To: HangThemHigh
Due to Stanford using an Intel compiler, these problems will run faster (2x reportedly) on Intel hardware with SSE2. Some AMD cpus have SSE2, but the Intel compiler doesn't take advantage of it. Interesting, since GCC 4.x generally produces code that is competitive with the Intel compiler speed-wise, without the Intel bias. The Pathscale AMD64 compilers (popular with high-performance apps) can suck an amazing amount of floating point performance out of AMD hardware, but you have to pay for the privilege. Most of these problems appear to be better suited to AMD than Intel, based on a loose survey of typical parameterization. Same ISA, significantly different design choices.
141 posted on
05/06/2006 6:37:06 PM PDT by
tortoise
(All these moments lost in time, like tears in the rain.)
To: tortoise
You obviously no a lot more about compilers than I do, but there has been a good bit of discussion about this on the Folding forums which might interest you. Unfortunately I have not kept links to the discussions.
Stanford has a rational for using the Intel compiler. I seem to recall they claimed it produces faster code than the AMD compiler, which was the only other candidate even close in terms of speed.
They say that are working to resolve the problem.
142 posted on
05/06/2006 7:19:01 PM PDT by
HangThemHigh
(Entropy's not what it used to be.)
To: tortoise
143 posted on
05/07/2006 5:07:31 AM PDT by
HangThemHigh
(Entropy's not what it used to be.)
FreeRepublic.com is powered by software copyright 2000-2008 John Robinson