Samsung Galaxy Note 3 apparently using "benchmark booster" to fix results

Samsung Galaxy Note 3 apparently using
We've seen benchmark results from the Samsung Galaxy Note 3 before, and they have always been impressive, much more impressive than results from other devices running the same exact chip, which is a red flag. Samsung has been caught "adjusting" benchmark results in the past, and it looks like the company is doing it again with a "benchmark booster" in the Note 3.

In the past, Samsung was caught overclocking the Galaxy S4 GPU in order to manipulate benchmark scores and make them look better than the competition, and Ars Technica has found evidence that the Note 3 is also boosting benchmark scores, but with a different method this time. Apparently, this time around when the Galaxy Note 3 sees a popular benchmarking app running (like Geekbench) it will invoke the "benchmark booster" which will make it so the multiple cores of the Snapdragon 800 cannot idle at all. 

The proof of this comes when running Stealthbench, which is exactly the same app as Geekbench just renamed to fool Samsung. When running Geekbench, the processor cores will not be allowed to idle, but with Stealthbench running, they can (as you can see in the first image below). Ars found a package called "DVFSHelper.java" which listed a number of benchmarking apps that will invoke the booster. It's unclear what other modifications, if any, are being done, but the results for single-core tests are the same, but multi-core Geekbench benchmarks could be as much as 20% higher because of Samsung's tweaks and Linpack could see results 50% higher than normal. 

Basically, it looks like more evidence that we should put less stock in benchmark numbers and more in real-world performance. 


source: Ars Technica

Related phones

Galaxy Note 3
  • Display 5.7" 1080 x 1920 pixels
  • Camera 13 MP / 2 MP front
  • Processor Qualcomm Snapdragon 800, Quad-core, 2300 MHz
  • Storage 64 GB + microSDHC
  • Battery 3200 mAh(21h 3G talk time)

FEATURED VIDEO

86 Comments

1. stealthd unregistered

If you can't beat em, cheat em. You just got Samsunged!

22. doejon

Posts: 411; Member since: Jul 31, 2012

but still note 3 is better than the other ones.. i don´t care if a app opens 0,0003 milisec faster xD

66. amats69

Posts: 1527; Member since: Nov 12, 2012

"But still note 3 is better than the other ones." but still samsung is CHEATING..

68. Wiki_jaan

Posts: 704; Member since: Jun 24, 2012

every one cheating not just samsung....

79. Long1

Posts: 399; Member since: May 18, 2013

"but still samsung is CHEATING.." No. They were CHEATING the Note 3, It has auto optimization to recognise when more power is needed and bring it on(which is great thing), and they rename some property to fool the device.

82. amats69

Posts: 1527; Member since: Nov 12, 2012

I congratulate you coz your fanboyism is at the highest level.you look like a member of a cult that follows whatever his/her master told him/her even if he/she knows it is already wrong..dude believe me, theres still life outside the shadow of samsung.....pathetic fanboy....

70. PunyPoop

Posts: 752; Member since: Jan 18, 2013

The Might Knight of Samsung! Worst than apple fanboys...

78. debugger919 unregistered

because you are a samsung fan boy, so what ever it does is good.

84. jdoee100

Posts: 334; Member since: Jun 04, 2013

"AnandTech just published a report that shows ASUS, HTC and other companies using the same 'benchmark detect' function to artificially bump up the numbers."(engadget) It's funny how it's always only Samsung that gets blamed for industry practices, when plenty of other manufacturers do the exact same thing.

37. phil2n

Posts: 519; Member since: Apr 30, 2012

I thought Sammy is only Superb in marketing, I was blind that they're also good on cheating and manipulating. I'm going to sue them for misleading costumers about benchmark test results. Tim Cook

67. Trex95

Posts: 2320; Member since: Mar 03, 2013

Tim cock using Nokia 3310 Lol .

41. Yes-Sir

Posts: 58; Member since: Sep 16, 2013

No they have not gone Samsunged! It shows how unknowledged and unaware you are, then again you come to PhoneArena, so you should be classed as a "Techie" What they have done was simple" they took the "Reference" device, installed SetCpu: a software to alter, modify the CPUs floating point. Normally all devices are set to "in demand" so 1 core remains active at 300Mhz, while other 3 sleeps. But on this case they went for "Performance" (mode) which the CPUs idles at 1.8Mhz at all times!

