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 : Intel Corporation (INTC) -- Ignore unavailable to you. Want to Upgrade?

To: Dan3 who wrote (122742)	12/13/2000 9:13:17 PM
From: Tenchusatsu	Read Replies (1) \| Respond to of 186894

Dan, <But if you have a compiler that recognizes the spec code sequences and substitutes updated algorithms as an "optimization" it is within the rules.> It is my understanding that you can optimize the SPEC source code in any way that you want (at least for the non-"base" scores), whether it's through a simple compiler switch or through profiling and hand-assembling critical code sections. Those algorithms you are talking about are the products of hand-optimization, not a bait-n-switch tactic by the compiler (which is a silly idea anyway). <Intel really is cheating on SPEC - but cheating within the rules, I suppose.> Lots of vendors rely on profiling and hand-optimized code when publishing SPEC scores. Intel isn't the only one. You can call it "cheating" if you want. I'll just call that "sour grapes." By the way, the high SPECfp scores that you see for Pentium 4 really is due in large part to the memory bandwidth. This isn't a big secret; lots of people already know that. I would argue that most FP programs out there also care about memory bandwidth (except Excel and other spreadsheets), but I doubt you'd believe me. At least not until Athlon gets there ... Tenchusatsu