Samsung caught CHEATING in Galaxy S4 GPU benchmarks

nikufellow

In the zone
Samsung I9500 Galaxy S4 ,
the Exynos 5 Octa versions
of the flagship, didn’t make
it to most of the markets
and remained the more
exotic option. Production
hiccups and delays plus
the lack of LTE radio in the
initial batch was the reason
Samsung switched to
Snapdragon 600 chipset
and make the I9505 its
worldwide flagship.
Anyway, the octa-core
version followed shortly
after the Snapdragon 600
was launched. We already
had a meeting with the
beast and found it slightly
faster in benchmarks than
its international sibling, but
weaker when it comes to
battery life. A new
development however,
suggests that the scores
posted by the Exynos GPU
are obtained by cheating
and the I9500 real life
performance is lower than
its scores indicate.
It’s a really simple yet
powerful trick – the GPU
clock get boosted when you
run specific benchmarks on
your octa-core I9500
Galaxy S4 and this way
you get about 10% higher
result than you should.
Samsung indeed didn’t
specify the PoweVR SGX
544MP3 clock, but as per
Anandtech’s investigation it
runs always at up to
480MHz unless you are
using benchmarks. When
you do open a benchmark
app the Galaxy S4 GPU
start ticking at 532MHz
and continues to do so
until you exit the app.This
means the Exynos-powered
Galaxy S4 only brings its
A-game to benchmarks and
does slightly worse
elsewhere. It’s not like the
smartphone is a slouch
with regular apps, but
that’s blatant cheating.
We are not quite sure what
is the reason for the lower
clock cap in non-
benchmark apps, but risks
of overheating or huge
battery drainage are the
main suspect.
The CPU of the Galaxy S4
variants was also found to
behave somewhat odd with
its clock speed forced to
the maximum possible the
minute you run a
benchmark. That’s not
technically cheating
though, as the power is
available if needed
elsewhere – there’s no
overclocking.
The cheat doesn’t apply on
all benchmarks, but
Anandtechso far discovered
the different behavior on
AnTuTu, Quadrant, GL
Benchmark 2.5,
BenchmarkPi and Linpack.
So, it seems Samsung is
trying to polish the octa-
core Galaxy S4 reputation.
Whatever the reason. we
are hoping it doesn’t
become a habbit – the
company’s flagships have
enough power and they
don’t need to use such
unfair practices to gain an
advantage.

Source :gsmarena blog
 

quagmire

Allllright !
True words: "Benchmarking has become benchmarketing."

BTW Intel was also caught "manipulating" the benchmarks of Clovertrail+ SOC phones.
Read this : AnTuTu and Intel - AnandTech Forums
 

amjath

Human Spambot
Relax people it was
*cdn.gsmarena.com/pics/13/07/octa-core-sgs4-benchmarks/gsmarena_002.jpg
by Samsung by is official report
Under ordinary conditions, the Galaxy S4 has been designed to allow a maximum GPU frequency of 533MHz. However, the maximum GPU frequency is lowered to 480MHz for certain gaming apps that may cause an overload, when they are used for a prolonged period of time in full-screen mode. Meanwhile, a maximum GPU frequency of 533MHz is applicable for running apps that are usually used in full-screen mode, such as the S Browser, Gallery, Camera, Video Player, and certain benchmarking apps, which also demand substantial performance.

The maximum GPU frequencies for the Galaxy S4 have been varied to provide optimal user experience for our customers, and were not intended to improve certain benchmark results.

We remain committed to providing our customers with the best possible user experience.

ì￾¼ë¶€ 벤치마í￾¬ 사ì￾´íŠ¸ì—￾ì„œ 제기한 갤럭시 S4 테스트ì—￾ 대해 ë§￾씀드립니다. [ì￾´ìŠˆì™€ 팩트] :: SAMSUNG TOMORROW
 
Top Bottom