43. Ninetysix

Posts: 2933; Member since: Oct 08, 2012

Rename the benchmark app and benchmarks go back to "normal" values. Not cheating alright.

85. tuminatr

Posts: 1092; Member since: Feb 23, 2009

I find this very funny, and you cant truely blame them. Is it gonna stop me from buying a note 3..... NOPE

55. samsungIScopycat

Posts: 27; Member since: Jul 05, 2012

samsung always issued with cheat benchmark....and now note 3... never change their policy... copy + cheat + fraud + corruption = samsung~

63. amats69

Posts: 1527; Member since: Nov 12, 2012

Nothing new on Samsung..im not surprise anymore in this news...Samsung only know is Copying, Cheating, And hiring students to give negative review on their competitor's product, but still all samsung fans worship samsung..shame on you sammy! .#shamesung

3. Katrina

Posts: 84; Member since: Aug 17, 2012

o sammy, cheating at this era? you think your consumers are uneducated fools? (well most people are lols) well your scores for N3 is lower than Z1 now ....

23. GTR722

Posts: 270; Member since: Oct 20, 2012

Samsung fans and Apple fans are just the same...they will love anything that comes with a "S" or and apple.

54. Long1

Posts: 399; Member since: May 18, 2013

No! Note 3 scores better even than G2 without this "booster" By the way , lets see Samsung's anwer, not this be same like the "region lock" iManipulation.

4. newbey123

Posts: 696; Member since: Mar 19, 2012

LOL, not really surprised at this.

57. SlimSoulja86

Posts: 660; Member since: Nov 03, 2011

And you just believe everything you're told.

5. ocilfa

Posts: 334; Member since: Aug 03, 2012

I don't see why this matters. The G2, Z1, and a few other devices have the same chip as the Note 3. Wouldn't it be obvious to see if the benches were BS anyways?

7. Katrina

Posts: 84; Member since: Aug 17, 2012

no man, samsung fan bois goes crazy and show off when they see the numbers are higher than other OEMs.

8. silencer271

Posts: 254; Member since: Apr 05, 2013

love how they find some reason why an android device benchmark is good but on the iphone they take it as the law. iLosers.

10. darkkjedii

Posts: 30786; Member since: Feb 05, 2011

Ooooooohhhh naughty Sammy. I'm tellin mama. Mamaaaaaaaaaa Sammy cheatinnnnnnnnn!!!

18. PapaSmurf

Posts: 10457; Member since: May 14, 2012

Ha! +1

30. darkkjedii

Posts: 30786; Member since: Feb 05, 2011

I'll get called a fanboy for that, even though I was actually joking in the way kids tell mom lol. My niece is notorious for saying "I'm telliiiiiiiiin" while running to tell. +1

11. willard12 unregistered

Wow, PA. This week alone, samsung has copied apple's golden device, had thousands of their devices traded-in for iphones on gazelle, region locked the note 3, their watch does nothing, and are apparently their note 3 phone really isn't as fast as the benchmark scores say they are. Samsung has finally convinced numbers to lie. Meanwhile, at PA, Apple's slo-mo is awesome, their benchmarks are great and real, iphones are selling like crack, and they are better than coke. Of course, since most of those samsung articles turned out to be wrong, I've found it's better to wait until the update is posted in order to get the full/real story.

13. Bootutu

Posts: 228; Member since: Jul 11, 2013

Over-clocking???, its still running at specified speed, they clearly said the Powervr 544MP3 runs at 532MHz, and there are several stock apps that runs at that speed too, what a misleading article by PA again, and because the S800 is running at full speed under stress you call that a problem, anyways its normal if you are ruling, people will nitpick just to show a negative point. P.S: PA should go and revise the Note 3 biased review

19. PapaSmurf

Posts: 10457; Member since: May 14, 2012

Hmmm. I partially agree, so I'll throw you a +1. P.S. They need to revise the review ASAP.

* Some comments have been hidden, because they don't meet the discussions rules.

Latest Stories

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.