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.

Politics : Formerly About Advanced Micro Devices -- Ignore unavailable to you. Want to Upgrade?

To: kash johal who wrote (94588)	2/22/2000 1:27:00 PM
From: Epinephrine	Read Replies (3) \| Respond to of 1571929

RE <seems to me that clearly the FPU will suck compared to Athlon.> Check this out from the same news bit on JC's: "If all this meandering thought is true, it means that Willamette can start half as many integer SIMD instructions per cycle as P5/P6/K6-x/K7/Jedi, but each integer SIMD instruction can do twice as much work. I think (and don't hold me to this) that the tradeoff is thus: The new method would probably be a little slower (as it depends on a greater degree of explicit paralellism) and would not see benefit with regular MMX code, but saves a buttload on die area (I presume), probably having the useful side effect of making the cpu more apt to ramp in MHz. (granted, if I'm wrong about it being only one per cycle, then all this supposin' goes down the drain and I'd have to upgrade my integer SIMD opinion to something quite high<g>)" I am not trying to seem like I am on some pipe dream about Willamette being a non-contender I am posting these bits so that Elmer and maybe even Paul can show me where I am making mistaken conclusions and where I am being foolish. Something that I think that they will be more than happy to do :) In any event I think you are right it is sure going to get interesting! Thanks Epinephrine