Most Android Handset Manufacturers are Cheating Benchmarks, expect Apple, according to a report.

[j0hnlam]

So apparently AnandTech (a news tech site that Linus really loves), had done an investigation in manufacturer's benchmarks and found that majority of these OEMs are cheating the system to make their hardware superior. 

 

 

With the investigation, it has been found that Apple and Motorola are the two OEMs that don't cheat on benchmarks. 

 

 

 

With the exception of Apple and Motorola, literally every single OEM we’ve worked with ships (or has shipped) at least one device that runs this silly CPU optimization. It’s possible that older Motorola devices might’ve done the same thing, but none of the newer devices we have on hand exhibited the behavior. It’s a systemic problem that seems to have surfaced over the last two years, and one that extends far beyond Samsung.

 

Now you may argue what is actually cheating and isn't but I think it's great to see a major tech publishing site putting this issue into the limelight. 

 

Despite all of the effort that OEMs put into benchmark optimizations, the gains are negligible. The impact on CPU tests revealed a 0 to 5 percent performance increase, and a less than 10 percent increase on GPU benchmarks.

 

Thoughts? (I can already predict the Apple hate comments about saying "Apple not cheating" is a lie...)

 

Link: http://www.anandtech.com/show/7384/state-of-cheating-in-android-benchmarks

[helping]

posted already.

 

also dat samsung

[mgsstar]

HTC and Samsung seems to be competing with each other because they have the most programs that are cheated on.

 

Like the Nexus devices and the Nvidia device are being good people and are not cheating on benchmarks in order to give users their true performance.

[martinrox1568]

Lol "except apple" who the fuck do they have to compete with with on iOS?

[loganryan99]

Apple would have no reason to cheat because people who use iPhones obviously don't care about benchmarks.

 

On a serious note, did benchmarks seriously matter to people even before this happened?

[j0hnlam]

Like the article stated, the CPU optimizations that OEMs used only give a performance increase of 0 to 5 percent. Very minimal. 

The benchmarks really don't matter even before this happened. 

[SMURG]

I really don't see how this is cheating in all honesty. If the phone can get that score, even with what is essentially an overclock, then what's the big deal? It's not like they goofed the numbers, they just increased their clock speeds in order to maximize the result. It's possible that these phones also ramp up their clock speeds for other stressing applications, so saying that this is "cheating" is like saying that Intel's "boost clock" technology is cheating too.