HTC admits its High Performance Mode pumped up benchmark scores for the HTC One (M8)

HTC admits its High Performance Mode pumped up benchmark scores for the HTC One (M8)
Last year, Samsung was accused of putting technology into its high-end Android phones that could determine when a benchmark test was being run, and boost the frequencies of the CPU on the device. Samsung did admit that "the Samsung Galaxy Note 3 maximizes its CPU/GPU frequencies when running features that demand substantial performance."

Now, it is HTC's turn to be accused of pumping up benchmark results. And the phone it is accused of cheating with is none other than the HTC One (M8). It seems that AnTuTu has added a new version of its popular benchmarking site called AnTuTu X, which is supposed to eliminate some of the cheating that has apparently has been going on. AnTuTu X is supposed to show a more authentic and realistic benchmark score.

When the HTC One (M8) was run through the AnTuTu 4 benchmark test, the phone scored a whopping 38,815. That topped the 34,898 scored by the Samsung Galaxy S5. The Sony Xperia Z2 scored an impressive 32,768 and the LG G Pro 2 produced a score of 30,068. 

But when the HTC One (M8) was put through the cheat-proof AnTuTu X, the results were much different. The Samsung Galaxy S5 was on top with a score of  35,357. The Sony Xperia Z2 finished second with a benchmark score of 32,508 while the LG G Pro 2 was next with a score of 29,787. The HTC One (M8)? HTC's new flagship had a score of 27,171.

Like Samsung, HTC admitted in an email response that it had designed the phone to "optimize in certain scenarios to produce the best possible performance." HTC actually has a name for this so-called feature, the High Performance Mode, which can be enabled or disabled in developer settings. The feature is not yet in HTC's U.S. models, but will soon be added via a software update.

Time eventually healed the wound created by Samsung's optimization of benchmark tests. Will HTC's actions also be forgotten after a few months?

source: ePrice (translated), CNET via GSMArena

Related phones

One (M8)
  • Display 5.0" 1080 x 1920 pixels
  • Camera 4 MP / 5 MP front
  • Processor Qualcomm Snapdragon 801, Quad-core, 2300 MHz
  • Storage 32 GB + microSDXC
  • Battery 2600 mAh(20h 3G talk time)



1. Calm_down_mon

Posts: 43; Member since: Dec 12, 2013

The old Samsung story >.> who the hell cares about these f-ing benchmark scores

24. The-Sailor-Man

Posts: 1095; Member since: Mar 25, 2014

Samsung was attaced , and we know why, But all the Android OEM's was doing , THE SAME. Even Anand wrote it. But the tech media want to point only Samsung.(why?) And Anand wrote that Note 3 beat all of them anyhow.

47. hmmm...

Posts: 81; Member since: Jun 05, 2013

Motorola did not cheat and Sony phones were not on the list of phones he tested.

54. AfterShock

Posts: 4147; Member since: Nov 02, 2012

Nor was Nexus phones bumped up for tests. So, not all phones are thrown into the ring of shame.

2. apple4never

Posts: 1064; Member since: May 08, 2013

LOL 27k? my dual core, midrange Xperia SP get 23k

18. fireblade

Posts: 717; Member since: Dec 27, 2013

please post the picture. No pic = hoax and you're a liar

33. apple4never

Posts: 1064; Member since: May 08, 2013

36. zekes

Posts: 230; Member since: Aug 14, 2012

Good phone it beat the hell out the note 2

68. apple4never

Posts: 1064; Member since: May 08, 2013

lol yup, i got it for free with a 150$ gift card too!!! and the fact that it beat the flagship it launched aside is startling and it got so close to the M7

3. Lt.Green

Posts: 397; Member since: Mar 13, 2014

Real time use is the true score, not benchmarks.

14. vincelongman

Posts: 5724; Member since: Feb 10, 2013

Unfortunately, you can't easily compare real life usage, so people use benchmarks Also most benchmarks I've tried only use 2-3 of my Nexus 5's cores and its only at ~70% usage for each of those cores. And a slightly lower benchmark score can also be good as it means you CPU is using less battery

23. sprockkets

Posts: 1612; Member since: Jan 16, 2012

If the same performance is attainable with a performance setting, I have nothing against it.

26. Arte-8800

Posts: 4562; Member since: Mar 13, 2014

My note 3 is set on performance mode, and not on demand . Via custom kernel.

