That stinks. Just as Intel AnTuTu cheat, this is not acceptable and I hope Samsung will get the treatment it deserves (and I am sure more people here will blame Samsung than they were people blaming Intel for cheating).
That stinks. Just as Intel AnTuTu cheat, this is not acceptable and I hope Samsung will get the treatment it deserves (and I am sure more people here will blame Samsung than they were people blaming Intel for cheating).
Repeating this doesn't make it true. Even Antutu revised their benchmark to remove this. The only question is where the source of cheating came from (it's theoretically possible that it's just an accident but doesn't seem likely).Using ICC vs generic compiler is not cheating.
For the purposes of keeping things intellectually honest i wanted to point this out. Since the Intel Antutu benchmark seems to reverberate in an echo chamber, while shenangians like this are diminished
For the purposes of keeping things intellectually honest i wanted to point this out. Since the Intel Antutu benchmark seems to reverberate in an echo chamber, while shenangians like this are diminished
Feel free to point out the arstechnica and Anandtech articles discussing Intel's AnTuTu shenanigans, I'd love to read them.
picking two sites and saying it was not reported is a false equivalency.
Considering they are two of the biggest tech sites on the planet that makes it even less equal, yeah.
Still in denial heh? :biggrin: Using a compiler that explicitly improves code for a loop in a benchmark is cheating, it's exactly the same as detecting a benchmark and up frequency.Using ICC vs generic compiler is not cheating.
I totally agree.A turbo that only works in benchmarks to make the product look better. Not much to say about it. And specially not since this is the second device to do so. Its simply really bad ethics.
Still in denial heh? :biggrin: Using a compiler that explicitly improves code for a loop in a benchmark is cheating, it's exactly the same as detecting a benchmark and up frequency.
Actually, if the benchmark is at all properly written and the DVFS as regulated by the operating system (and user settings) is at all reasonable then this is exactly what should happen when you launch a benchmark. If you're running something that pegs the CPU it should be given as much CPU time as possible within the operating limits of the device.However, if you load up just about any popular CPU benchmarking app, the Note 3 CPU locks into 2.3GHz mode, the fastest speed possible, and none of the cores ever shut off. Stopping the CPU from idling shouldn't in and of itself affect the benchmark scores a whole lot, so this was our first sign that something was wrong. Benchmarks exist to measure the performance of a phone during normal usage, and a device should never treat a benchmark app differently than a normal app.
Using ICC benefits all x86 applications. Its also the fastest compiler for any x86 CPU.
While you can claim the usage could be misleading using optimized vs non optimized compilers. Then sure.
But its still not a cheat, since its something the enduser can get.
No it's not something you can get, because only the loop in that particular benchmark gets optimized so it benefits only the benchmark. Even turning 64-bit code generation made the loop dozens of times slower, how do you explain that?Using ICC benefits all x86 applications. Its also the fastest compiler for any x86 CPU.
While you can claim the usage could be misleading using optimized vs non optimized compilers. Then sure.
But its still not a cheat, since its something the enduser can get.
It must suck to have to cheat on your benchmarks.
Yet another reason I bought an iPhone 5S.
haha amazing that half the posts here were on intel yet again. can we just all agree that everyone seems to dope their benchmarks
Except AMD/Nvidia. They would crucified if they were caught doing this. Gamers take that stuff seriously. Don't know about you mobile folks.
Quack 3, anyone? lol
Lol.. well I was going to say not anymore anyway.
Actually, if the benchmark is at all properly written and the DVFS as regulated by the operating system (and user settings) is at all reasonable then this is exactly what should happen when you launch a benchmark. If you're running something that pegs the CPU it should be given as much CPU time as possible within the operating limits of the device.
The problem here is the benchmark must not have enough of a warmup period to convince the OS under normal conditions that it needs to ramp up the clock speed fast enough. So by the time the OS realizes that it needs to give the app full frequency the benchmark is already over, or at least has ran enough for the score to be heavily impacted. This is exactly why any competent benchmark has a long warmup period, and another reason why AnTuTu is garbage. But it looks like even GB3 doesn't have enough of one.