrian contenders regu­
larly are among the top tiers.
No network operator from the
alpine republic ever scored
worse than the grade “good“.
Austrian customers have
every reason to be happy, as
they can choose from three very
good providers.And this at
­considerably lower costs than
for example in Germany.The
Austrian network operators also
look pretty good when it comes
to the roll-out of LTE. In autumn
2016,A1,T-Mobile Austria and
Drei (the Austrian subsidiary of
Hutchison Three) already offered
4G to a large part of the Austrian
population and could focus on
filling the few remaining gaps.
So, we highly anticipated the
results of the measurements
that P3 took in eleven larger
Austrian cities, on appro­xi­­
mately 2700 kilometres of
­connecting roads as well as in
Austrian railways.
Voice connections
Assessing voice telephony,A1
scores first. Especially in the
drive tests – the test calls made
from car to car – that P3 took in
larger cities, this operator per­
formed a tiny bit better than its
competitors.At the time of tes­
ting,A1 was the only Austrian
operator who already supported
VoLTE.This may have helped
particularly in the inner cities
with their good LTE coverage.
But the high score that A1
achieved in this category, would
not have been possible had this
operator not also performed
­exceptionally well in its conven­
tional telephony service.
OPERATOR A1 Drei T-Mobile
DATA (Roads; Drivetest)
Web-Page Download (Live/StaƟc)
Success RaƟo (%/%) 97.9/98.2 97.9/98.8 96.4/95.9
Ø Session Time (s/s) 2.5/1.3 2.4/1.3 2.8/1.5
File Download (3 MB)
Success RaƟo/Ø Session Time (%/s) 98.6/1.5 97.7/1.3 97.9/2.5
90%/10% faster than (kbit/s) 11934/61856 11215/60914 5452/43059
File Upload (1 MB)
Success RaƟo/Ø Session Time (%/s) 97.7/1.6 98.4/1.2 96.5/1.9
90%/10% faster than (kbit/s) 2863/14304 3294/15742 2131/17022
File Download (10 Seconds)
Success RaƟo (%) 98.8 97.9 97.2
Ø Throughput (kbit/s) 49500 50032 28171
90%/10% faster than (kbit/s) 14417/87661 17012/93075 6238/58081
File Upload (10 Seconds)
Success RaƟo (%) 98.8 98.6 98.4
Ø Throughput (kbit/s) 26980 31521 14835
90%/10% faster than (kbit/s) 2476/45416 5615/44914 2094/29461
Youtube Videos
Success RaƟo/Start Time (%/s) 97.8/1.8 99.5/1.7 97.3/1.9
Playouts without InterrupƟons (%) 100.0 100.0 99.0
Ø Video ResoluƟon (p) 626 636 596
OPERATOR A1 Drei T-Mobile
DATA (CiƟes; Drivetest)
Web-Page Download (Live/StaƟc)
Success RaƟo (%/%) 99.8/99.9 99.3/99.4 99.2/99.5
Ø Session Time (s/s) 2.3/1.1 2.4/1.2 2.6/1.3
File Download (3 MB)
Success RaƟo/Ø Session Time (%/s) 99.7/0.9 99.7/1.0 99.7/1.4
90%/10% faster than (kbit/s) 20466/62647 17394/62016 11331/56738
File Upload (1 MB)
Success RaƟo/Ø Session Time (%/s) 99.8/0.9 99.4/0.8 99.7/1.0
90%/10% faster than (kbit/s) 8583/16097 8667/16360 6552/20305
File Download (10 Seconds)
Success RaƟo (%) 99.9 99.7 99.8
Ø Throughput (kbit/s) 62742 51652 45196
90%/10% faster than (kbit/s) 23730/115195 20937/89926 14096/84799
File Upload (10 Seconds)
Success RaƟo (%) 99.9 99.8 99.3
Ø Throughput (kbit/s) 34237 36577 25496
90%/10% faster than (kbit/s) 14082/46060 20326/45098 7436/42033
Youtube Videos
Success RaƟo/Start Time (%/s) 99.9/1.7 99.5/1.7 99.7/1.8
Playouts without InterrupƟons (%) 100.0 99.9 99.9
Ø Video ResoluƟon (p) 645 649 638
DATA (CiƟes; Walktest)
Web-Page Download (Live/StaƟc)
Success RaƟo (%/%) 99.4/100.0 99.5/100.0 98.6/99.2
Ø Session Time (s/s) 2.4/1.1 2.4/1.2 2.7/1.4
File Download (3 MB)
Success RaƟo/Ø Session Time (%/s) 100.0/0.8 100.0/0.8 99.8/1.2
90%/10% faster than (kbit/s) 20430/58968 22067/58394 15464/53812
File Upload (1 MB)
Success RaƟo/Ø Session Time (%/s) 100.0/0.9 100.0/0.9 98.4/1.2
90%/10% faster than (kbit/s) 9281/16247 8153/15696 5589/20752
File Download (10 Seconds)
Success RaƟo (%) 100.0 100.0 99.2
Ø Throughput (kbit/s) 61643 55793 52971
90%/10% faster than (kbit/s) 25388/108707 23493/92364 20198/93411
File Upload (10 Seconds)
Success RaƟo (%) 100.0 100.0 99.4
Ø Throughput (kbit/s) 35294 35516 27018
90%/10% faster than (kbit/s) 14731/46139 17435/45180 5546/43022
Youtube Videos
Success RaƟo/Start Time (%/s) 99.6/1.7 100.0/1.7 99.1/1.7
Playouts without InterrupƟons (%) 100.0 100.0 100.0
Ø Video ResoluƟon (p) 652 666 652
DATA (Towns; Drivetest)
Web-Page Download (Live/StaƟc)
Success RaƟo (%/%) 99.7/99.3 99.1/99.7 99.2/99.7
Ø Session Time (s/s) 2.5/1.4 2.4/1.2 2.7/1.3
File Download (3 MB)
Success RaƟo/Ø Session Time (%/s) 99.7/1.4 99.7/1.0 99.7/1.7
90%/10% faster than (kbit/s) 10838/53982 17583/61856 8366/46720
File Upload (1 MB)
Success RaƟo/Ø Session Time (%/s) 98.4/1.7 100.0/0.9 99.7/1.2
90%/10% faster than (kbit/s) 2527/13865 7775/15311 5344/17676
File Download (10 Seconds)
Success RaƟo (%) 99.5 99.7 99.7
Ø Throughput (kbit/s) 42224 52360 30313
90%/10% faster than (kbit/s) 14119/82550 20509/86300 9052/59538
File Upload (10 Seconds)
Success RaƟo (%) 100.0 99.7 98.7
Ø Throughput (kbit/s) 24080 34217 20782
90%/10% faster than (kbit/s) 2902/44232 15027/45149 4504/30440
Youtube Videos
Success RaƟo/Start Time (%/s) 100.0/1.8 99.1/1.7 99.4/1.8
Playouts without InterrupƟons (%) 100.0 100.0 99.7
Ø Video ResoluƟon (p) 613 657 618
Bludenz
Imst
Zirl
Worgl
Zell am See
Sankt Johann
im Pongau
Bischofshofen
Hallein
Seewalchen
am AƩersee
Laarkirchen
Kremsmunster
Enns
Melk Brunn
am Gebirge
Koƫngbrunn
Gleisdorf
Völkermarkt
Velden
am Wörthersee
SpiƩal an
der Drau
Kufstein
Linz
Traun
St. Pölten
Baden
Salzburg
Feldkirch
Villach
Wien
Graz
Klagenfurt
Innsbruck
Drivetest
Walktest
Roads
Train
In the walk tests that our
­testing teams conducted in city
centres and public buildings, all
three candidates scored almost
equally on a very high level. Last
year‘s winner is slightly ahead in
the discipline of voice connec­
tions in rural ­regions, namely in
smaller towns and on connecting
roads. But even there, the ­overall
difference between the three
networks is only minuscule.
Data communication
When we look at the tests of
data connections, the results
are very similar. Here again,
A1 shows a slight advance in
­larger cities with Drei following
closely.
For web page access tests,
T-Mobile Austria falls a fraction
behind – but still offers top
­results that might be totally
­sufficient for a test win in some
other countries.All in all, the
­success ratios, session times
and data throughputs that can be
seen in the adjacent ­tables are
fantastic values.
In the inner city walk tests,
Drei becomes the frontrunner by
a very thin margin. But here
again, the two other providers
follow at a distance of only a ­­
few points.
As we could already observe
in the voice tests, Drei scores
slightly better than both of its
competitors in small towns and
on the connecting roads. But, as
in the other disciplines before,
the contest takes place at a very
high level. >>
Austria
OPERATOR A1 Drei T-Mobile
VOICE (CiƟes; Drivetest)
Call Success RaƟo (%) 99.0 98.6 98.9
Call Setup Time (s) 3.5 4.6 5.7
Speech Quality (MOS-LQO) 3.9 3.8 3.8
VOICE (CiƟes; Walktest)
Call Success RaƟo (%) 99.5 99.8 99.8
Call Setup Time (s) 3.4 4.5 5.6
Speech Quality (MOS-LQO) 4.0 3.9 3.8
VOICE (Towns; Drivetest)
Call Success RaƟo (%) 97.9 99.1 98.8
Call Setup Time (s) 3.7 4.6 5.8
Speech Quality (MOS-LQO) 3.8 3.8 3.8
VOICE (Roads; Drivetest)
Call Success RaƟo (%) 97.3 99.0 98.0
Call Setup Time (s) 3.9 4.6 6.8
Speech Quality (MOS-LQO) 3.7 3.8 3.7
VOICE (Train; Walktest)
Call Success RaƟo (%) 93.8 92.1 91.5
Call Setup Time (s) 4.0 4.7 6.7
Speech Quality (MOS-LQO) 3.6 3.7 3.7
600 of1000 Points400 of 1000 Points
VOICE DATA
of180
of270
of80
of120
of50
of75
of60
of90
88%
93%
84%
A1 Drei T-Mobile
87%
92%
87%
97%
96%
93%
94%
89%
88%
91%
92%
83%
of30
of45
82%
77%
71%
79%
81%
77%
89%
93%
89%
95%
96%
90%
93%
91%
95%
CiƟes
Drivetest
CiƟes
Walktest
Towns
Drivetest
Roads
Drivetest
Train
Walktest
600 of1000 Points400 of 1000 Points
VOICE DATA
of180
of270
of80
of120
of50
of75
of60
of90
88%
93%
84%
A1 Drei T-Mobile
87%
92%
87%
97%
96%
93%
94%
89%
88%
91%
92%
83%
of30
of45
82%
77%
71%
79%
81%
77%
89%
93%
89%
95%
96%
90%
93%
91%
95%
CiƟes
Drivetest
CiƟes
Walktest
Towns
Drivetest
Roads
Drivetest
Train
Walktest
600 of1000 Points400 of 1000 Points
VOICE DATA
of180
of270
of80
of120
of50
of75
of60
of90
88%
93%
84%
A1 Drei T-Mobile
87%
92%
87%
97%
96%
93%
94%
89%
88%
91%
92%
83%
of30
of45
82%
77%
71%
79%
81%
77%
89%
93%
89%
95%
96%
90%
93%
91%
95%
CiƟes
Drivetest
CiƟes
Walktest
Towns
Drivetest
Roads
Drivetest
Train
Walktest
network test
98 1/2017
Single review In spite of its strong competitors, Swisscom ­habitually ranked
first in Switzerland. However, this year things are a little different.
In the connect network test, the
Swiss operators have had sur­
prises in store time and again.
Traditionally, the bar is set extre­
mely high in Switzerland – quite
often all three network providers
achieved the grade “very good“.
Of course we have raised our
requirements once again this year
– and still the present test winner
succeeded in climbing just a little
over the 950 point mark within
our 1000 point scoring scheme.
Thus, for the first time in the net­
work test for Germany,Austria
and Switzerland, we have to
award the grade “outstanding“.
This is short of a sensation and
it goes along with a surprising
change at the top. But the other two
Helvetian candidates still are gra­
ded “very good“ – and their achie­
vements are far from being close
calls. But one thing after another.
Voice connections
Very much the same as in the
other countries, the testing staff
of P3 examined the quality and
stability of voice connections in
Switzerland by conducting drive
tests and walk tests. ­During
­these tests, Sunrise quickly
­turned out to be ahead in most
of the examined scena­rios.
This is quite clear for phone calls
out of cars in larger ­cities. But in
more rural areas (smaller towns
as well as on the tested connec­
ting roads), the gap between
Sunrise and Swisscom shrinks
down to one or two points.
In the results of our walk
tests conducted in large Swiss
cities, the competitors Sunrise
and Swisscom are actually on
par.
Salt keeps some distance to
the two leading contestants, but
still achieves very good results.
While Sunrise and Swisscom
­improved over last year‘s results,
the third Swiss operator that was
formerly known as “Orange“
­more or less stays at the same
level than last year.
Up to now, Swisscom is the
only operator in Switzerland who
supports VoLTE.This modern
voice standard that was applied
for a part of the test calls, con­
tributes to Swisscom‘s excellent
results in the voice category.
How­ever, in the final scoring
­Sunrise stays still close ahead.
Data communications
The standings and tendencies
that we could observe in the
voice category repeat them­selves
almost identically in the data
­­discipline. Sunrise accounts
for the biggest improvement
over last year‘s results in this
cate­gory.This is quite obvious in
the larger cities where Sunrise‘s
lead over the also very strong
Swisscom is a little more pro­
nounced in the drive tests than
in the walk tests.
For example, Sunrise achieves
impressive success ­rations of
100 per cent for file downloads
or Youtube playbacks in the cars.
