Silicon Investor (SI) -- The First Internet Community

STOCKTALK

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor. We ask that you disable ad blocking while on Silicon Investor in the best interests of our community. If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.

Technology Stocks : Advanced Micro Devices - Moderated (AMD) -- Ignore unavailable to you. Want to Upgrade?

To: Charles R who wrote (19088)	11/14/2000 6:25:55 PM
From: fyodor_	Respond to of 275872

<Chuck: I gather there is significant advantage to packing twice as many vector operations in SSE2 as you can in SSE.> I don't know that it is possible to pack twice as many single precision vector operations in SSE2 as in SSE. It is my understanding that SSE2 does nothing to enhance single precision floating point math (directly, anyway). Both games and SPECfp would benefit from it, however, if it the possibility you describe was present. -fyo

To: Charles R who wrote (19088)	11/15/2000 11:14:35 AM
From: jcholewa	Read Replies (1) \| Respond to of 275872

> I am not familiar with either the composition of SpecFP code or the software architecture for games but I gather there > is significant advantage to packing twice as many vector operations in SSE2 as you can in SSE. Yeah. It's too bad that SSE2 doesn't have this advantage for single precision fp. The only thing it affects is integer (it effectively makes P4 as good as PIII or Athlon in vector integer, but only if you use the new instructions) and double precision floating-point. Games would have no difference from PIII, as they would rely on single precision fp. Well, not counting microarchitectural changes.     -JC