Stefan Mileschin

Apple fanzine Appleinsider discovered the much touted Samsung Galaxy S4 figures had been inflated by benchmark fixing.

Daniel Eran Dilger said that Samsung had been increasing the clock speed of its Android-based Galaxy S4 when running certain benchmarks.

Versions of Samsung's Galaxy S4 equipped with the company's new Exynos 5 Octa SoC were announced as Samsung's answer to Apple's A6X.

Both chips are manufactured at Samsung's System LSI chip fab and Apple uses its own DIY "Swift" CPU core design, paired with GPU core tech it licenses from Imagination Technologies, providing four PowerVR SGX 554MP4 cores to power graphics on the iPad 4.

Samsung's Exynos 5 Octa used ARM's stock Cortex-A15 CPU core design, and pairs four of them with essentially the same Imagination GPU design as Apple.

So there was a bit of spat to see which of them was faster. Samsung appeared to win hands down, so the Apple fanboys looked at Samsung's chip.

They found that Samsung was specifically enabling full speed GPU performance only when running specific benchmarks, and then reverting to slower 480MHz speeds when being used to play games or run other apps.

It is worth pointing out that Samsung never publicly claimed the maximum GPU frequencies for the Exynos 5 Octa.

When running LBenchmark 2.5.1, AnTuTu and Quadrant the device "triggers a GPU clock not available elsewhere: 532MHz".

The CPU side behaved similarly when running benchmarks, but it used maximum performance modes

AppleInsider sniffed that there are strings for Fusion3 and Adonis indicating that Samsung isn't just cheating on benchmarks for one of its processors, it's part of the company's culture.

It is possible that a spec war similar to what happened on PCs is taking place in the smartphone industry. Every chip maker has been accused of doing this at one time or another.

What is interesting here is that the Apple press did not look for similar code under the bonnet of Apple's toy. We are not saying it exists, but if you are going to accuse one side of breaking the rules of engagement you have to also look at the other side too.

Nor did they point out that even if the fudging was taken into account, the Samsung set-up was still faster.

http://news.techeye.net/chips/samsun...ged-benchmarks