In contrast, in the smaller towns
and on connecting roads, >>
Zürich
Luzern
Uster
Schaĭausen
Thun
Lugano
Lausanne
Yverdon-les-Bains
Winthertur
Bern
Geneve
Basel
Chur
Cham
Baden
Dubendorf
Volketswil
EffreƟkon
DieƟkon
Sarnen
Bellinzona
Spiez
Bulle
Vevey
Renens
Morges
Gland
Meyrin
Carouge
Worb
Solothurn
PraƩeln
MuƩenz
Drivetest
Walktest
Roads
Train
OPERATOR A1 Drei T-Mobile
DATA (Train; Walktest)
Web-Page Download (Live/StaƟc)
Success RaƟo (%/%) 92.2/94.0 91.9/93.0 92.1/94.1
Ø Session Time (s/s) 2.6/1.7 2.5/1.3 2.8/1.7
File Download (3 MB)
Success RaƟo/Ø Session Time (%/s) 97.4/2.4 92.2/1.3 95.2/2.8
90%/10% faster than (kbit/s) 6536/51337 13429/57362 4678/43353
File Upload (1 MB)
Success RaƟo/Ø Session Time (%/s) 91.7/3.2 93.3/1.6 88.8/1.9
90%/10% faster than (kbit/s) 862/13106 2782/13785 1903/17779
File Download (10 Seconds)
Success RaƟo (%) 95.4 92.4 96.2
Ø Throughput (kbit/s) 34828 45022 25450
90%/10% faster than (kbit/s) 6889/73222 16495/78822 4906/52455
File Upload (10 Seconds)
Success RaƟo (%) 92.5 90.4 94.3
Ø Throughput (kbit/s) 17759 25002 14753
90%/10% faster than (kbit/s) 735/39427 3200/42636 1401/29780
Youtube Videos
Success RaƟo/Start Time (%/s) 94.3/2.0 92.0/1.9 89.7/2.0
Playouts without InterrupƟons (%) 100.0 100.0 100.0
Ø Video ResoluƟon (p) 576 630 593
It was well worth
the effort: In this
year‘s connect
network test,A1 takes back
the gold medal from last
year‘s winner Drei. Particu­
larly good voice results make
A1 the overall winner in the
Alpine Republic. But equally in
terms of data communica­
tions and connectivity in rail­
ways, the A1 network scores
very strong. Moreover, with a
total score of 918 out of a
possible maximum of 1000
A1 is also a top tier when
comparing the results from
all three involved countries.
Compared ­­
to last year,
T-Mobile Austria took the
­biggest step forward.
The actual reason that this
contender still scores razor-
thin behind its competitors
for most of the ­in­dicators is
the extreme strength of all
Austrian providers. Still,
T-Mobile Austria turns out to
be in excellent shape. Its
total result would have made
this provider a strong number
two in Germany and would
actually come quite close to
the performance of the parent
company based there.
Drei was also ­able
to improve on last
year‘s results on
the whole. But it is
by a very close margin of only
three points, that the Hutchi­
son-owned provider makes
second place. Drei scores
better than its competitors
particularly in smaller towns
and on connecting roads.
­Examining mobile connectivity
in trains, the overall winner
A1 and Drei are basically on
par.And bear in mind that a
second place in Austria would
be equivalent to a test win in
many other countries.
Mobile communications
on Austrian railways
When they talk on the phone or
surf the web in trains,Austrian
customers have once more good
reason to be pleased.
The measurement values that
P3‘s teams gathered on hund­
reds of railway kilometres, certify
very good results for the three
Austrian operators – even if
­indicators like success ratios or
data rates somewhat drop in this
category when for instance com­
pared to those obtained on the
connecting roads.
A1 scores best for conveying
voice calls in trains, while Drei
turns out to be the data cham­
pion. Regarding voice telephony
in trains,T-Mobile falls slightly
behind, while this is not true for
data communications ­– where
this operator offers absolutely no
cause for complaint.
On the whole, in the railway
category Austria scores some­
what behind Switzerland, but is
clearly ahead of the results from
Germany.
918 915 876TOTAL
very good very good very good-RaƟng
Shown voice, data and total scores are rounded.
VOICE
max. 400
max. 1000 Points
DATA
max. 600
365 360 346
553
530555
A1 Drei
T-Mobile
Switzerland
OPERATOR Sunrise Swisscom Salt
VOICE (CiƟes; Drivetest)
Call Success RaƟo (%) 99.8 99.1 99.2
Call Setup Time (s) 3.4 3.3 5.0
Speech Quality (MOS-LQO) 3.9 3.9 3.5
VOICE (CiƟes; Walktest)
Call Success RaƟo (%) 99.9 99.7 99.1
Call Setup Time (s) 3.3 3.3 4.9
Speech Quality (MOS-LQO) 3.9 4.0 3.5
VOICE (Towns; Drivetest)
Call Success RaƟo (%) 99.9 99.3 99.5
Call Setup Time (s) 3.6 3.3 5.0
Speech Quality (MOS-LQO) 3.8 3.9 3.4
VOICE (Roads; Drivetest)
Call Success RaƟo (%) 99.3 98.5 96.2
Call Setup Time (s) 3.7 3.4 5.3
Speech Quality (MOS-LQO) 3.8 3.9 3.4
VOICE (Train; Walktest)
Call Success RaƟo (%) 98.3 97.1 96.2
Call Setup Time (s) 4.0 3.6 5.2
Speech Quality (MOS-LQO) 3.8 3.8 3.4
network test
1110 1/2017
OPERATOR Sunrise Swisscom Salt
DATA (CiƟes; Drivetest)
Web-Page Download (Live/StaƟc)
Success RaƟo (%/%) 99.8/99.9 98.4/99.3 99.0/99.2
Ø Session Time (s/s) 2.5/1.3 2.4/1.2 2.9/1.6
File Download (3 MB)
Success RaƟo/Ø Session Time (%/s) 100.0/1.2 99.8/1.0 99.7/2.0
90%/10% faster than (kbit/s) 13796/58451 16771/62827 6805/37891
File Upload (1 MB)
Success RaƟo/Ø Session Time (%/s) 99.7/0.8 99.1/0.8 98.8/1.5
90%/10% faster than (kbit/s) 6819/26846 6375/27778 2821/13722
File Download (10 Seconds)
Success RaƟo (%) 99.9 99.1 99.6
Ø Throughput (kbit/s) 56170 65199 38994
90%/10% faster than (kbit/s) 17097/98956 18817/123092 8781/77076
File Upload (10 Seconds)
Success RaƟo (%) 99.7 99.1 99.7
Ø Throughput (kbit/s) 27109 26860 20218
90%/10% faster than (kbit/s) 8022/45763 7070/44770 4219/39494
Youtube Videos
Success RaƟo/Start Time (%/s) 100.0/1.5 99.5/1.4 99.9/1.6
Playouts without InterrupƟons (%) 100.0 100.0 100.0
Ø Video ResoluƟon (p) 676 674 646
DATA (CiƟes; Walktest)
Web-Page Download (Live/StaƟc)
Success RaƟo (%/%) 99.5/99.7 97.9/99.1 98.8/98.8
Ø Session Time (s/s) 2.6/1.4 2.5/1.2 2.9/1.5
File Download (3 MB)
Success RaƟo/Ø Session Time (%/s) 99.6/1.0 99.6/0.8 99.8/2.2
90%/10% faster than (kbit/s) 16891/60560 20599/70012 5593/38388
File Upload (1 MB)
Success RaƟo/Ø Session Time (%/s) 99.6/1.0 99.2/0.7 99.0/1.3
90%/10% faster than (kbit/s) 4837/26144 7741/28070 3364/15009
File Download (10 Seconds)
Success RaƟo (%) 99.8 98.8 99.8
Ø Throughput (kbit/s) 59246 72930 43371
90%/10% faster than (kbit/s) 17024/117345 23343/138987 8045/76761
File Upload (10 Seconds)
Success RaƟo (%) 99.6 100.0 99.6
Ø Throughput (kbit/s) 27846 30210 24091
90%/10% faster than (kbit/s) 4983/46361 10290/45322 5097/45404
Youtube Videos
Success RaƟo/Start Time (%/s) 100.0/1.6 99.6/1.4 99.4/1.6
Playouts without InterrupƟons (%) 100.0 100.0 100.0
Ø Video ResoluƟon (p) 677 681 648
DATA (Towns; Drivetest)
Web-Page Download (Live/StaƟc)
Success RaƟo (%/%) 99.5/99.3 98.9/99.5 99.3/99.4
Ø Session Time (s/s) 2.6/1.4 2.4/1.3 2.8/1.6
File Download (3 MB)
Success RaƟo/Ø Session Time (%/s) 100.0/1.2 99.5/1.0 99.8/1.7
90%/10% faster than (kbit/s) 11404/56272 17905/63141 8696/38326
File Upload (1 MB)
Success RaƟo/Ø Session Time (%/s) 99.5/1.1 99.8/0.9 99.0/1.6
90%/10% faster than (kbit/s) 4286/25438 5276/26499 2487/13647
File Download (10 Seconds)
Success RaƟo (%) 100.0 99.8 99.5
Ø Throughput (kbit/s) 50685 57979 39136
90%/10% faster than (kbit/s) 13006/95800 16417/109851 8924/77090
File Upload (10 Seconds)
Success RaƟo (%) 100.0 99.5 99.2
Ø Throughput (kbit/s) 22788 25217 19457
90%/10% faster than (kbit/s) 5822/44056 7049/42839 2704/39110
Youtube Videos
Success RaƟo/Start Time (%/s) 100.0/1.6 99.7/1.4 99.5/1.6
Playouts without InterrupƟons (%) 100.0 100.0 100.0
Ø Video ResoluƟon (p) 675 670 646
OPERATOR Sunrise Swisscom Salt
DATA (Roads; Drivetest)
Web-Page Download (Live/StaƟc)
Success RaƟo (%/%) 98.6/99.0 96.3/98.3 96.6/97.9
Ø Session Time (s/s) 2.6/1.4 2.5/1.3 2.8/1.7
File Download (3 MB)
Success RaƟo/Ø Session Time (%/s) 99.2/1.7 99.0/1.1 98.4/2.1
90%/10% faster than (kbit/s) 8734/60333 13263/64971 8049/39177
File Upload (1 MB)
Success RaƟo/Ø Session Time (%/s) 99.0/1.3 98.9/1.2 96.4/1.8
90%/10% faster than (kbit/s) 3119/21623 3439/26756 2109/13629
File Download (10 Seconds)
Success RaƟo (%) 99.7 98.0 98.4
Ø Throughput (kbit/s) 46109 59593 47507
90%/10% faster than (kbit/s) 11612/100874 16575/112413 10961/88886
File Upload (10 Seconds)
Success RaƟo (%) 98.7 99.0 96.2
Ø Throughput (kbit/s) 17638 23561 20622
90%/10% faster than (kbit/s) 4035/37289 4060/44437 2944/41313
Youtube Videos
Success RaƟo/Start Time (%/s) 99.7/1.6 99.3/1.5 97.6/1.6
Playouts without InterrupƟons (%) 100.0 100.0 99.8
Ø Video ResoluƟon (p) 664 675 651
OPERATOR Sunrise Swisscom Salt
DATA (Train; Walktest)
Web-Page Download (Live/StaƟc)
Success RaƟo (%/%) 97.2/97.5 96.7/98.4 96.4/98.6
Ø Session Time (s/s) 2.8/1.7 2.7/1.6 3.1/2.0
File Download (3 MB)
Success RaƟo/Ø Session Time (%/s) 98.9/2.2 97.8/1.9 98.9/3.7
90%/10% faster than (kbit/s) 6959/44594 5299/53097 3173/31360
File Upload (1 MB)
Success RaƟo/Ø Session Time (%/s) 97.1/1.5 97.8/1.0 98.2/1.6
90%/10% faster than (kbit/s) 2910/23022 5907/24406 2957/13769
File Download (10 Seconds)
Success RaƟo (%) 97.1 97.8 99.6
Ø Throughput (kbit/s) 31914 40576 21238
90%/10% faster than (kbit/s) 7834/65680 11614/85242 3799/46908
File Upload (10 Seconds)
Success RaƟo (%) 98.2 98.2 99.3
Ø Throughput (kbit/s) 20141 22821 19025
90%/10% faster than (kbit/s) 2518/39363 6755/36717 4275/35336
Youtube Videos
Success RaƟo/Start Time (%/s) 98.8/1.8 100.0/1.8 97.2/1.9
Playouts without InterrupƟons (%) 100.0 100.0 99.2
Ø Video ResoluƟon (p) 645 662 604
Single review
Both in
the voice and data measure­
ments, Sunrise achieves ex­
cellent results. Compared to
the previous year, the opera­
tor improved in both catego­
ries. Due to a distinct gain in
points in data communica­
tions, Sunrise brings in the
overall win. Interestingly, the
gap to last year‘s winner is
quite wide – although Swiss­
com achieved very good results
this time as well. For the first
time in the connect mobile net­
work test for Germany,Austria
and Switzerland we have to
award the grade “outstanding“.
The smal­
lest Helve­
tic mobile network provider
achieves stable – and very
good – overall results, both
in the voice and the data
­categories.
As they basically ­remain
at the same level than in the
­previous year, Salt comes
in third in the overall Swiss
ranking.
But when we compare the
results of all three countries,
Salt is still a top tier. More­
over, Salt attacks both of its
competitors quite successfully
with its aggressive pricing.
The
test
results
clearly show that last year‘s
winner Swisscom has impro­
ved as well during the last
twelve months.
However, in the final sco­
ring Swisscom was beaten by
the second largest contender
Sunrise.
Nevertheless, Swiss­com‘s
customers can rest assured
because with its excellent
test results this Swiss ope­
rator would still immediately
lead the field in Germany
or Austria.
951 933 878TOTAL
outstanding very good very good-RaƟng
Shown voice, data and total scores are rounded.
VOICE
max. 400
max. 1000 Points
DATA
max. 600
390 380 352
561
526
553
Sunrise Swisscom
Salt
600 of 1000 Points400 of 1000 Points
VOICE DATA
of180
of270
of80
of120
of50
of75
of60
of90
96%
94%
79%
Sunrise Swisscom Salt
98%
96%
91%
99%
98%
90%
98%
95%
90%
94%
92%
87%
of30
of45
94%
92%
84%
90%
93%
87%
93%
93%
88%
93%
93%
88%
92%
88%
94%
CiƟes
Drivetest
CiƟes
Walktest
Towns
Drivetest
Roads
Drivetest
Train
Walktest
600 of 1000 Points400 of 1000 Points
VOICE DATA
of180
of270
of80
of120
of50
of75
of60
of90
96%
94%
79%
Sunrise Swisscom Salt
98%
96%
91%
99%
98%
90%
98%
95%
90%
94%
92%
87%
of30
of45
94%
92%
84%
90%
93%
87%
93%
93%
88%
93%
93%
88%
92%
88%
94%
CiƟes
Drivetest
CiƟes
Walktest
Towns
Drivetest
Roads
Drivetest
Train
Walktest
the leading duo is almost on par
– once again on a very high level.
