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: Petz who wrote (56674)	9/28/2001 7:10:30 PM
From: Jim McMannis	Read Replies (2) \| Respond to of 275872

RE:"I think the best way to look at AMD's rating scheme is that the model name is an extremely conservative estimate of equivalent P4 performance" OK but I still think it was based on the thunderbird. Has AMD ever said what it was based on? Jim

To: Petz who wrote (56674)	9/28/2001 7:19:55 PM
From: andreas_wonisch	Read Replies (2) \| Respond to of 275872

Petz, Re: I think the best way to look at AMD's rating scheme is that the model name is an extremely conservative estimate of equivalent P4 performance, even on Elmer's so-called "New Code." Good point. I did a quick calculation on the Athlon SPECint2000 scores and came to the conclusion that a 1.53 GHz Athlon XP with KT266A will probably be as fast as a 1.8 or 1.9 GHz Pentium 4 (on base). And that's a best-case scenario for Intel. The only code where it will be still beaten by a 1.8 GHz P4 is of SPECfp2000 type (i.e. heavily optimized and very bandwidth dependant). So it's a good guess that on almost any application/benchmark imaginable an Athlon XP 1800+ will actually beat a 1.8 GHz P4 or be very close in performance. And AMD's new rating makes only sense if this is actually the case. If there are some benchmarks where the 1.8 GHz P4 wins and some where the Athlon edges out, the new rating will be a disaster. But if AMD is able to make a good point that almost every software runs faster, it will be easier accepted. Andreas PS: AMD might even make SPECint their new standard to measure CPU performance. It is an industry-wide accepted benchmark and Athlon already does very favorable compared to P4 there (clock-for-clock parity). According to the submitted benchmarks a Palomino is 11% faster in SPECint clock-for-clock. Add to that another 10% KT266A bonus and you'll get the 1.8/1.53 ratio.