4. vintage

Posts: 15; Member since: Feb 12, 2014

(htc fans) the camera may be a little out dated but we beat every one else In benchmarks. And it's metal. Hahahahah

15. V-Dawg

Posts: 6; Member since: Feb 27, 2014

Dude, you might want to look again, the HTC did'nt beat any of the three devices listed above in the legit (normal grounds of testing) mode. Actually the score is way way way low. You are also wrong about the camera, the tech of dual cam is somewhat new and they say the've improved the hardware of the cam (Not in terms of MP), but the ram is still DDR2.

17. PapaSmurf

Posts: 10457; Member since: May 14, 2012

Right? Lol.

21. vintage

Posts: 15; Member since: Feb 12, 2014

?? I was being sarcastic smart a** if u look I said (HTC fan) which means I was speakn as to be one think you please if your going to try and correct some one please read carefully next time thank you and papa smurf just have a seat and do something with your life

27. pwnarena

Posts: 1129; Member since: Feb 15, 2013

you've just encountered two people who have consitently exhibited questionable abilities in reading english texts. i think maherk and rd_nest will follow.

34. vintage

Posts: 15; Member since: Feb 12, 2014

Haha I can tell I thought my post was self explanitory if I do say so my self but I guess not I would think with the 6000+ comments papasmurf made he would learn how to read

44. giani

Posts: 269; Member since: Nov 13, 2012

The ram is actually DDR3. So now the world can relax. For more details check Anandtech 's review.

60. vintage

Posts: 15; Member since: Feb 12, 2014

Actually that comment needs to go number 15 I never said anything about DDR2 or DDR3.

79. V-Dawg

Posts: 6; Member since: Feb 27, 2014

pwnarena @ oh is that so, and where else have you seen me replying a comment. Just because people don't say what you want them to say, don't thrash them.

62. Ashoaib

Posts: 3298; Member since: Nov 15, 2013

Hahaha.. you reminded me a proverb "out of a pan, into the fire"... its ok, forgive both of them.. sometime misunderstanding happen

66. vintage

Posts: 15; Member since: Feb 12, 2014

Who says I'm mad I'm not mad at all but at the same time please don't try and correct me if you don't understand my comment. And you can't forgive some one you were never mad at.

78. V-Dawg

Posts: 6; Member since: Feb 27, 2014

Yeah right!!!! just because you typed in "HTC FANS" at the start doesn't mean youre being sarcastic. Anyway I apologize and next time try to be accurately sarcastic.

81. vintage

Posts: 15; Member since: Feb 12, 2014

Can you shut up I don't care about your opinion thank you just because your slow doesn't mean I wasn't being sarcastic please have a seat. And the 14 People that liked my comments must have got it

64. Berzerk000

Posts: 4275; Member since: Jun 26, 2011

It's DDR3 RAM. Check Anandtech's review.

5. Mxyzptlk unregistered

Benchmarks are useless most of the time and don't prove anything. It's just like using a carrier coverage map over using something like Sensorly for real world feedback.

28. Arte-8800

Posts: 4562; Member since: Mar 13, 2014

To you it might not, to others, like who actually finds out when playing games shows how many FPS, refresh rates, the speed of the memory> bandwith, vectors, and etc,,,,,these are advanced users! unlike me and you and others are novice to this. When your car goes on service it uses an app just like antutu, to figure out what's it's performance is and what is it know after few years, ,, rolling road map Even CPU engineers use similar CPU app like antutu to abstract gains and flows of CPU plot algorithm

59. chebner

Posts: 249; Member since: Oct 17, 2011

I know some users like to know what fps they're getting at any split second. But what matters more, insane fps or that the game plays smoothly? With a PC plugged into the wall power consumption doesn't really matter. But in a phone where battery power is finite anyone with common sense should prefer minimum fps to provide smooth game play vs max fps, the power savings will allow you to play longer. What the help are you talking about hooking your car up to? That's just a diagnostic computer that tells you what is wrong. Only a dynometer can tell you how much power your car has. Most people probably don't know what a dynometer is let alone had their car tested. Yes CPU engineers use benchmarks to test against a known baseline so they can tell how their chip is performing, but these tests have little to do with real world performance.

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 or use the Reprints & Permissions tool that appears at the bottom of each web page. Visit for samples and additional information.