In the data measurements,
Salt has to settle for the third
rank once more – but this again
only means that Sunrise and
Swisscom were able to improve
over last year‘s results, while Salt
kept its still strong performance
stable. In this context, we should
bear in mind that Salt attacks its
two strong competitors with a
particularly aggressive pricing.
This makes Salt‘s offerings all
the more interesting.Above all,
Salt‘s results are definitely top
notch when compared to those
from other countries.
Mobile connectivity
in Swiss railways
The tendencies observed in
the voice and data measure­
ments conducted in larger cities
and smaller towns as well as on
the connecting roads, prevail for
the tests of phone calls and data
connectivity in Swiss railways too.
Again, in this category Swiss
customers have every reason to
be happy.Their operators achieve
the best results within the three
countries at a distinct ­distance.
Indeed all three Swiss opera­
tors achieve a remarkably high
performance and reliability in the
challenging task of providing
connectivity to moving trains in
the demanding Helvetic topology.
When looking at the detailed
results in railway tests, Sunrise
is slightly ahead in the voice ca­
tegory, while Swisscom scores a
tad higher in the data measure­
ments. Salt again comes in third
regarding both voice and data
communications while still show­
ing very good results.
600 of 1000 Points400 of 1000 Points
VOICE DATA
of180
of270
of80
of120
of50
of75
of60
of90
96%
94%
79%
Sunrise Swisscom Salt
98%
96%
91%
99%
98%
90%
98%
95%
90%
94%
92%
87%
of30
of45
94%
92%
84%
90%
93%
87%
93%
93%
88%
93%
93%
88%
92%
88%
94%
CiƟes
Drivetest
CiƟes
Walktest
Towns
Drivetest
Roads
Drivetest
Train
Walktest
600 of 1000 Points400 of 1000 Points
VOICE DATA
of180
of270
of80
of120
of50
of75
of60
of90
96%
94%
79%
Sunrise Swisscom Salt
98%
96%
91%
99%
98%
90%
98%
95%
90%
94%
92%
87%
of30
of45
94%
92%
84%
90%
93%
87%
93%
93%
88%
93%
93%
88%
92%
88%
94%
CiƟes
Drivetest
CiƟes
Walktest
Towns
Drivetest
Roads
Drivetest
Train
Walktest
network test
1312 1/2017
As in previous years, connect’s
partner for the network measure­
ments, P3 communications, used
two vehicles to test drive the cho­
sen cities, towns and roads. In
Germany and Austria each car
carried six Samsung Galaxy S5
smartphones to measure voice
services and three Samsung Ga­
laxy Note 4 performing the data
service tests. In order to reflect
the advanced roll-out of LTE with
“3 Carrier Aggregation“ (the
combination of three carrier fre­
quencies) in Switzerland, we
used three Samsung Galaxy S7
for the data measurements there.
The same setup of devices was
utilized in the walk tests. For this
effort, the smartphones were
­installed in trolleys and back­
packs with additional batteries.
The devices’ firmware was
each operator’s current firmware
version. If such software was not
available the most current firm­
ware from Samsung was used.
Voice telephony
Voice services were measured
with the smartphones performing
calls alternating between the two
measurement cars (“mobile-to-
mobile“).An additional car served
as a mobile remote station for the
calls of the walk test teams.
Background data traffic was
transmitted by one of the smart­
phones simultaneously to each
call to reflect a realistic usage
scenario.Audio quality was as­
sessed by using POLQA (Percep­
tual Objective Listening Quality
Assessment) wide band scoring.
All devices were configured in
“LTE preferred” mode.Thus in
the three German Networks as
well as with A1 in Austria and
Swisscom in Switzerland, the
modern Voice-over-LTE (VoLTE)
service could be used.Within
networks not yet supporting
­VoLTE, the smartphones were
forced to switch to 3G or 2G
technology, the so-called circuit-
switched-fall-back (CSFB).
Data connectivity
To assess cellular data perfor­
mance a sequence of tests were
executed.As a dynamic web-
browsing test, each country’s top
web sites (according to the Alexa
ranking) were downloaded in the
so-called live web-browsing test.
Additionally a static web site was
tested, the industry standard ETSI
(European Telecommunications
Standards Institute) “Kepler“
­reference page. HTTP downloads
and uploads were performed with
3 MB and 1 MB files, simulating
small file transfers.The networks’
peak performance was tested
with a ten second download and
upload of a single, very large file.
The Youtube measurements
considered the new “adaptive
­resolution“ feature of this video
platform. In order to offer a per­
sistent video experience,Youtube
adapts the video streams‘ reso­
lution dynamically to the band­
width that is currently available.
Our scoring therefore considers
the success ratio, the time until
the playback starts, the percen­
tage of video playouts that take
place without interruptions as
well as the videos‘ average
­resolution or line number count
respectively.
Indoor and train
measurements
The walk tests consisted of the
same tasks as were performed in
the cars. For this effort two teams
measured in public transport and
in public places, like coffee
shops, museums, train stations
and airport terminals.Travelling
from city to city by public trans­
port allowed the assessment of
cellular network quality within
the long distance trains.
Logistics
The tests were performed in
­Austria, Germany and Switzer­
land around the same period of
time (Germany: October 21 –
­November 12; Austria: October 7
– 27; Switzerland: October 14
– November 1).All measure­
ments were done between 8 AM
and 10 PM. Both cars were
­­always in the same cities, but on
different routes to avoid any in­
terference of one car’s measure­
ment by the other car’s. Both
­vehicles followed a given route,
including fixed location measure­
ments at “areas of interest” such
as well-visited public places.
Measurements there lasted one
hour. Locations such as train
­stations, airports, much-frequen­
ted public parks or high-density
urban areas typically demonstrate
how networks respond when a
high number of users compete
for their share of bandwidth
within the network’s available
­radio frequencies.
The measurements included
17 larger cities and 26 smaller
towns in Germany, while the walk
tests frequented six cities. In
Austria the drive tests covered 11
big cities and 20 smaller towns,
the walk test team visited five
­cities. In Switzerland, the test
route included 13 big cities and
20 smaller towns with the walk
tests conducted in four cities.
Travel between the cities mainly
used highways, but smaller state
and county roads were driven as
well. For each connect test P3
communications follows a well-
defined process to generate four
independent and representative
city and route plans.The connect
editors choose randomly one of
these four alternatives.
Test efforts and results
Overall 25,000 km were driven
for the connect P3 mobile net­
work test in 2016. In Germany
the approximately 12,100 km of
driven routes alongside the cities
and areas visited represent 13.4
million inhabitants, equaling
METHODOLOGY
Professional and critical: Bernd Theiss, head of test and technology at connect (on
the left), and Hakan Ekmen, managing director of P3 communications (on the right).
Four Samsung Galaxy Note 4
­­measured the data performance in
German and Austrian networks.
Here are the reasons why we tested and evaluated the
merging networks of E-Plus and O2 as a single O2 network.
After Telefónica/O2 bought
out its former competitor
E-Plus in October 2014, the
merger of both networks
goes at full speed. Previous
E-Plus customers are being
transferred to O2 tariffs, and
Telefónica must sell off some
base stations that have been
occupied by both operators
according to the German
­regulatory authority.The re­
maining base station sites
are already designated “O2“.
Cells formerly belonging to
E-Plus are no longer visible
as a discrete mobile network.
Instead, at the moment there
are “old“ O2/E-Plus cells
along with “new“ ones.
Given this situation, con­
nect and P3 decided to only
examine O2 in their network
test that we con­ducted in
late 2016.
As we know from our
­readers and from our own
experiences, difficulties
­definitely ­occur in the course
of the network merge.
These problems that include
failing handovers between
two differently ­configured
network cells, are clearly
­recognized in the results
of this year‘s P3 connect
mobile network test.
o2 And E-plus
A merger with some obstacles:
Combining two separate mo­
bile networks into one is far
from being routine work.
U­navoidable problems that are
resulting from this endeavour
are clearly recognized by our
test results.
around 16.7 per cent of
Germany’s population. ­Austria
was measured by driving
5,900 km covering about
3 million inhabitants (approx.
36 per cent of the Austrian
population). In Switzerland,
the test teams drove approx.
7,000 km, covering 1.9
­million people representing
around 22.5% of the Swiss
population. Certainly a huge
effort, but necessary to gain
the required statistical rele­
vance and confidence in the
test results.
Scoring
The results of the voice test
contribute 40 per cent of the
total score, those of the data
tests make up 60 per cent.
For the overall result we apply
a 1000 point scheme in order
to represent sufficiently detailed
results.
Moreover this scheme allows
us to better compare the results
of network tests that we have
conducted in different countries
(find all results and additional
information at www.connect-
testmagazine.com).
network test
14 1/2017
fairness AND transparency
This year, some of the candidates massively tried to influence the conditions and parameters in the run-up of our
test. The connect and P3 staff responsible for the testing project have of course fended off these attempts.
As in previous years, connect and P3 met in
early 2016 in order to define the conditions
and parameters for this year‘s network test.
In this preceding test design phase, we for
example identify new test criteria, discard or
confirm old ones and determine their influ­
ence on the overall score.We define the
timeframe as well as a preselection of
smartphone models that we intend to use
for the measurements.We then communi­
cate these preliminary ­definitions in advance
to the CTOs of the network operators.
Feedback is appreciated
In this process we appreciate feedback about
aspects like suitable tariffs that facilitate un­
obstructed measurements of the best perfor­
mance possible.After all, our objective is to
evaluate the network experience of the most
demanding customers.We also agree on the
firmware versions used in the measurement
smartphones, as each mobile network operator
makes ­adjustments to most popular devices
to ­ensure a smooth interplay with their network.
But this time, some contenders apparently
took part in the discussions with the single
intent to enforce measurement conditions
that would favour their own network. For
­example, there have been attempts to im­
pose a smartphone model on us that all in all
works less reliably than others – presumably
because the involved provider expected an
­advantage for its own network from this.
One operator insinuated flaws in the test
design more than once – until extensive
measurements conducted both by P3 com­
munications as well as by the connect test
lab disproved all of them. Permanent chan­
ges in the reasonings of some operators led
connect to the assump­tion that one or the
other of them would not have minded blowing
the rapidly ­approaching deadline of our test.
The more danger, the more honour
We cannot help but understand such
­attempts as a compliment for the high rele­
vance that the operators assign to our test.
And of course we remain true to ourselves
concerning these issues.After all, it is our
standard to conduct a test that provides
deep insights into the quality and perfor­
mance of the examined mobile networks.
However, we will draw one conclusion from
this year‘s experience: In the future, we will
publish obvious attempts to abuse our trans­
parent approach to testing the very same
way as we document our test procedures.
Looking back at the results of the connect network tests since 2010 provides especially one ­insight:
Despite of the constantly rising requirements, the level of the overall results has steadily ­improved.
We reckon that our demanding and well renowned network test is not entirely blameless.
Historical Development
Customers‘ expectations are
constantly growing – expan­
ding data volumes and rising
transmission speeds are re­
garded to be absolutely normal.
P3 and connect take account
of this development by con­
stantly raising the requirements
and thresholds of our tests.
Network test
as a driving force
The adjoining glance at the
development of results in
­Germany,Austria and Switzer­
land in recent years shows
a clear overall tendency:
­Despite the growing require­
ments, all tested networks
improved steadily.
In all modesty, we believe
that the high relevance and
­challenging demands of our
annual network tests are an
important driving force of
this development.
conclusion
Hannes Rügheimer,
connect author
The operators enthusiastically fight
for the top rank in the connect net-
work test. The fact that almost all
candidates managed to improve in
spite of the rising requirements is
clearly supporting our claim that our
critical tests contribute to the overall
enhancement of the mobile net-
works’ quality.
Against this background, the
­repeated test victory of Deutsche
Telekom in Germany was by no
­means self-evident. It rather reflects
the considerable efforts that Tele-
kom takes in order to maintain and
extend its network. Vodafone also
worked flat out, but remains on the
second rank. O2’s result shows
some room for development but can
be explained by the ongoing integra-
tion with the former E-Plus network.
In the alpine countries,there were
rigorous fights as well. This led to a
change at the top ranks in both
countries. In Austria,A1 managed to
gain back the crown from last year’s
winner Drei. Particularly its very
strong voice results secured the win
to A1. Although, also Drei noticeably
improved over last year, the Hutchi-
son-owned company fell back be-
hind A1 at a very narrow gap. Com-
pared to the previous year, T-Mobile
Austria made the biggest step for-
ward,but still was not able to pass its
two extremely strong competitors.
In Switzerland we also see some
movement at the top.Especially with
its distinct rise in points in the data
category, Sun­rise manages to out-
play last year’s winner Swisscom.
Thus,for the first time a candidate of
theconnectnetworktestin­Germany,
Austria and Switzerland is awarded
the grade “outstanding“. And even
though Swisscom only ranks second
this ­time, the company has still im-
proved considerably compared to the
pre­vious year.Salt takes the third rank
of the Helvetic network providers but
still achieved very good results.
The top quality of mobile connecti-
vity on Swiss and also on Austrian
railways was especially noteworthy.
This is something that German rail-
way customers can only dream about.
GERMANY AUSTRIA SWITZERLAND
Overall Results Voice and Data Telekom Vodafone Telefónica A1 Drei T-Mobile Sunrise Swisscom Salt
VOICE max. 400 Points 366 352 270 365 360 346 390 380 352
CiƟes Drivetest 180 94% 92% 71% 94% 89% 88% 98% 95% 90%
CiƟes Walktest 60 95% 90% 76% 97% 96% 93% 99% 98% 90%
Towns Drivetest 80 95% 91% 78% 87% 92% 87% 98% 96% 91%
Roads Drivetest 50 92% 88% 49% 88% 93% 84% 96% 94% 79%
Train Walktest 30 57% 54% 33% 82% 77% 71% 94% 92% 84%
DATEN max. 600 Points 521 489 373 553 555 530 561 553 526
CiƟes Drivetest 270 90% 87% 68% 95% 93% 91% 94% 92% 88%
CiƟes Walktest 90 87% 84% 56% 95% 96% 90% 93% 93% 88%
Towns Drivetest 120 87% 81% 65% 89% 93% 89% 93% 93% 88%
Roads Drivetest 75 89% 82% 62% 91% 92% 83% 94% 92% 87%
Train Walktest 45 61% 46% 30% 79% 81% 77% 90% 93% 87%
Total max. 1000 Points 887 841 643 918 915 876 951 933 878
-RATING very good good sufficient very good very good very good outstanding very good very good
Sunrise
Swisscom
Salt
40%
60%
80%
100%
2010 2011 2012 2013 2014 2015 2016
Vodafone
Telefónica (O2)
Telefónica (E-Plus)
Telekom
40%
60%
80%
100%
2010 2011 2012 2013 2014 2015 2016
Drei
T-Mobile
A1
40%
60%
80%
100%
2010 2011 2012 2013 2014 2015 2016
887 841 643TOTAL
very good good sufficient-RaƟng
Shown voice, data and total scores are rounded.
