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 (76793)	10/25/1999 11:32:00 AM
From: Steve Porter	Read Replies (1) \| Respond to of 1572771

Process Boy, Yes, but JC has got wind that c't (you know the german mag) got INCORRECT results on SPEC (i.e. the output of the program was wrong) when running on a Coppermine system with 1 RIMM. I don't have all the details and I would hate to scoop JC on here, so we'll wait and see what he digs up. But this has shades of the Pentium FPU bug all over again. Steve PS: the story first broke here: jc-news.com

To: Process Boy who wrote (76793)	10/25/1999 12:32:00 PM
From: Michael DaKota	Respond to of 1572771

"the new compiler uses prefetching to boost the PIII scores (Athlon supports the same opcodes, and benefits to an even greater degree from prefetching, and the results can only be published after the compiler is released. Basically, the end result is that the 35.3 specint and the 30.4 specfp (much higher, especially in fp, than the scores with current standard compilers) for the 733 do not count until sometime next quarter. Likewise, the Athlon-700 projected scores (>35 specint and between 34 and 42 specfp) do not count until a compiler that supports prefetching for the Athlon pops up" JC's news . FHWL, michael.