Slashdot Mirror
AMD Rejects SYSmark Benchmark
Deathspawner writes "In an unusual move, Advanced Micro Devices has issued a press release rejecting its endorsement for the industry recognized benchmark SYSmark 2012. Developed by BAPCo and backed by industry heavyweights such as Dell, Intel and Hewlett-Packard, AMD has stated that BAPCo both has tuned SYSmark to create bias in favor of its competitor, and that its benchmarks are not relevant for the audience it targets. Also noted is a complete lack of heterogeneous CPU+GPU testing. Techgage tears apart AMD's claims to see if they are valid, while also evaluating the overall usefulness of SYSmark and the impact it can have on consumers."
Nvidia and Via quit too.
That's great! Why doesn't AMD go and write a compiler of thier own and give it away for free?
Its called Open64.
"His name was James Damore."
Well if they used the Intel compiler then it is for all intents and purposes useless as Intel has been rigging their compiler with a "Genuine_Intel" flag and if the flag isn't detected dropping all SSE and above optimizations and instead running in slow ass x87 mode. Last I read despite being ordered to change their behavior Intel is STILL putting out compilers with the evil bit on and haven't done anything to alert previous customers of their douchebaggery.
So I wouldn't be so quick to just dismiss out of hand, after all, who would have thought that Intel would rig their own compiler to cheat? I can't even imagine how many programs out there have been compiled using the Intel compilers which makes every single program written using their tool chain rigged against AMD and Via.
ACs don't waste your time replying, your posts are never seen by me.
Um, they do have their own compiler: http://developer.amd.com/tools/open64/Pages/default.aspx
Seems to be both free to download, and comes with source code so you can go over it if you wish.
...and make a compiler.
They did. It's even GPL licensed.
http://developer.amd.com/tools/open64/Pages/default.aspx
upon the advice of my lawyer, i have no sig at this time
The most recent versions of Intel's compiler clearly document the evil behavior in the description of the code generation options, so at least it's not hidden any more. You can also mostly disable the evil behavior, if you are willing to sacrifice the runtime code-path selection that allows you to use SSEx on hardware that supports it while retaining compatibility with earlier machines.
Still, any benchmark using Intel's compiler can't be trusted unless it is fully open-source, including the exact compiler flags used.