VOICE
max. 400
max. 1000 Points
DATA
max. 600
366 352 270
521
373
489
Telekom Vodafone
Telefónica
918 915 876TOTAL
very good very good very good-RaƟng
Shown voice, data and total scores are rounded.
VOICE
max. 400
max. 1000 Points
DATA
max. 600
365 360 346
553
530555
A1 Drei
T-Mobile
951 933 878TOTAL
outstanding very good very good-RaƟng
Shown voice, data and total scores are rounded.
VOICE
max. 400
max. 1000 Points
DATA
max. 600
390 380 352
561
526
553
Sunrise Swisscom
Salt
TESTSIEGER TESTSIEGER TESTSIEGERBEST IN TEST BEST IN TEST BEST IN TEST
network test
1716 1/2017

More Related Content

PDF
Disruptive innovation and an exemplary Cx
PDF
Telefonica Internal Conference 2012
PDF
北投地質調查和導覽計畫
PPT
Factpendientesmod2 presentacin1ingles2muestra 090221132635-phpapp02
PDF
AbMole Inhibitor catalog July 2014
PDF
SurveyofBP
DOCX
青平台參與式預算 審議團隊招募簡章
PPTX
Indeed ID - One minute presentation.
Disruptive innovation and an exemplary Cx
Telefonica Internal Conference 2012
北投地質調查和導覽計畫
Factpendientesmod2 presentacin1ingles2muestra 090221132635-phpapp02
AbMole Inhibitor catalog July 2014
SurveyofBP
青平台參與式預算 審議團隊招募簡章
Indeed ID - One minute presentation.

Viewers also liked (14)

PPTX
Social Media
PDF
愛走動愛相隨
PDF
友善步道 彩繪北投
PPTX
The Pursuit of Happiness: What Americans Do For Fun
PDF
台北市北投區青少年課餘學習平台計畫
DOCX
Bismillaahi
PDF
Los Alamos Paper
PDF
為愛走唱 行動那卡西
PPTX
IT Automatisierung mit Salt Stack | Thorsten Kramm @ FrosCon 2014
PDF
Multi Model Performance Improvement
PDF
How Fit Flops Work
PDF
北投參與式預算提案 社區環境清潔
PPT
Teaching Current Events in the ESL Classroom
PDF
Pmi conclave15 coopetition as a project strategy
Social Media
愛走動愛相隨
友善步道 彩繪北投
The Pursuit of Happiness: What Americans Do For Fun
台北市北投區青少年課餘學習平台計畫
Bismillaahi
Los Alamos Paper
為愛走唱 行動那卡西
IT Automatisierung mit Salt Stack | Thorsten Kramm @ FrosCon 2014
Multi Model Performance Improvement
How Fit Flops Work
北投參與式預算提案 社區環境清潔
Teaching Current Events in the ESL Classroom
Pmi conclave15 coopetition as a project strategy
Ad

Similar to Netztest-connect-2017-01-ENGLISH-final (20)

PDF
Mobile Speed Connectivity tests in UK
PDF
Fastest mobile networks 2013 pc mag 1
PDF
VoLTE Testing Explained
PDF
VoLTE - New Interconnection Models are coming
PDF
The Netherlands - Analyst & Journalist Briefing 2014
PPTX
The VoLTE User Experience--Better or Worse
PDF
Multimedia Communications - Portugal Telecom Prospective
PPTX
VoLTE.pptx
PDF
Next Generation Service Platforms for Multimedia and Value Added Services
PDF
M&A on the German Telco Market - Status autumn 2013
PDF
Baltics - Analyst & Journalist Briefing 2014
PDF
Quick Summary of LTE Voice Summit 2014 #LTEVoice
PDF
Analyst & Journalist Meeting Tele2 2013 - Joachim Horn - MNO Rollout
PPTX
European survey on consumer preferences in telecom and the digital evolution
PPSX
Presentation on telefonica company
PPTX
German Telecoms Market Q3/2015
PDF
Ericsson ConsumerLab: Smartphone Usage Experience Report
PDF
Special Topic on 5G Evolution
PDF
Mobile Telecoms Tech & Market Disruptions - April 2015 Version
PDF
SDP Global Summit 2009
Mobile Speed Connectivity tests in UK
Fastest mobile networks 2013 pc mag 1
VoLTE Testing Explained
VoLTE - New Interconnection Models are coming
The Netherlands - Analyst & Journalist Briefing 2014
The VoLTE User Experience--Better or Worse
Multimedia Communications - Portugal Telecom Prospective
VoLTE.pptx
Next Generation Service Platforms for Multimedia and Value Added Services
M&A on the German Telco Market - Status autumn 2013
Baltics - Analyst & Journalist Briefing 2014
Quick Summary of LTE Voice Summit 2014 #LTEVoice
Analyst & Journalist Meeting Tele2 2013 - Joachim Horn - MNO Rollout
European survey on consumer preferences in telecom and the digital evolution
Presentation on telefonica company
German Telecoms Market Q3/2015
Ericsson ConsumerLab: Smartphone Usage Experience Report
Special Topic on 5G Evolution
Mobile Telecoms Tech & Market Disruptions - April 2015 Version
SDP Global Summit 2009
Ad

Netztest-connect-2017-01-ENGLISH-final

  • 1. germany austria switzerland The great 2017 mobile network test he network operators in Germany, Aus- tria and Switzerland wait for the results of our annual mobile network benchmark with the highest tension imaginable. As in the years before, we have conducted this benchmark as part of our well-proven co­ operation with the Aachen-based network testing specialist P3 communications. However, we have never before experien­ ced such harsh disputes with some of the test candidates in the forefront of our test. They argued about questions like which smartphone models were to be used for the measure- ments, how the test routes should be put to- gether or how we should balance the indivi- dual voice and data results. Eventually, all their reasoning could be ­traced back to one pur­pose: Some of the candidates hoped to gain advantages in those areas where they believed to be particularly strong and having the edge over their competitors. We did show good sportsmanship inter­ preting the pressure built up by some of the candidates in the run-up of our test as proof for the high relevance and acceptance of our benchmarks within the whole industry. At the end of 2016, we conducted our annual mobile network benchmark for the 23rd time. Together with our renowned benchmarking partner P3 communications, once again we have investigated, which mobile operators in Germany, Austria and Switzerland are a cut above the rest – with utmost effort and our objective, customer-oriented testing methodology. 60% DATA Youtube Quality Success RaƟo Web Page Download File Up- and Download 40% VOICE Call Success RaƟo Call Setup Time Speech Quality Quality benchmark for many years Traditionally, fairness and transparency play a very big role in our network test (also see page 16). In any case, with all decisions con- cerning methodology and scoring, our high- est priority is always to ensure the signifi- cance of our benchmark for you, our readers. As a matter of principle, connect and P3 value objective and authoritative results about the actual quality and performance of the cellular networks second to none. Eve- rybody interested can find an exact descrip- tion of our methodology on pages 14 and 15. So, on the following pages you can read the answers to many thrilling questions: Has Deutsche Telekom been able to defend its top rank in Germany? How does O2 rank, after its owner Telefónica has started to com- bine this network‘s former radio cells with those of E-Plus, which they bought in 2014 (also see page 15). Which operators come out on top inAustria and Switzerland, where the contest is traditionally conducted on an especially high performance level? Not all contestants will like all of the answers and results. This again proves to us that we have done a good job. Because this is the only way for us to guarantee that truly everybody can fully rely on our test results. Hannes Rügheimer Practical relevance in mind In the scoring of our test results, we ac­ count for the steadily growing importance of data communica­ tions. the benchmark Köln Frankfurt am MainWiesbaden München Berlin Düsseldorf Wuppertal Bielefeld Oldenburg Duisburg Essen Kassel Halle Dresden Magdeburg StuƩgart Augsburg Hamburg Drivetest Walktest Roads Train Unna Ascheberg Greven Bramsche Vechta Seevetal Wedemark Bad Oeynhausen Schwerte Soest Brandenburg Dessau-Rosslau Wiedemar Grimma Plauen Hof Bayreuth Lauf an der Pegnitz Pfaffenhofen an der Ilm Dachau Guenzburg Kirchheim unter Teck Idstein Limburg Troisdorf Leonberg gs ruƩga BaB I WW e T Bludenz Imst Zirl Worgl Zell am See Sankt Johann im Pongau Bischofshofen Hallein Seewalchen am AƩersee Laarkirchen Kremsmunster Enns Melk Brunn am Gebirge Koƫngbrunn Gleisdorf Völkermarkt Velden am Wörthersee SpiƩal an der Drau Kufstein Linz Traun St. Pölten Baden Salzburg Feldkirch Villach Wien Graz Klagenfurt InnsbruckZürich Luzern Uster Schaĭausen Thun Lugano Lausanne Yverdon-les-Bains Winthertur Bern Geneve Basel Chur Cham Baden Dubendorf Volketswil EffreƟkon DieƟkon Sarnen Bellinzona Spiez Bulle Vevey Renens Morges Gland Meyrin Carouge Worb Solothurn PraƩeln MuƩenz T SINCE network test 3
  • 2. 600 of 1000 Points400 of 1000 Points VOICE DATA of180 of270 of80 of120 of50 of75 of60 of90 92% 88% 49% Telekom Vodafone Telefónica 95% 91% 78% 95% 90% 76% 94% 92% 71% 89% 82% 62% of30 of45 57% 54% 33% 61% 46% 30% 87% 81% 65% 87% 84% 56% 87% 68% 90% CiƟes Drivetest CiƟes Walktest Towns Drivetest Roads Drivetest Train Walktest 600 of 1000 Points400 of 1000 Points VOICE DATA of180 of270 of80 of120 of50 of75 of60 of90 92% 88% 49% Telekom Vodafone Telefónica 95% 91% 78% 95% 90% 76% 94% 92% 71% 89% 82% 62% of30 of45 57% 54% 33% 61% 46% 30% 87% 81% 65% 87% 84% 56% 87% 68% 90% CiƟes Drivetest CiƟes Walktest Towns Drivetest Roads Drivetest Train Walktest The quality and reliability of voice connections represent 40 per cent of the final score. Which operator offers the best network in this respect? Data tests account for 60 per cent of the final score.Who delivers the best performance in this category?Voice Data In some users‘ communications habits, voice telephony only plays a minor role. But conven­ tional phone calls are far from being outdated. Otherwise, the three German network opera­ tors would probably not have gone to the lengths of imple­ menting VoLTE – telephony based on sending data packets over the LTE network. Therefore, quality and perfor­ mance of voice telephony still played an important part in the drive tests and walk tests con­ ducted by P3: For this purpose, both cars that were driving through 17 large and many smaller German ­­cities carried six Samsung smartphones each.They permanently called their counterparts in the other vehicle. In order to simulate the everyday smartphone use, the phones would constantly trans­ fer data in the background ­during the ­telephony tests. An identical device configu­ra­ tion was used in the backpacks Data communication is the most prestigious category in connect‘s network test. Firstly, the results of these tests repre­ sent 60 per cent of the final score.And secondly, the test parcours to be completed by our candidates incorporates a large number of practice-orien­ ted applications. For instance, the smartphones frequently ­access the most popular web sites according to the renowned Alexa ranking, as well as the static „ETSI reference web ­page“ also known as „Kepler page“. Measuring the speed and relia­ bilty of data transfers is the aim of our upload and download tests. We monitor uploads with test files sized 3 MB and downloads with 1 MB files.Additionally, we verify which amount of data travels over the network within ten seconds. Another scope of our testing are Youtube videos.The popular ­video platform does not distin­ guish between standard defini­ tion (SD) and high definition (HD) carried by the test staff who walked around in city centres and public buildings conducting the walk tests. The devices had been confi­ gured to make sure that part of the connections would be esta­ blished via VoLTE and another part would be transmitted via conventional circuit-switched telephony. Distinct ranking order The tests in the city centres ­already showed a clear picture: Both in the drive tests and in the walk tests, Deutsche Tele­ kom is ahead.Vodafone follows at a distance of a few points. O2 is clearly defeated what can be seen in the table below by means of lower success ­ratios, longer call setup times and also a lower average speech quality. As a result of their test drives through smaller cities and on connecting roads, the P3 tes­ ting team found basically the same ranking order: Deutsche Telekom leads,Vodafone fol­ lows at a comparatively small distance, and O2 comes in last with a considerably larger gap. While Telefónicas network keeps up quite well in smaller towns, its distance to the lea­ ding two contenders grows larger on the connecting roads. On the whole, compared to last year‘s test, O2 improved in the voice category. For the voice calls ­examined by the testing staff in trains, even Telekom and ­Vodafone showed some weak- nesses. But O2 scores again far behind them. Regarding some measurement values like call setup times and speech quality,Vodafone is ­narrowly ahead. But while the Düsseldorf-based operator could claim a stage win in the voice category last year, in the 2017 test this title goes to Telekom. resolutions any more. It rather ­dynamically adapts the video resolution to the bandwidth that is currently available. In order to respect this new strategy in our tests, we examined the success ratio of video playbacks, the start times, the percentage of playouts that took place with­- out interruptions as well as the videos‘ average resolution or number of lines respectively. Both P3 test cars checked these indicators as part of their drivetest, and also the walk test teams had the same agenda. All data measurements in Ger­ many were executed with the LTE Cat.6 smartphone Samsung Galaxy Note 4. Strong Deutsche Telekom In big cities, the results were similar to the voice category: Again, a very strong Telekom takes the lead,Vodafone follows with good results, and O2 is clearly defeated. Separate ana­ lyses show that Vodafone >> OPERATOR Telekom Vodafone Telefónica