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: Elmer who wrote (66975)	7/30/1999 12:51:00 AM
From: Joey Smith	Read Replies (2) \| Respond to of 1575181

Elmer: re:Actually, I can report that Coppermine will be within a few % of the MIA-thlon in FP but exactly how they accomplish this I can't say. That can't be. Scumbria and Charles believe Intel wont have a competitive product for Athlon for AT LEAST 12-18 months <gg> joey

To: Elmer who wrote (66975)	7/30/1999 1:06:00 AM
From: Cirruslvr	Read Replies (1) \| Respond to of 1575181

Elmer - RE: "Actually, I can report that Coppermine will be within a few % of the MIA-thlon in FP but exactly how they accomplish this I can't say." I don't see how the SPECfp score can increase at least 30%, unless 64Kb of L1 cache is used or the SPECfp program is designed to take advantage of up to 256KB cache or if it is SSE optimized. I was about to post SPECfp scores comparing the Celeron and PII 333 and 400 to try to show what I wrote above, but after getting the data here, I realized I couldn't make any judgement over how much the 128Kb cache hurts the Celeron because the PII 400's 100MHz variable would have skewed the results. I was going to use the Celeron and PII 333 and 400 because those are the only overlapping MHz. When the Celeron 500 comes out next week, we might be able to garner some kind of guesstimate of how much we expect 128Kb more cache to help, but again, the 66MHz bus of the Celeron will skew the data. It looks like we will have to stick with the Dixon vs. non on chip L2 cache mobile PII comparisons which don't take into account any potential gain from SSE. BTW, If Cuontimemine can only get within a few % of the non-optimized Athlon in SPECfp WITH SSE optimizations, the world (or at least the people who care ;) ) will be better able to see how POWERFUL the Athlon's FPU is! Also, if it is still a few % slower WITH SSE, AMD will have the highest performing x86 FPU based on SPECfp scores until Intel can come out with Willa.., IF that can even beat the Athlon!

To: Elmer who wrote (66975)	7/30/1999 1:36:00 AM
From: Charles R	Respond to of 1575181

<The person who I heard make this claim was VERY high up inside Intel and his name would probably be recognized. > So high up that he lost touch with ground?