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: Dan3 who wrote (20090)	11/21/2000 3:03:22 PM
From: Jim McMannis	Respond to of 275872

RE:"Anand is capable of making his own SPEC runs. My guess is that one consideration is that his runs resulted in scores substantially lower than Intel's published scores. Rather than claim (explicitly or implicitly) that Intel's SPEC runs were not reproducible, he just dropped the issue for now." Now that's a new twist. Anand had mercy on Intel? <G> Jim

To: Dan3 who wrote (20090)	11/21/2000 3:21:02 PM
From: Daniel Schuh	Read Replies (1) \| Respond to of 275872

My guess is that one consideration is that his runs resulted in scores substantially lower than Intel's published scores. That's one of my takes, too. If you look at the fine print in spec.org , Intel almost always publishes scores based on some compiler version with an alleged availability date months in the future. For example, the current top gigamine scores explained in spec.org Tested by: Intel Corporation Test date: Jun-2000 Hardware Avail: Apr-2000 Software Avail: Oct-2000 If Anand insists on running the benchmarks himself, and I'm sure he does, he either has to work with what's actually available or get the Intel compiler operation in to do a custom installation. In the event of the later, I bet he'd have some amusing tales to tell. I imagine he just decided to bag it, rather than go through the hassle (with the added bonus of random flamage from disinterested parties afterwards :-\|). Cheers, Dan.

To: Dan3 who wrote (20090)	11/21/2000 3:51:12 PM
From: andreas_wonisch	Respond to of 275872

Dan, Re: My guess is that one consideration is that his runs resulted in scores substantially lower than Intel's published scores. With the 4.0 compiler probably. C't benched a 1.5 GHz P4 with both version 4.0 and 5.0 of Intel's compiler in SPECfp2000 and got 403 with 4.0 and 502 with 5.0 (both scores are base). This is a very impressive 25%(!) increase but still 10% below Intel's published numbers. Also note that a 1.2 GHz Athlon with Compaq's newest compiler is almost as fast (ca. 370) as a 1.5 Pentium 4 with Intel's 4.0 compiler in SPECfp. Andreas