Samsung denies cheating on benchmark tests

The evidence seems pretty hard to argue, but Samsung wants to try anyway. Multiple sources have found that Samsung (as well as most other Android manufacturers) artificially boost CPU frequencies when the system detects a benchmarking test running. Samsung is denying that its devices are trying to cheat the tests, and is actually putting the blame on the tests themselves...
This is a discussion for a news article. To read the whole news, click here

102 Comments

65. Shatter

Posts: 2036; Member since: May 29, 2013

If Samsung uses it/makes an app its automatically bloatware. If apple or their OEM made the same app it would be amazing software that they always bragged about.

70. ngo2dd

Posts: 896; Member since: Jul 08, 2011

Expect it has been show time and time again that in the S4 that if disabled the Samsung app the phone run faster. So the phone is bloated with app that make it run slower. So those app turn into bloatware. Man you are one if the biggest shill here.

88. Pancholo

Posts: 380; Member since: Feb 27, 2012

Let me add quickly that, technically, iOS 7 is also bloatware for the moment being, as it makes your device lag. Commence the hatin'.

67. procopiojose

Posts: 132; Member since: Oct 26, 2012

the benchmark tests doesn't really trigger the cpu to "run forrest run" mode. remember that new generation socs implements power-saving features that will consistently give low scores cause it doesn't require all that power to do the tasks.

68. _Bone_

Posts: 2155; Member since: Oct 29, 2012

Where even AnandTech is missing the point: Samsung devices do absolutely NOT fake any these results, every benchmark is GENUINE and done by the HW, only these short burst type of apps get an extra kick at the silicon maximum, which is what benchmarking is all about, see it through the end. I guess when only these apps operate at "100%" and others at "90%", you could feel mad, but the other way to look at it is far more productive: there are thermal and battery benefits at 90%, and the Note 3 is an battery monsterhouse. The greatest thing Asus ever did was to dowclock the Snapdragon 600 on the Nexus 7, the thing is also battery and buttery smooth. Now is it cheating that with some ROM work you can get it pulling 1.9GHz performance? No, it's just what the HW maximum is, yet a lower clock is a better balance. It's not cheating if nothing is faked..

91. Jobayer

Posts: 167; Member since: Feb 22, 2013

yea r8, whatever helps u sleep at night man >_

96. _Bone_

Posts: 2155; Member since: Oct 29, 2012

idiot

84. nothingmuch

Posts: 201; Member since: May 03, 2013

""FlexRender is designed to help the Adreno GPU boost speed and save power by dynamically switching between drawing pixels in direct or deferred rendering mode"" http://www.qualcomm.com/snapdragon/processors/800 ^^^this is just one of the comments listed on this site, isn't the GPU just one of the things that these benchmarks test??? just think of what else the processor can help itself boost! now think on this for a minute if Samsung claims that they didn't cheat maybe they didn't, just maybe that the "white list" might be on the processor itself..... now wouldn't that be some sh**! Again which major brands are being accused of the same thing? Hmmmm....with the same chip? kinda obvious here that the source of the "boosting performance" might be in the processor itself, not the one who incorporates it.

87. yudi.nemesis unregistered

So now you blaming the chipset?? Maybe you can explain to me why? Nexus and Motorola devices with the same Qualcomm chipsets not have “Boosting Performance” in android benchmark.

100. Trex95

Posts: 2364; Member since: Mar 03, 2013

They will keep lies till they will believe there lies lol.

101. Ronny666

Posts: 128; Member since: Jun 08, 2013

We are unable to detect this in iphones mainly because it is running on closed operating system unlike androids. Most OEMs optimise their devices for benchmarking apps so that it can display the full potential of their hardware. That's not called cheating unless if they overclocked it higher than the advertised speed mainly for benchmarking tools. In Samsung's case, the GPU does hit at the highest frequency when running certain apps not just benchmark apps but most of the time it runs on lower frequency so that it doesn't generate too much of heat and suffer shorter life span. The normal GPU speed is adequate for running most applications even graphic extensive games. The basic thing that Anandtech missed out is, benchmark results does not directly translate to real-world usage. If you are on dope then u r cheating but if you are wearing feather light spikes in a competition to perform better than your usual training, then it is not called cheating.

102. remixfa

Posts: 14605; Member since: Dec 19, 2008

Its been a while since I've posted. So I figured this would be a good thread to visit. I just wanted to point out that I called Samsung and a few others out on cheating 2 YEARS AGO. Way to go, finally catching up! lol.
This copy is for your personal, non-commercial use only. You can order presentation-ready copies for distribution to your colleagues, clients or customers at https://www.parsintl.com/phonearena or use the Reprints & Permissions tool that appears at the bottom of each web page. Visit https://www.parsintl.com/ for samples and additional information.