DATA (CiƟes; Drivetest) Web-Page Download (Live/StaƟc) Success RaƟo (%/%) 99.4/99.7 98.8/99.2 94.9/96.3 Ø Session Time (s/s) 2.9/1.4 3.0/1.4 3.4/2.1 File Download (3 MB) Success RaƟo/Ø Session Time (%/s) 99.7/1.3 99.5/1.9 99.1/6.0 90%/10% faster than (kbit/s) 12884/50934 7219/52516 1784/27666 File Upload (1 MB) Success RaƟo/Ø Session Time (%/s) 99.7/1.3 99.2/1.4 96.7/2.7 90%/10% faster than (kbit/s) 5690/11586 3567/12864 1453/8719 File Download (10 Seconds) Success RaƟo (%) 99.8 99.2 98.3 Ø Throughput (kbit/s) 55738 39675 13591 90%/10% faster than (kbit/s) 16703/107234 7880/94372 1993/32908 File Upload (10 Seconds) Success RaƟo (%) 99.7 99.1 98.5 Ø Throughput (kbit/s) 27892 15841 8492 90%/10% faster than (kbit/s) 7885/42458 3735/29749 1382/17214 Youtube Videos Success RaƟo/Start Time (%/s) 99.8/1.7 99.8/1.8 98.1/2.1 Playouts without InterrupƟons (%) 100.0 99.9 99.3 Ø Video ResoluƟon (p) 578 614 457 DATA (CiƟes; Walktest) Web-Page Download (Live/StaƟc) Success RaƟo (%/%) 99.3/99.3 98.5/98.9 91.3/92.1 Ø Session Time (s/s) 2.9/1.4 3.1/1.6 3.4/2.2 File Download (3 MB) Success RaƟo/Ø Session Time (%/s) 99.6/1.4 99.1/1.9 95.5/6.3 90%/10% faster than (kbit/s) 13578/51513 6801/56497 1569/37915 File Upload (1 MB) Success RaƟo/Ø Session Time (%/s) 98.6/1.5 97.7/2.0 91.6/3.3 90%/10% faster than (kbit/s) 3127/11409 1951/13029 1026/9050 File Download (10 Seconds) Success RaƟo (%) 99.4 99.2 96.2 Ø Throughput (kbit/s) 57742 48961 17254 90%/10% faster than (kbit/s) 15848/111341 7252/108922 1471/47759 File Upload (10 Seconds) Success RaƟo (%) 99.2 98.9 96.5 Ø Throughput (kbit/s) 25097 16423 8773 90%/10% faster than (kbit/s) 3455/42305 2141/39359 654/28672 Youtube Videos Success RaƟo/Start Time (%/s) 99.4/1.7 99.0/1.9 98.2/2.3 Playouts without InterrupƟons (%) 100.0 100.0 99.0 Ø Video ResoluƟon (p) 572 609 465 DATA (Towns; Drivetest) Web-Page Download (Live/StaƟc) Success RaƟo (%/%) 99.0/99.6 98.2/98.5 94.3/95.5 Ø Session Time (s/s) 2.9/1.4 3.2/1.6 3.5/2.2 File Download (3 MB) Success RaƟo/Ø Session Time (%/s) 99.8/1.6 99.0/2.3 96.7/4.4 90%/10% faster than (kbit/s) 8710/48387 6378/36364 2908/29376 File Upload (1 MB) Success RaƟo/Ø Session Time (%/s) 99.2/1.7 97.7/1.8 94.4/3.6 90%/10% faster than (kbit/s) 3230/11227 2298/11594 1045/7779 File Download (10 Seconds) Success RaƟo (%) 99.8 99.4 96.2 Ø Throughput (kbit/s) 36873 20028 17021 90%/10% faster than (kbit/s) 10172/73053 6410/39497 3904/35924 File Upload (10 Seconds) Success RaƟo (%) 99.4 99.0 98.2 Ø Throughput (kbit/s) 20759 11329 6804 90%/10% faster than (kbit/s) 4144/41682 2670/19807 988/16864 Youtube Videos Success RaƟo/Start Time (%/s) 99.6/1.8 99.4/1.8 98.2/2.2 Playouts without InterrupƟons (%) 100.0 100.0 99.8 Ø Video ResoluƟon (p) 556 580 484 OPERATOR Telekom Vodafone Telefónica VOICE (CiƟes; Drivetest) Call Success RaƟo (%) 99.4 98.8 95.9 Call Setup Time (s) 4.0 3.9 5.5 Speech Quality (MOS-LQO) 3.8 3.9 3.7 VOICE (CiƟes; Walktest) Call Success RaƟo (%) 99.4 98.5 96.7 Call Setup Time (s) 4.0 4.0 5.3 Speech Quality (MOS-LQO) 3.9 3.9 3.7 VOICE (Towns; Drivetest) Call Success RaƟo (%) 99.5 98.7 97.2 Call Setup Time (s) 4.0 3.9 5.7 Speech Quality (MOS-LQO) 3.8 3.9 3.6 VOICE (Roads; Drivetest) Call Success RaƟo (%) 98.6 97.4 88.3 Call Setup Time (s) 4.3 4.2 6.4 Speech Quality (MOS-LQO) 3.8 3.8 3.3 VOICE (Train; Walktest) Call Success RaƟo (%) 84.2 83.1 76.5 Call Setup Time (s) 5.1 5.4 6.9 Speech Quality (MOS-LQO) 3.6 3.6 3.1 Germany 600 of 1000 Points400 of 1000 Points VOICE DATA of180 of270 of80 of120 of50 of75 of60 of90 92% 88% 49% Telekom Vodafone Telefónica 95% 91% 78% 95% 90% 76% 94% 92% 71% 89% 82% 62% of30 of45 57% 54% 33% 61% 46% 30% 87% 81% 65% 87% 84% 56% 87% 68% 90% CiƟes Drivetest CiƟes Walktest Towns Drivetest Roads Drivetest Train Walktest network test 54 1/2017
  • 3. Modern cars rely heavily on connectivity. How does this actually work out on German roads? Railways used to be the blind spot of German mobile network operators. Is this still true this year? Connecting Roads It was approximately 6600 kilo­ metres that P3‘s two test vehic­ les covered this year on German connecting roads – on top of the 5500 kilometres that each car covered driving through large and smaller cities.The point of this exercise: Gaining closer insights about the quality and reliability of the mobile networks on this parti­ cular type of roads. Distinct ranking order on the roads While Telekom and Vodafone ­were almost at level in the voice tests conducted on connecting roads, their offset becomes more obvious in the data cate­gory. ­Especially regarding the success ratios of web page access as well The testing staff spent about 33 hours on 15 different ICE trains during their railway tests con­ ducted in 2016.Yet, the measu­ rements did not only take place in these flagship trains of Deut­ sche Bahn, but also considered regional railway connections. The test results from German trains should not surprise any­ body who has read the outcome of the other categories: In the trains, Deutsche Telekom once again leads the pack,Vodafone scores second with viable re­ sults, and O2 brings up the rear. Especially when comparing the partial results obtained on the railways, looking over the borders – specifically at the re­ sults of the respective tests in Austria and Switzerland – may make German railway custo­ mers quite envious. Both alpine countries are considerably ahead in this respect.And even Telekom, which scores best in the measurements taken in German railways, cannot mea­ sure up with the results of the Austrian and Swiss operators – by far. Deutsche Bahn has as downloads and uploads,Tele­ kom clearly ranks first and keeps its competitor from Düsseldorf at a distance. Still, this match takes place at a very high level when looking at the distance of O2. Similar to the large and small ­cities before,Telefónica also loses valuable points on the connecting roads and thus falls back further behind the leading two German operators. The test results in this category are quite obvious: Car drivers who need robust data connec­ tions on the road – whether for navig­ation, for communication or mobile entertainment purposes – currently cannot pass the ­mobile networks of Deutsche ­Telekom or Vodafone. ­recently started an initiative to enhance connectivity especially in its ICE trains in close coope­ ration with the German mobile network operators. But this does not seem to have much ­impact on this year‘s mobile network test. Much need for improvement When looking at the details, there are many similarities bet­ ween the railways and the con­ necting roads:While Telekom and Vodafone were almost at level regarding voice phone calls in trains, their offset grows in the data category. Here, ­Telekom offers the best results – but still shows a lot of room for improvement.Voda­fone‘s results are mid-level, and O2 once again comes in last. ­Success ratios, like those for web surfing, of around 85 per cent at Telekom, about 77 per cent at Vodafone and approxi­ mately 63 per cent at O2 ­convey a clear message:When it comes to connectivity in Ger­ man railways, there remains a lot of work to be done. Telekom wins this year‘s ­network test at a distinct dis­ tance to runner-up Vodafone. Both in the voice and data ­categories, the Bonn-based operator turns out to be the strongest conveyor. Although we have raised our requirements, last year‘s win­ ner Telekom not only mana­ ged to defend its position, but actually continued to improve its score.Therefore Telekom absolutely deserves the first place in Germany – and this for the sixth time in a row. Compared to the results of last year, both the networks of O2 and E-Plus managed to improve. This is particularly underlined by the results of the voice measurements.A valid expla­ nation for O2 not scoring any better may be the problems caused by the ongoing inte­ gration of both networks. So we hope in the best ­interest of O2‘s customers that this integration will ­con­tinue to foster noticeable ­improvements. Vodafone also improved both in the voice and data cate­ gories compared to last year‘s results. But still Telekom out­ performed their Düsseldorf- based competitor in this year‘s voice tests.Anyway, the 2017 network test em­ phasizes Vodafone‘s clear ­improvements in the data scores. In this context, Vodafone‘s excellent Youtube results are particularly eye- catching.All in all, these ­results entail a second rank with a good overall score. OPERATOR Telekom Vodafone Telefónica DATA (Roads; Drivetest) Web-Page Download (Live/StaƟc) Success RaƟo (%/%) 99.0/99.3 95.9/96.3 90.1/91.5 Ø Session Time (s/s) 3.0/1.5 3.2/1.6 3.5/2.3 File Download (3 MB) Success RaƟo/Ø Session Time (%/s) 99.3/2.0 97.0/2.5 93.8/4.8 90%/10% faster than (kbit/s) 6409/43956 5908/41958 2359/29851 File Upload (1 MB) Success RaƟo/Ø Session Time (%/s) 98.6/1.9 96.8/1.8 90.8/3.8 90%/10% faster than (kbit/s) 2475/10344 2632/12214 959/8032 File Download (10 Seconds) Success RaƟo (%) 99.5 96.9 95.4 Ø Throughput (kbit/s) 30575 26531 15444 90%/10% faster than (kbit/s) 7496/59980 6780/55954 2801/36096 File Upload (10 Seconds) Success RaƟo (%) 99.2 96.3 92.5 Ø Throughput (kbit/s) 15882 12376 6963 90%/10% faster than (kbit/s) 3060/33803 2635/20692 937/17988 Youtube Videos Success RaƟo/Start Time (%/s) 99.6/1.8 97.4/1.9 92.7/2.2 Playouts without InterrupƟons (%) 100.0 100.0 99.9 Ø Video ResoluƟon (p) 543 593 492 OPERATOR Telekom Vodafone Telefónica DATA (Train; Walktest) Web-Page Download (Live/StaƟc) Success RaƟo (%/%) 84.5/85.5 76.7/78.9 62.7/61.7 Ø Session Time (s/s) 3.4/1.9 3.7/2.1 4.5/3.7 File Download (3 MB) Success RaƟo/Ø Session Time (%/s) 88.2/5.4 81.2/4.9 72.6/11.8 90%/10% faster than (kbit/s) 2440/29183 2491/36934 726/12771 File Upload (1 MB) Success RaƟo/Ø Session Time (%/s) 87.0/2.9 78.8/3.2 62.4/7.3 90%/10% faster than (kbit/s) 1151/8555 1058/10974 526/5915 File Download (10 Seconds) Success RaƟo (%) 87.9 81.2 72.6 Ø Throughput (kbit/s) 21106 16802 7094 90%/10% faster than (kbit/s) 3825/40116 2488/37275 911/13041 File Upload (10 Seconds) Success RaƟo (%) 87.1 81.5 73.4 Ø Throughput (kbit/s) 13661 9099 2894 90%/10% faster than (kbit/s) 999/29577 767/20322 216/8658 Youtube Videos Success RaƟo/Start Time (%/s) 88.1/2.1 78.4/2.4 89.3/3.1 Playouts without InterrupƟons (%) 100.0 99.2 98.4 Ø Video ResoluƟon (p) 499 522 402 Data on Railways Single review 887 841 643TOTAL very good good sufficient-RaƟng Shown voice, data and total scores are rounded. VOICE max. 400 max. 1000 Points DATA max. 600 366 352 270 521 373 489 Telekom Vodafone Telefónica could definitely improve over last year‘s results in the data ­category. To some extent, this supports the Düsseldorf operator counterbalancing its shortfall in the data score.Also,Vodafone achieves especially good results at You­tube playback – in this ­category, the Düsseldorfers ­partly draw level with Telekom. Still, this is not sufficient to grant Vodafone a partial victory.The reason is that Vodafone scores a little worse than Telekom in the ­discipline of web page access and – slightly less distinctive – regarding file uploads and ­downloads. However, when we look at O2, their deficits are even more ob­ vious.While the inner city walk tests show a success ratio of 99.3 per cent for web page access in the Telekom network and 98.5 percent at ­Vodafone, this value drops to 91.3 per cent in Tele­ fónica‘s network. Statistically, this means that ­almost one out of ten web page views will fail when O2 customers are walking through city centres. When observing indicators like the success ratios or average speeds of file downloads in the big city drive and walk tests, the results show the same trend that looks increasingly familiar this year:Telekom leads,Voda­fone follows at not too big a distance, and O2 clearly comes in last. For example, according to our walk tests, file downloads run at average data rates of more than 13578 kbps in 90 per cent of the cases in the Telekom network. In the Vodafone network, it is still more than 6801 kps, and O2 achieves only a minimum of 1568 kbps. So the latter candi­ date accomplishes not much ­more than a tenth of the speed offered by test winner Telekom. Drive tests in smaller towns The drive tests conducted in small towns gave equivalent re­ sults:Again,Telekom achieves the best measurement values, Vodafone follows at a distinct but not huge distance – and Telefónica once again comes in last. Similar to the inner city drive tests, the success ratios of web page access via the O2 network are worse than in the networks of the leading two providers, but still better than O2‘s walk test results. Our download and upload tests show comparable results. Once more,Vodafone turns out to be a Youtube star in smaller towns. Its top performance in this category is at the same level as test winner Telekom. Both opera­ tors seem to have a very perfor­ mant “peering“ to the content de­ livery network of Google‘s video platform. O2‘s weak overall scores can be explained to some extent with the ongoing integration of the former E-Plus network and the distortions coming along with it. While O2 at least improved in the voice category, the 2017 network test indicates a stagnation for this operator in the data category. network test 76 1/2017
