[Follow-Up] Samsung Denies Allegations of SGS4 Benchmark Manipulation

[dgstorm]

Just yesterday we shared that Samsung was caught using sneaky methods to manipulate and inflate benchmark scores on their Galaxy S4 line of phones. A number of tech websites, including the highly reputable Anandtech discovered what seems like an obvious deception. From their testing the SGS4 would increase its GPU clock speed from 480MHz to 533MHz, but it seemed to only do it when running benchmarks like AnTuTu & Quadrant.

Since the revelation was published yesterday, Samsung quickly moved to respond. They began with a very vague initial response, which stated, "[We] did not use a specific tool on purpose to achieve higher benchmark scores."

Later on they shared a slightly less vague denial and posted it on their official website. Here's a quote with their full response,

Under ordinary conditions, the GALAXY S4 has been designed to allow a maximum GPU frequency of 533MHz. However, the maximum GPU frequency is lowered to 480MHz for certain gaming apps that may cause an overload, when they are used for a prolonged period of time in full-screen mode. Meanwhile, a maximum GPU frequency of 533MHz is applicable for running apps that are usually used in full-screen mode, such as the S Browser, Gallery, Camera, Video Player, and certain benchmarking apps, which also demand substantial performance.

The maximum GPU frequencies for the GALAXY S4 have been varied to provide optimal user experience for our customers, and were not intended to improve certain benchmark results.

Samsung Electronics remains committed to providing our customers with the best possible user experience.

Their response seems to have some plausibility. The only problem with this denial is that during testing at Anandtech, they could find no other app which used the higher clock speed bump for the GPU. This still leaves things in a grey cloud of doubt. What do you think? Did Samsung doctor the results, or is this just a case of mistaken intention?

Source: Samsung Tomorrow

 

[WishRyder]

I'm wondering why this matters? If the GPU is capable of it, I'm sure apps can be written to take advantage of it. Just because the benchmark is the only one at this time, doesn't make the results any more invalid than any other benchmark.

Furthermore, on the mobile side, ROMs have been manipulating benchmark performance since the beginning. On the desktop side, Nvidia and AMD have been tweaking their drivers to manipulate certain benchmarks for years. Vice versa, certain benchmarks have been optimized to certain hardware as well.

I could go into further detail, but the point is, people shouldn't rely too heavily on benchmarks.

 

[leeshor]

That makes sense to me. Where they use the word overload just substitute overheat.

 

[skennelly]

I remember a day when all of these benchmark apps had zero credibility and "real world performance" was the ultimate test. Guys like kejar and jrummy used to bake "benchmark enhancements" into their roms. Seems pointless. Anyone that's seen the S4 knows the phone is lightning fast, that's all that matters.