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: Process Boy who wrote (57351)	5/5/1999 11:15:00 PM
From: Cirruslvr	Read Replies (3) \| Respond to of 1572778

PB - RE: "What do you mean by "as expected"? There are some fairly decent rumors floating around about Coppermine too. What IF the Coppermine performs better than you expect?" Au contraire, my Intel friend. I basically know what to expect. We practically do know how well Coppermine will perform, if it contains 32K L1 cache. From intel.com, Intel has benchmarks which compare the mobile PII to the Mobile PII PE, otherwise known as the Coppermine w/o KNI. To see the numbers for yourself, look at the chart at the bottom of the thing instead of the graphs. (mobile) Winstone 99 MultimediaMark 99 3DMarkCPU SPECint95 SPECfp95 JMark 2.0 PII 266 1.06 1.13 1.12 1.11 1.10 1.14 PII 266PE 1.11 1.13 1.13 1.19 1.26 1.19 % PE is 4.7% * * 7.2% 14.5% 4.4% faster PII 300 1.10 1.26 1.24 1.24 1.19 1.28 PII 300PE 1.16 1.26 1.25 1.33 1.35 1.34 % PE is 5.5% * * 8.9% 13.4% 4.7% faster * - % difference is negligible The faster bus should speed things up some too, but we have no numbers on that. Any estimates as to what we can expect? In integer applications, Coppermine should be as fast as a next speed bin PIII. (Sound familiar?) In FPU-heavy applications, Coppermine should be as fast as a same clock speed PIII (again, sound familiar?), except in the heavily L2 cache dependent SPECfp95. When I said "as expected" with the K7, I mean the claim AMD has made. I never knew that SPECfp95 was THAT L2 cache dependent. If full speed cache makes Coppermine 13-14% faster than PIII w/half speed cache, I wonder how much of a % jump the K6-III would have over a K6-2 when the difference in L2 speed is 2X greater than the difference in the Intel chips. Does anyone have any numbers on this???