  • 4. In comparison to the previous year, all Austrian operators im- proved – on an already very high level.Who wins the race in the alpine republic this time? Since connect‘s network test has been including the alpine countries,Austrian network operators used to have a neck- and-neck contest on the high­ est level. Compared to the other countries in the DACH region, the Austrian contenders regu­ larly are among the top tiers. No network operator from the alpine republic ever scored worse than the grade “good“. Austrian customers have every reason to be happy, as they can choose from three very good providers.And this at ­considerably lower costs than for example in Germany.The Austrian network operators also look pretty good when it comes to the roll-out of LTE. In autumn 2016,A1,T-Mobile Austria and Drei (the Austrian subsidiary of Hutchison Three) already offered 4G to a large part of the Austrian population and could focus on filling the few remaining gaps. So, we highly anticipated the results of the measurements that P3 took in eleven larger Austrian cities, on appro­xi­­ mately 2700 kilometres of ­connecting roads as well as in Austrian railways. Voice connections Assessing voice telephony,A1 scores first. Especially in the drive tests – the test calls made from car to car – that P3 took in larger cities, this operator per­ formed a tiny bit better than its competitors.At the time of tes­ ting,A1 was the only Austrian operator who already supported VoLTE.This may have helped particularly in the inner cities with their good LTE coverage. But the high score that A1 achieved in this category, would not have been possible had this operator not also performed ­exceptionally well in its conven­ tional telephony service. OPERATOR A1 Drei T-Mobile DATA (Roads; Drivetest) Web-Page Download (Live/StaƟc) Success RaƟo (%/%) 97.9/98.2 97.9/98.8 96.4/95.9 Ø Session Time (s/s) 2.5/1.3 2.4/1.3 2.8/1.5 File Download (3 MB) Success RaƟo/Ø Session Time (%/s) 98.6/1.5 97.7/1.3 97.9/2.5 90%/10% faster than (kbit/s) 11934/61856 11215/60914 5452/43059 File Upload (1 MB) Success RaƟo/Ø Session Time (%/s) 97.7/1.6 98.4/1.2 96.5/1.9 90%/10% faster than (kbit/s) 2863/14304 3294/15742 2131/17022 File Download (10 Seconds) Success RaƟo (%) 98.8 97.9 97.2 Ø Throughput (kbit/s) 49500 50032 28171 90%/10% faster than (kbit/s) 14417/87661 17012/93075 6238/58081 File Upload (10 Seconds) Success RaƟo (%) 98.8 98.6 98.4 Ø Throughput (kbit/s) 26980 31521 14835 90%/10% faster than (kbit/s) 2476/45416 5615/44914 2094/29461 Youtube Videos Success RaƟo/Start Time (%/s) 97.8/1.8 99.5/1.7 97.3/1.9 Playouts without InterrupƟons (%) 100.0 100.0 99.0 Ø Video ResoluƟon (p) 626 636 596 OPERATOR A1 Drei T-Mobile DATA (CiƟes; Drivetest) Web-Page Download (Live/StaƟc) Success RaƟo (%/%) 99.8/99.9 99.3/99.4 99.2/99.5 Ø Session Time (s/s) 2.3/1.1 2.4/1.2 2.6/1.3 File Download (3 MB) Success RaƟo/Ø Session Time (%/s) 99.7/0.9 99.7/1.0 99.7/1.4 90%/10% faster than (kbit/s) 20466/62647 17394/62016 11331/56738 File Upload (1 MB) Success RaƟo/Ø Session Time (%/s) 99.8/0.9 99.4/0.8 99.7/1.0 90%/10% faster than (kbit/s) 8583/16097 8667/16360 6552/20305 File Download (10 Seconds) Success RaƟo (%) 99.9 99.7 99.8 Ø Throughput (kbit/s) 62742 51652 45196 90%/10% faster than (kbit/s) 23730/115195 20937/89926 14096/84799 File Upload (10 Seconds) Success RaƟo (%) 99.9 99.8 99.3 Ø Throughput (kbit/s) 34237 36577 25496 90%/10% faster than (kbit/s) 14082/46060 20326/45098 7436/42033 Youtube Videos Success RaƟo/Start Time (%/s) 99.9/1.7 99.5/1.7 99.7/1.8 Playouts without InterrupƟons (%) 100.0 99.9 99.9 Ø Video ResoluƟon (p) 645 649 638 DATA (CiƟes; Walktest) Web-Page Download (Live/StaƟc) Success RaƟo (%/%) 99.4/100.0 99.5/100.0 98.6/99.2 Ø Session Time (s/s) 2.4/1.1 2.4/1.2 2.7/1.4 File Download (3 MB) Success RaƟo/Ø Session Time (%/s) 100.0/0.8 100.0/0.8 99.8/1.2 90%/10% faster than (kbit/s) 20430/58968 22067/58394 15464/53812 File Upload (1 MB) Success RaƟo/Ø Session Time (%/s) 100.0/0.9 100.0/0.9 98.4/1.2 90%/10% faster than (kbit/s) 9281/16247 8153/15696 5589/20752 File Download (10 Seconds) Success RaƟo (%) 100.0 100.0 99.2 Ø Throughput (kbit/s) 61643 55793 52971 90%/10% faster than (kbit/s) 25388/108707 23493/92364 20198/93411 File Upload (10 Seconds) Success RaƟo (%) 100.0 100.0 99.4 Ø Throughput (kbit/s) 35294 35516 27018 90%/10% faster than (kbit/s) 14731/46139 17435/45180 5546/43022 Youtube Videos Success RaƟo/Start Time (%/s) 99.6/1.7 100.0/1.7 99.1/1.7 Playouts without InterrupƟons (%) 100.0 100.0 100.0 Ø Video ResoluƟon (p) 652 666 652 DATA (Towns; Drivetest) Web-Page Download (Live/StaƟc) Success RaƟo (%/%) 99.7/99.3 99.1/99.7 99.2/99.7 Ø Session Time (s/s) 2.5/1.4 2.4/1.2 2.7/1.3 File Download (3 MB) Success RaƟo/Ø Session Time (%/s) 99.7/1.4 99.7/1.0 99.7/1.7 90%/10% faster than (kbit/s) 10838/53982 17583/61856 8366/46720 File Upload (1 MB) Success RaƟo/Ø Session Time (%/s) 98.4/1.7 100.0/0.9 99.7/1.2 90%/10% faster than (kbit/s) 2527/13865 7775/15311 5344/17676 File Download (10 Seconds) Success RaƟo (%) 99.5 99.7 99.7 Ø Throughput (kbit/s) 42224 52360 30313 90%/10% faster than (kbit/s) 14119/82550 20509/86300 9052/59538 File Upload (10 Seconds) Success RaƟo (%) 100.0 99.7 98.7 Ø Throughput (kbit/s) 24080 34217 20782 90%/10% faster than (kbit/s) 2902/44232 15027/45149 4504/30440 Youtube Videos Success RaƟo/Start Time (%/s) 100.0/1.8 99.1/1.7 99.4/1.8 Playouts without InterrupƟons (%) 100.0 100.0 99.7 Ø Video ResoluƟon (p) 613 657 618 Bludenz Imst Zirl Worgl Zell am See Sankt Johann im Pongau Bischofshofen Hallein Seewalchen am AƩersee Laarkirchen Kremsmunster Enns Melk Brunn am Gebirge Koƫngbrunn Gleisdorf Völkermarkt Velden am Wörthersee SpiƩal an der Drau Kufstein Linz Traun St. Pölten Baden Salzburg Feldkirch Villach Wien Graz Klagenfurt Innsbruck Drivetest Walktest Roads Train In the walk tests that our ­testing teams conducted in city centres and public buildings, all three candidates scored almost equally on a very high level. Last year‘s winner is slightly ahead in the discipline of voice connec­ tions in rural ­regions, namely in smaller towns and on connecting roads. But even there, the ­overall difference between the three networks is only minuscule. Data communication When we look at the tests of data connections, the results are very similar. Here again, A1 shows a slight advance in ­larger cities with Drei following closely. For web page access tests, T-Mobile Austria falls a fraction behind – but still offers top ­results that might be totally ­sufficient for a test win in some other countries.All in all, the ­success ratios, session times and data throughputs that can be seen in the adjacent ­tables are fantastic values. In the inner city walk tests, Drei becomes the frontrunner by a very thin margin. But here again, the two other providers follow at a distance of only a ­­ few points. As we could already observe in the voice tests, Drei scores slightly better than both of its competitors in small towns and on the connecting roads. But, as in the other disciplines before, the contest takes place at a very high level. >> Austria OPERATOR A1 Drei T-Mobile VOICE (CiƟes; Drivetest) Call Success RaƟo (%) 99.0 98.6 98.9 Call Setup Time (s) 3.5 4.6 5.7 Speech Quality (MOS-LQO) 3.9 3.8 3.8 VOICE (CiƟes; Walktest) Call Success RaƟo (%) 99.5 99.8 99.8 Call Setup Time (s) 3.4 4.5 5.6 Speech Quality (MOS-LQO) 4.0 3.9 3.8 VOICE (Towns; Drivetest) Call Success RaƟo (%) 97.9 99.1 98.8 Call Setup Time (s) 3.7 4.6 5.8 Speech Quality (MOS-LQO) 3.8 3.8 3.8 VOICE (Roads; Drivetest) Call Success RaƟo (%) 97.3 99.0 98.0 Call Setup Time (s) 3.9 4.6 6.8 Speech Quality (MOS-LQO) 3.7 3.8 3.7 VOICE (Train; Walktest) Call Success RaƟo (%) 93.8 92.1 91.5 Call Setup Time (s) 4.0 4.7 6.7 Speech Quality (MOS-LQO) 3.6 3.7 3.7 600 of1000 Points400 of 1000 Points VOICE DATA of180 of270 of80 of120 of50 of75 of60 of90 88% 93% 84% A1 Drei T-Mobile 87% 92% 87% 97% 96% 93% 94% 89% 88% 91% 92% 83% of30 of45 82% 77% 71% 79% 81% 77% 89% 93% 89% 95% 96% 90% 93% 91% 95% CiƟes Drivetest CiƟes Walktest Towns Drivetest Roads Drivetest Train Walktest 600 of1000 Points400 of 1000 Points VOICE DATA of180 of270 of80 of120 of50 of75 of60 of90 88% 93% 84% A1 Drei T-Mobile 87% 92% 87% 97% 96% 93% 94% 89% 88% 91% 92% 83% of30 of45 82% 77% 71% 79% 81% 77% 89% 93% 89% 95% 96% 90% 93% 91% 95% CiƟes Drivetest CiƟes Walktest Towns Drivetest Roads Drivetest Train Walktest 600 of1000 Points400 of 1000 Points VOICE DATA of180 of270 of80 of120 of50 of75 of60 of90 88% 93% 84% A1 Drei T-Mobile 87% 92% 87% 97% 96% 93% 94% 89% 88% 91% 92% 83% of30 of45 82% 77% 71% 79% 81% 77% 89% 93% 89% 95% 96% 90% 93% 91% 95% CiƟes Drivetest CiƟes Walktest Towns Drivetest Roads Drivetest Train Walktest network test 98 1/2017
  • 5. Single review In spite of its strong competitors, Swisscom ­habitually ranked first in Switzerland. However, this year things are a little different. In the connect network test, the Swiss operators have had sur­ prises in store time and again. Traditionally, the bar is set extre­ mely high in Switzerland – quite often all three network providers achieved the grade “very good“. Of course we have raised our requirements once again this year – and still the present test winner succeeded in climbing just a little over the 950 point mark within our 1000 point scoring scheme. Thus, for the first time in the net­ work test for Germany,Austria and Switzerland, we have to award the grade “outstanding“. This is short of a sensation and it goes along with a surprising change at the top. But the other two Helvetian candidates still are gra­ ded “very good“ – and their achie­ vements are far from being close calls. But one thing after another. Voice connections Very much the same as in the other countries, the testing staff of P3 examined the quality and stability of voice connections in Switzerland by conducting drive tests and walk tests. ­During ­these tests, Sunrise quickly ­turned out to be ahead in most of the examined scena­rios. This is quite clear for phone calls out of cars in larger ­cities. But in more rural areas (smaller towns as well as on the tested connec­ ting roads), the gap between Sunrise and Swisscom shrinks down to one or two points. In the results of our walk tests conducted in large Swiss cities, the competitors Sunrise and Swisscom are actually on par. Salt keeps some distance to the two leading contestants, but still achieves very good results. While Sunrise and Swisscom ­improved over last year‘s results, the third Swiss operator that was formerly known as “Orange“ ­more or less stays at the same level than last year. Up to now, Swisscom is the only operator in Switzerland who supports VoLTE.This modern voice standard that was applied for a part of the test calls, con­ tributes to Swisscom‘s excellent results in the voice category. How­ever, in the final scoring ­Sunrise stays still close ahead. Data communications The standings and tendencies that we could observe in the voice category repeat them­selves almost identically in the data ­­discipline. Sunrise accounts for the biggest improvement over last year‘s results in this cate­gory.This is quite obvious in the larger cities where Sunrise‘s lead over the also very strong Swisscom is a little more pro­ nounced in the drive tests than in the walk tests. For example, Sunrise achieves impressive success ­rations of 100 per cent for file downloads or Youtube playbacks in the cars. In contrast, in the smaller towns and on connecting roads, >> Zürich Luzern Uster Schaĭausen Thun Lugano Lausanne Yverdon-les-Bains Winthertur Bern Geneve Basel Chur Cham Baden Dubendorf Volketswil EffreƟkon DieƟkon Sarnen Bellinzona Spiez Bulle Vevey Renens Morges Gland Meyrin Carouge Worb Solothurn PraƩeln MuƩenz Drivetest Walktest Roads Train OPERATOR A1 Drei T-Mobile DATA (Train; Walktest) Web-Page Download (Live/StaƟc) Success RaƟo (%/%) 92.2/94.0 91.9/93.0 92.1/94.1 Ø Session Time (s/s) 2.6/1.7 2.5/1.3 2.8/1.7 File Download (3 MB) Success RaƟo/Ø Session Time (%/s) 97.4/2.4 92.2/1.3 95.2/2.8 90%/10% faster than (kbit/s) 6536/51337 13429/57362 4678/43353 File Upload (1 MB) Success RaƟo/Ø Session Time (%/s) 91.7/3.2 93.3/1.6 88.8/1.9 90%/10% faster than (kbit/s) 862/13106 2782/13785 1903/17779 File Download (10 Seconds) Success RaƟo (%) 95.4 92.4 96.2 Ø Throughput (kbit/s) 34828 45022 25450 90%/10% faster than (kbit/s) 6889/73222 16495/78822 4906/52455 File Upload (10 Seconds) Success RaƟo (%) 92.5 90.4 94.3 Ø Throughput (kbit/s) 17759 25002 14753 90%/10% faster than (kbit/s) 735/39427 3200/42636 1401/29780 Youtube Videos Success RaƟo/Start Time (%/s) 94.3/2.0 92.0/1.9 89.7/2.0 Playouts without InterrupƟons (%) 100.0 100.0 100.0 Ø Video ResoluƟon (p) 576 630 593 It was well worth the effort: In this year‘s connect network test,A1 takes back the gold medal from last year‘s winner Drei. Particu­ larly good voice results make A1 the overall winner in the Alpine Republic. But equally in terms of data communica­ tions and connectivity in rail­ ways, the A1 network scores very strong. Moreover, with a total score of 918 out of a possible maximum of 1000 A1 is also a top tier when comparing the results from all three involved countries. Compared ­­ to last year, T-Mobile Austria took the ­biggest step forward. The actual reason that this contender still scores razor- thin behind its competitors for most of the ­in­dicators is the extreme strength of all Austrian providers. Still, T-Mobile Austria turns out to be in excellent shape. Its total result would have made this provider a strong number two in Germany and would actually come quite close to the performance of the parent company based there. Drei was also ­able to improve on last year‘s results on the whole. But it is by a very close margin of only three points, that the Hutchi­ son-owned provider makes second place. Drei scores better than its competitors particularly in smaller towns and on connecting roads. ­Examining mobile connectivity in trains, the overall winner A1 and Drei are basically on par.And bear in mind that a second place in Austria would be equivalent to a test win in many other countries. Mobile communications on Austrian railways When they talk on the phone or surf the web in trains,Austrian customers have once more good reason to be pleased. The measurement values that P3‘s teams gathered on hund­ reds of railway kilometres, certify very good results for the three Austrian operators – even if ­indicators like success ratios or data rates somewhat drop in this category when for instance com­ pared to those obtained on the connecting roads. A1 scores best for conveying voice calls in trains, while Drei turns out to be the data cham­ pion. Regarding voice telephony in trains,T-Mobile falls slightly behind, while this is not true for data communications ­– where this operator offers absolutely no cause for complaint. On the whole, in the railway category Austria scores some­ what behind Switzerland, but is clearly ahead of the results from Germany. 