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: Steve Porter who wrote (61627)	6/13/1999 3:55:00 PM
From: grok	Read Replies (1) \| Respond to of 1583867

RE: <Well all I know is that the test was run in 'true' spec fashion. That means no heavy compiler tweaking (sun, dec, hp and all do this). It was pretty much an identical run for both machines and therefore spec was used to measure the difference in performance between the two machines and wasn't used to look for an absolute number that you can compare to alpha spec mark etc.> I think that the Intel compiler has already had the heavy tweaking needed to get it to deliver good Specmarks. But now it is available to everyone including AMD so the same binaries were used which is fair. However, it is more fair to Intel since the compiler is optimized for the P6 core which issues complex-simple-simple instructions superscalarly. The K7 issues three symmetrically complex instructions and so could achieve even higher advantage if the compiler wasn't trying hard to avoid these complex-complex-complex triplets. Maybe K7 would look better relative to Xeon if some other compiler were used. Even better if there were a K7 optimized compiler.