• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

altivec vs. sse2 benchmarks

jhu

Lifer
i can't find any direct performance benchmarks comparing the two. anyone want to help me write one?
 
Running DNETC with an altivec core offers a 400% speed increase, running it with an SSE2 core offers a 25% increase.
 
i was thinking of common calculations that scientists would encounter, but i don't really know any. also, i don't have a ppc computer.
 
It's kinda tough to do direct comparisons of the 2 because you always have to wonder how well coded for SSE or altivec the program is. I've had some experience programming SSE and the problem is, c/c++ compilers can't handle sse as well as it can plain floating point or integer code. Because of this, you have to resort to using compiler "intrinsics" but doing that sucks compared to writing regular math forumulas.
 
Back
Top