918 915 876TOTAL very good very good very good-RaƟng Shown voice, data and total scores are rounded. VOICE max. 400 max. 1000 Points DATA max. 600 365 360 346 553 530555 A1 Drei T-Mobile Switzerland OPERATOR Sunrise Swisscom Salt VOICE (CiƟes; Drivetest) Call Success RaƟo (%) 99.8 99.1 99.2 Call Setup Time (s) 3.4 3.3 5.0 Speech Quality (MOS-LQO) 3.9 3.9 3.5 VOICE (CiƟes; Walktest) Call Success RaƟo (%) 99.9 99.7 99.1 Call Setup Time (s) 3.3 3.3 4.9 Speech Quality (MOS-LQO) 3.9 4.0 3.5 VOICE (Towns; Drivetest) Call Success RaƟo (%) 99.9 99.3 99.5 Call Setup Time (s) 3.6 3.3 5.0 Speech Quality (MOS-LQO) 3.8 3.9 3.4 VOICE (Roads; Drivetest) Call Success RaƟo (%) 99.3 98.5 96.2 Call Setup Time (s) 3.7 3.4 5.3 Speech Quality (MOS-LQO) 3.8 3.9 3.4 VOICE (Train; Walktest) Call Success RaƟo (%) 98.3 97.1 96.2 Call Setup Time (s) 4.0 3.6 5.2 Speech Quality (MOS-LQO) 3.8 3.8 3.4 network test 1110 1/2017
  • 6. OPERATOR Sunrise Swisscom Salt DATA (CiƟes; Drivetest) Web-Page Download (Live/StaƟc) Success RaƟo (%/%) 99.8/99.9 98.4/99.3 99.0/99.2 Ø Session Time (s/s) 2.5/1.3 2.4/1.2 2.9/1.6 File Download (3 MB) Success RaƟo/Ø Session Time (%/s) 100.0/1.2 99.8/1.0 99.7/2.0 90%/10% faster than (kbit/s) 13796/58451 16771/62827 6805/37891 File Upload (1 MB) Success RaƟo/Ø Session Time (%/s) 99.7/0.8 99.1/0.8 98.8/1.5 90%/10% faster than (kbit/s) 6819/26846 6375/27778 2821/13722 File Download (10 Seconds) Success RaƟo (%) 99.9 99.1 99.6 Ø Throughput (kbit/s) 56170 65199 38994 90%/10% faster than (kbit/s) 17097/98956 18817/123092 8781/77076 File Upload (10 Seconds) Success RaƟo (%) 99.7 99.1 99.7 Ø Throughput (kbit/s) 27109 26860 20218 90%/10% faster than (kbit/s) 8022/45763 7070/44770 4219/39494 Youtube Videos Success RaƟo/Start Time (%/s) 100.0/1.5 99.5/1.4 99.9/1.6 Playouts without InterrupƟons (%) 100.0 100.0 100.0 Ø Video ResoluƟon (p) 676 674 646 DATA (CiƟes; Walktest) Web-Page Download (Live/StaƟc) Success RaƟo (%/%) 99.5/99.7 97.9/99.1 98.8/98.8 Ø Session Time (s/s) 2.6/1.4 2.5/1.2 2.9/1.5 File Download (3 MB) Success RaƟo/Ø Session Time (%/s) 99.6/1.0 99.6/0.8 99.8/2.2 90%/10% faster than (kbit/s) 16891/60560 20599/70012 5593/38388 File Upload (1 MB) Success RaƟo/Ø Session Time (%/s) 99.6/1.0 99.2/0.7 99.0/1.3 90%/10% faster than (kbit/s) 4837/26144 7741/28070 3364/15009 File Download (10 Seconds) Success RaƟo (%) 99.8 98.8 99.8 Ø Throughput (kbit/s) 59246 72930 43371 90%/10% faster than (kbit/s) 17024/117345 23343/138987 8045/76761 File Upload (10 Seconds) Success RaƟo (%) 99.6 100.0 99.6 Ø Throughput (kbit/s) 27846 30210 24091 90%/10% faster than (kbit/s) 4983/46361 10290/45322 5097/45404 Youtube Videos Success RaƟo/Start Time (%/s) 100.0/1.6 99.6/1.4 99.4/1.6 Playouts without InterrupƟons (%) 100.0 100.0 100.0 Ø Video ResoluƟon (p) 677 681 648 DATA (Towns; Drivetest) Web-Page Download (Live/StaƟc) Success RaƟo (%/%) 99.5/99.3 98.9/99.5 99.3/99.4 Ø Session Time (s/s) 2.6/1.4 2.4/1.3 2.8/1.6 File Download (3 MB) Success RaƟo/Ø Session Time (%/s) 100.0/1.2 99.5/1.0 99.8/1.7 90%/10% faster than (kbit/s) 11404/56272 17905/63141 8696/38326 File Upload (1 MB) Success RaƟo/Ø Session Time (%/s) 99.5/1.1 99.8/0.9 99.0/1.6 90%/10% faster than (kbit/s) 4286/25438 5276/26499 2487/13647 File Download (10 Seconds) Success RaƟo (%) 100.0 99.8 99.5 Ø Throughput (kbit/s) 50685 57979 39136 90%/10% faster than (kbit/s) 13006/95800 16417/109851 8924/77090 File Upload (10 Seconds) Success RaƟo (%) 100.0 99.5 99.2 Ø Throughput (kbit/s) 22788 25217 19457 90%/10% faster than (kbit/s) 5822/44056 7049/42839 2704/39110 Youtube Videos Success RaƟo/Start Time (%/s) 100.0/1.6 99.7/1.4 99.5/1.6 Playouts without InterrupƟons (%) 100.0 100.0 100.0 Ø Video ResoluƟon (p) 675 670 646 OPERATOR Sunrise Swisscom Salt DATA (Roads; Drivetest) Web-Page Download (Live/StaƟc) Success RaƟo (%/%) 98.6/99.0 96.3/98.3 96.6/97.9 Ø Session Time (s/s) 2.6/1.4 2.5/1.3 2.8/1.7 File Download (3 MB) Success RaƟo/Ø Session Time (%/s) 99.2/1.7 99.0/1.1 98.4/2.1 90%/10% faster than (kbit/s) 8734/60333 13263/64971 8049/39177 File Upload (1 MB) Success RaƟo/Ø Session Time (%/s) 99.0/1.3 98.9/1.2 96.4/1.8 90%/10% faster than (kbit/s) 3119/21623 3439/26756 2109/13629 File Download (10 Seconds) Success RaƟo (%) 99.7 98.0 98.4 Ø Throughput (kbit/s) 46109 59593 47507 90%/10% faster than (kbit/s) 11612/100874 16575/112413 10961/88886 File Upload (10 Seconds) Success RaƟo (%) 98.7 99.0 96.2 Ø Throughput (kbit/s) 17638 23561 20622 90%/10% faster than (kbit/s) 4035/37289 4060/44437 2944/41313 Youtube Videos Success RaƟo/Start Time (%/s) 99.7/1.6 99.3/1.5 97.6/1.6 Playouts without InterrupƟons (%) 100.0 100.0 99.8 Ø Video ResoluƟon (p) 664 675 651 OPERATOR Sunrise Swisscom Salt DATA (Train; Walktest) Web-Page Download (Live/StaƟc) Success RaƟo (%/%) 97.2/97.5 96.7/98.4 96.4/98.6 Ø Session Time (s/s) 2.8/1.7 2.7/1.6 3.1/2.0 File Download (3 MB) Success RaƟo/Ø Session Time (%/s) 98.9/2.2 97.8/1.9 98.9/3.7 90%/10% faster than (kbit/s) 6959/44594 5299/53097 3173/31360 File Upload (1 MB) Success RaƟo/Ø Session Time (%/s) 97.1/1.5 97.8/1.0 98.2/1.6 90%/10% faster than (kbit/s) 2910/23022 5907/24406 2957/13769 File Download (10 Seconds) Success RaƟo (%) 97.1 97.8 99.6 Ø Throughput (kbit/s) 31914 40576 21238 90%/10% faster than (kbit/s) 7834/65680 11614/85242 3799/46908 File Upload (10 Seconds) Success RaƟo (%) 98.2 98.2 99.3 Ø Throughput (kbit/s) 20141 22821 19025 90%/10% faster than (kbit/s) 2518/39363 6755/36717 4275/35336 Youtube Videos Success RaƟo/Start Time (%/s) 98.8/1.8 100.0/1.8 97.2/1.9 Playouts without InterrupƟons (%) 100.0 100.0 99.2 Ø Video ResoluƟon (p) 645 662 604 Single review Both in the voice and data measure­ ments, Sunrise achieves ex­ cellent results. Compared to the previous year, the opera­ tor improved in both catego­ ries. Due to a distinct gain in points in data communica­ tions, Sunrise brings in the overall win. Interestingly, the gap to last year‘s winner is quite wide – although Swiss­ com achieved very good results this time as well. For the first time in the connect mobile net­ work test for Germany,Austria and Switzerland we have to award the grade “outstanding“. The smal­ lest Helve­ tic mobile network provider achieves stable – and very good – overall results, both in the voice and the data ­categories. As they basically ­remain at the same level than in the ­previous year, Salt comes in third in the overall Swiss ranking. But when we compare the results of all three countries, Salt is still a top tier. More­ over, Salt attacks both of its competitors quite successfully with its aggressive pricing. The test results clearly show that last year‘s winner Swisscom has impro­ ved as well during the last twelve months. However, in the final sco­ ring Swisscom was beaten by the second largest contender Sunrise. Nevertheless, Swiss­com‘s customers can rest assured because with its excellent test results this Swiss ope­ rator would still immediately lead the field in Germany or Austria. 951 933 878TOTAL outstanding very good very good-RaƟng Shown voice, data and total scores are rounded. VOICE max. 400 max. 1000 Points DATA max. 600 390 380 352 561 526 553 Sunrise Swisscom Salt 600 of 1000 Points400 of 1000 Points VOICE DATA of180 of270 of80 of120 of50 of75 of60 of90 96% 94% 79% Sunrise Swisscom Salt 98% 96% 91% 99% 98% 90% 98% 95% 90% 94% 92% 87% of30 of45 94% 92% 84% 90% 93% 87% 93% 93% 88% 93% 93% 88% 92% 88% 94% CiƟes Drivetest CiƟes Walktest Towns Drivetest Roads Drivetest Train Walktest 600 of 1000 Points400 of 1000 Points VOICE DATA of180 of270 of80 of120 of50 of75 of60 of90 96% 94% 79% Sunrise Swisscom Salt 98% 96% 91% 99% 98% 90% 98% 95% 90% 94% 92% 87% of30 of45 94% 92% 84% 90% 93% 87% 93% 93% 88% 93% 93% 88% 92% 88% 94% CiƟes Drivetest CiƟes Walktest Towns Drivetest Roads Drivetest Train Walktest the leading duo is almost on par – once again on a very high level. In the data measurements, Salt has to settle for the third rank once more – but this again only means that Sunrise and Swisscom were able to improve over last year‘s results, while Salt kept its still strong performance stable. In this context, we should bear in mind that Salt attacks its two strong competitors with a particularly aggressive pricing. This makes Salt‘s offerings all the more interesting.Above all, Salt‘s results are definitely top notch when compared to those from other countries. Mobile connectivity in Swiss railways The tendencies observed in the voice and data measure­ ments conducted in larger cities and smaller towns as well as on the connecting roads, prevail for the tests of phone calls and data connectivity in Swiss railways too. Again, in this category Swiss customers have every reason to be happy.Their operators achieve the best results within the three countries at a distinct ­distance. Indeed all three Swiss opera­ tors achieve a remarkably high performance and reliability in the challenging task of providing connectivity to moving trains in the demanding Helvetic topology. When looking at the detailed results in railway tests, Sunrise is slightly ahead in the voice ca­ tegory, while Swisscom scores a tad higher in the data measure­ ments. Salt again comes in third regarding both voice and data communications while still show­ ing very good results. 600 of 1000 Points400 of 1000 Points VOICE DATA of180 of270 of80 of120 of50 of75 of60 of90 96% 94% 79% Sunrise Swisscom Salt 98% 96% 91% 99% 98% 90% 98% 95% 90% 94% 92% 87% of30 of45 94% 92% 84% 90% 93% 87% 93% 93% 88% 93% 93% 88% 92% 88% 94% CiƟes Drivetest CiƟes Walktest Towns Drivetest Roads Drivetest Train Walktest 600 of 1000 Points400 of 1000 Points VOICE DATA of180 of270 of80 of120 of50 of75 of60 of90 96% 94% 79% Sunrise Swisscom Salt 98% 96% 91% 99% 98% 90% 98% 95% 90% 94% 92% 87% of30 of45 94% 92% 84% 90% 93% 87% 93% 93% 88% 93% 93% 88% 92% 88% 94% CiƟes Drivetest CiƟes Walktest Towns Drivetest Roads Drivetest Train Walktest network test 1312 1/2017
