Now we know another reason as to why the Galaxy S5 at MWC benchmarked lower than the Note 3: Samsung has removed the performance boosting code from its devices running Android 4.4.2. Sure, the Note 3 has an additional gigabyte of RAM to boot, but without that benchmark ''optimization'' in play, the S5 results are more realistic. Some of you will remember a while back I was all in a fluster about the artificial benchmark boosting that Samsung had going on in various Android 4.3 devices, including the Galaxy S4 and Note 3.
While several tech writers, including Brian Klug at AnandTech and Ron Amadeo at Ars Technica looked into it further and uncovered the optimization running in both the Galaxy S4 and Galaxy Note 3, several of our readers seemed to think optimization of a device in benchmark tests was totally acceptable. I, and many others like me, felt otherwise, and saw the optimization as a way to artificially skew benchmark results to produce better results than are actually possible in real-life, which, while common in the PC world, is still an attempt to delude consumers in a seemingly ''objective'' test into thinking they're getting something better than they actually are.
To confirm these findings, Primate Labs' John Poole was able to pull data from the Geekbench Results browser to see how the scores varied from update to update. Poole was able to confirm both that a benchmark boost was added to many of Samsung's devices beginning with the Android 4.3 update and that the boosting does not seem to be present in the Android 4.4 update. The team has also added some detection code in a recent Geekbench update that would note when boosted scores were uploaded to that database—Samsung's Android 4.3 software sets the detector off, while 4.4 does not. - Andrew Cunningham, Ars Technica
Samsung clearly took note of the negative press and has now removed the ''benchmark booster'' code from its firmware. Previously, when certain benchmark tests were launched, all of the cores in the devices would fire at full throttle for as long as the benchmark test was running. While this may sound entirely logical for a benchmark test, the problem lay in the fact that this performance is not experienced in normal usage, so while it is representative of ''possible'' performance, it's not performance that you could ever achieve. None of this matters now though, as it seems that all Samsung devices making the jump to Android 4.4.2 will have real-world benchmark performance results from here on out. And that's a much nicer result for everyone.
If you want more details on the specifics, head on over to Ars Technica for the full story.
Are you happy to see ''honest'' benchmarking from Samsung? Do you place any value in benchmark test scores?