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: pgerassi who wrote (101392)	4/1/2000 12:40:00 AM
From: Elmer	Read Replies (2) \| Respond to of 1572004

Re: "Dear Elmer: Ok, I see your point. But you are incorrect as well. The Coppermine was running on 256MB PC100 on a BX. 19.8 SPECbase 95fp on that platform is probable given previous published benches" No 19.8 is not probable given the published benchmarks. Dell publishes the number 32.1 based on the latest available compiler, as claimed by AMD. There is a major discrepency between the two scores and Intel is not the one making the claim, Dell is. spec.org Notice also that Dell publishes the compiler options used, as required by SPEC. AMD refuses to publish the compiler options used when measuring such low CuMine scores. A very questionable act to say the least. That's why no one else has been able to measure anything that low. They don't know how AMD did it and AMD won't tell. Does this seem honest to you? Fujitsu posts 30.2 spec.org Even Compaq posts 30.6 Does Compaq strike you as anti AMD? spec.org Re: "A German Web site ran SPEC on Coppermine and Athlon at same clock on gcc running on Red Hat Linux, I believe. Coppermine was trounced. Do you know of any other published benches were run on both machines using the same third party compiler such as gcc, Visual C++ (No Intel SPEC plugin), WATCOM, or Solaris. This would settle the Fairness test. Pete" The German benchmarks I saw showed CuMine still winning on INT but I don't remember seeing any FP scores. Perhaps you have a link? In either case, I don't think it is unfair for Intel to develop a commercial compiler that optimizes code for CuMine. There's nothing stopping AMD from doing the same for their processor. The fact is that based on available silicon and available compiler technology, CuMine beats Athlon in both INT and FP when the best available hardware and the best available compiler technology for each device is used. I know of no other tests which use any other third party compilers. If anyone does, I hope they will post a link. EP