  • 7. As in previous years, connect’s partner for the network measure­ ments, P3 communications, used two vehicles to test drive the cho­ sen cities, towns and roads. In Germany and Austria each car carried six Samsung Galaxy S5 smartphones to measure voice services and three Samsung Ga­ laxy Note 4 performing the data service tests. In order to reflect the advanced roll-out of LTE with “3 Carrier Aggregation“ (the combination of three carrier fre­ quencies) in Switzerland, we used three Samsung Galaxy S7 for the data measurements there. The same setup of devices was utilized in the walk tests. For this effort, the smartphones were ­installed in trolleys and back­ packs with additional batteries. The devices’ firmware was each operator’s current firmware version. If such software was not available the most current firm­ ware from Samsung was used. Voice telephony Voice services were measured with the smartphones performing calls alternating between the two measurement cars (“mobile-to- mobile“).An additional car served as a mobile remote station for the calls of the walk test teams. Background data traffic was transmitted by one of the smart­ phones simultaneously to each call to reflect a realistic usage scenario.Audio quality was as­ sessed by using POLQA (Percep­ tual Objective Listening Quality Assessment) wide band scoring. All devices were configured in “LTE preferred” mode.Thus in the three German Networks as well as with A1 in Austria and Swisscom in Switzerland, the modern Voice-over-LTE (VoLTE) service could be used.Within networks not yet supporting ­VoLTE, the smartphones were forced to switch to 3G or 2G technology, the so-called circuit- switched-fall-back (CSFB). Data connectivity To assess cellular data perfor­ mance a sequence of tests were executed.As a dynamic web- browsing test, each country’s top web sites (according to the Alexa ranking) were downloaded in the so-called live web-browsing test. Additionally a static web site was tested, the industry standard ETSI (European Telecommunications Standards Institute) “Kepler“ ­reference page. HTTP downloads and uploads were performed with 3 MB and 1 MB files, simulating small file transfers.The networks’ peak performance was tested with a ten second download and upload of a single, very large file. The Youtube measurements considered the new “adaptive ­resolution“ feature of this video platform. In order to offer a per­ sistent video experience,Youtube adapts the video streams‘ reso­ lution dynamically to the band­ width that is currently available. Our scoring therefore considers the success ratio, the time until the playback starts, the percen­ tage of video playouts that take place without interruptions as well as the videos‘ average ­resolution or line number count respectively. Indoor and train measurements The walk tests consisted of the same tasks as were performed in the cars. For this effort two teams measured in public transport and in public places, like coffee shops, museums, train stations and airport terminals.Travelling from city to city by public trans­ port allowed the assessment of cellular network quality within the long distance trains. Logistics The tests were performed in ­Austria, Germany and Switzer­ land around the same period of time (Germany: October 21 – ­November 12; Austria: October 7 – 27; Switzerland: October 14 – November 1).All measure­ ments were done between 8 AM and 10 PM. Both cars were ­­always in the same cities, but on different routes to avoid any in­ terference of one car’s measure­ ment by the other car’s. Both ­vehicles followed a given route, including fixed location measure­ ments at “areas of interest” such as well-visited public places. Measurements there lasted one hour. Locations such as train ­stations, airports, much-frequen­ ted public parks or high-density urban areas typically demonstrate how networks respond when a high number of users compete for their share of bandwidth within the network’s available ­radio frequencies. The measurements included 17 larger cities and 26 smaller towns in Germany, while the walk tests frequented six cities. In Austria the drive tests covered 11 big cities and 20 smaller towns, the walk test team visited five ­cities. In Switzerland, the test route included 13 big cities and 20 smaller towns with the walk tests conducted in four cities. Travel between the cities mainly used highways, but smaller state and county roads were driven as well. For each connect test P3 communications follows a well- defined process to generate four independent and representative city and route plans.The connect editors choose randomly one of these four alternatives. Test efforts and results Overall 25,000 km were driven for the connect P3 mobile net­ work test in 2016. In Germany the approximately 12,100 km of driven routes alongside the cities and areas visited represent 13.4 million inhabitants, equaling METHODOLOGY Professional and critical: Bernd Theiss, head of test and technology at connect (on the left), and Hakan Ekmen, managing director of P3 communications (on the right). Four Samsung Galaxy Note 4 ­­measured the data performance in German and Austrian networks. Here are the reasons why we tested and evaluated the merging networks of E-Plus and O2 as a single O2 network. After Telefónica/O2 bought out its former competitor E-Plus in October 2014, the merger of both networks goes at full speed. Previous E-Plus customers are being transferred to O2 tariffs, and Telefónica must sell off some base stations that have been occupied by both operators according to the German ­regulatory authority.The re­ maining base station sites are already designated “O2“. Cells formerly belonging to E-Plus are no longer visible as a discrete mobile network. Instead, at the moment there are “old“ O2/E-Plus cells along with “new“ ones. Given this situation, con­ nect and P3 decided to only examine O2 in their network test that we con­ducted in late 2016. As we know from our ­readers and from our own experiences, difficulties ­definitely ­occur in the course of the network merge. These problems that include failing handovers between two differently ­configured network cells, are clearly ­recognized in the results of this year‘s P3 connect mobile network test. o2 And E-plus A merger with some obstacles: Combining two separate mo­ bile networks into one is far from being routine work. U­navoidable problems that are resulting from this endeavour are clearly recognized by our test results. around 16.7 per cent of Germany’s population. ­Austria was measured by driving 5,900 km covering about 3 million inhabitants (approx. 36 per cent of the Austrian population). In Switzerland, the test teams drove approx. 7,000 km, covering 1.9 ­million people representing around 22.5% of the Swiss population. Certainly a huge effort, but necessary to gain the required statistical rele­ vance and confidence in the test results. Scoring The results of the voice test contribute 40 per cent of the total score, those of the data tests make up 60 per cent. For the overall result we apply a 1000 point scheme in order to represent sufficiently detailed results. Moreover this scheme allows us to better compare the results of network tests that we have conducted in different countries (find all results and additional information at www.connect- testmagazine.com). network test 14 1/2017
  • 8. fairness AND transparency This year, some of the candidates massively tried to influence the conditions and parameters in the run-up of our test. The connect and P3 staff responsible for the testing project have of course fended off these attempts. As in previous years, connect and P3 met in early 2016 in order to define the conditions and parameters for this year‘s network test. In this preceding test design phase, we for example identify new test criteria, discard or confirm old ones and determine their influ­ ence on the overall score.We define the timeframe as well as a preselection of smartphone models that we intend to use for the measurements.We then communi­ cate these preliminary ­definitions in advance to the CTOs of the network operators. Feedback is appreciated In this process we appreciate feedback about aspects like suitable tariffs that facilitate un­ obstructed measurements of the best perfor­ mance possible.After all, our objective is to evaluate the network experience of the most demanding customers.We also agree on the firmware versions used in the measurement smartphones, as each mobile network operator makes ­adjustments to most popular devices to ­ensure a smooth interplay with their network. But this time, some contenders apparently took part in the discussions with the single intent to enforce measurement conditions that would favour their own network. For ­example, there have been attempts to im­ pose a smartphone model on us that all in all works less reliably than others – presumably because the involved provider expected an ­advantage for its own network from this. One operator insinuated flaws in the test design more than once – until extensive measurements conducted both by P3 com­ munications as well as by the connect test lab disproved all of them. Permanent chan­ ges in the reasonings of some operators led connect to the assump­tion that one or the other of them would not have minded blowing the rapidly ­approaching deadline of our test. The more danger, the more honour We cannot help but understand such ­attempts as a compliment for the high rele­ vance that the operators assign to our test. And of course we remain true to ourselves concerning these issues.After all, it is our standard to conduct a test that provides deep insights into the quality and perfor­ mance of the examined mobile networks. However, we will draw one conclusion from this year‘s experience: In the future, we will publish obvious attempts to abuse our trans­ parent approach to testing the very same way as we document our test procedures. Looking back at the results of the connect network tests since 2010 provides especially one ­insight: Despite of the constantly rising requirements, the level of the overall results has steadily ­improved. We reckon that our demanding and well renowned network test is not entirely blameless. Historical Development Customers‘ expectations are constantly growing – expan­ ding data volumes and rising transmission speeds are re­ garded to be absolutely normal. P3 and connect take account of this development by con­ stantly raising the requirements and thresholds of our tests. Network test as a driving force The adjoining glance at the development of results in ­Germany,Austria and Switzer­ land in recent years shows a clear overall tendency: ­Despite the growing require­ ments, all tested networks improved steadily. In all modesty, we believe that the high relevance and ­challenging demands of our annual network tests are an important driving force of this development. conclusion Hannes Rügheimer, connect author The operators enthusiastically fight for the top rank in the connect net- work test. The fact that almost all candidates managed to improve in spite of the rising requirements is clearly supporting our claim that our critical tests contribute to the overall enhancement of the mobile net- works’ quality. Against this background, the ­repeated test victory of Deutsche Telekom in Germany was by no ­means self-evident. It rather reflects the considerable efforts that Tele- kom takes in order to maintain and extend its network. Vodafone also worked flat out, but remains on the second rank. O2’s result shows some room for development but can be explained by the ongoing integra- tion with the former E-Plus network. In the alpine countries,there were rigorous fights as well. This led to a change at the top ranks in both countries. In Austria,A1 managed to gain back the crown from last year’s winner Drei. Particularly its very strong voice results secured the win to A1. Although, also Drei noticeably improved over last year, the Hutchi- son-owned company fell back be- hind A1 at a very narrow gap. Com- pared to the previous year, T-Mobile Austria made the biggest step for- ward,but still was not able to pass its two extremely strong competitors. In Switzerland we also see some movement at the top.Especially with its distinct rise in points in the data category, Sun­rise manages to out- play last year’s winner Swisscom. Thus,for the first time a candidate of theconnectnetworktestin­Germany, Austria and Switzerland is awarded the grade “outstanding“. And even though Swisscom only ranks second this ­time, the company has still im- proved considerably compared to the pre­vious year.Salt takes the third rank of the Helvetic network providers but still achieved very good results. The top quality of mobile connecti- vity on Swiss and also on Austrian railways was especially noteworthy. This is something that German rail- way customers can only dream about. GERMANY AUSTRIA SWITZERLAND Overall Results Voice and Data Telekom Vodafone Telefónica A1 Drei T-Mobile Sunrise Swisscom Salt VOICE max. 400 Points 366 352 270 365 360 346 390 380 352 CiƟes Drivetest 180 94% 92% 71% 94% 89% 88% 98% 95% 90% CiƟes Walktest 60 95% 90% 76% 97% 96% 93% 99% 98% 90% Towns Drivetest 80 95% 91% 78% 87% 92% 87% 98% 96% 91% Roads Drivetest 50 92% 88% 49% 88% 93% 84% 96% 94% 79% Train Walktest 30 57% 54% 33% 82% 77% 71% 94% 92% 84% DATEN max. 600 Points 521 489 373 553 555 530 561 553 526 CiƟes Drivetest 270 90% 87% 68% 95% 93% 91% 94% 92% 88% CiƟes Walktest 90 87% 84% 56% 95% 96% 90% 93% 93% 88% Towns Drivetest 120 87% 81% 65% 89% 93% 89% 93% 93% 88% Roads Drivetest 75 89% 82% 62% 91% 92% 83% 94% 92% 87% Train Walktest 45 61% 46% 30% 79% 81% 77% 90% 93% 87% Total max. 1000 Points 887 841 643 918 915 876 951 933 878 -RATING very good good sufficient very good very good very good outstanding very good very good Sunrise Swisscom Salt 40% 60% 80% 100% 2010 2011 2012 2013 2014 2015 2016 Vodafone Telefónica (O2) Telefónica (E-Plus) Telekom 40% 60% 80% 100% 2010 2011 2012 2013 2014 2015 2016 Drei T-Mobile A1 40% 60% 80% 100% 2010 2011 2012 2013 2014 2015 2016 887 841 643TOTAL very good good sufficient-RaƟng Shown voice, data and total scores are rounded. VOICE max. 400 max. 1000 Points DATA max. 600 366 352 270 521 373 489 Telekom Vodafone Telefónica 918 915 876TOTAL very good very good very good-RaƟng Shown voice, data and total scores are rounded. VOICE max. 400 max. 1000 Points DATA max. 600 365 360 346 553 530555 A1 Drei T-Mobile 951 933 878TOTAL outstanding very good very good-RaƟng Shown voice, data and total scores are rounded. VOICE max. 400 max. 1000 Points DATA max. 600 390 380 352 561 526 553 Sunrise Swisscom Salt TESTSIEGER TESTSIEGER TESTSIEGERBEST IN TEST BEST IN TEST BEST IN TEST network test 1716 1/2017