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: Petz who wrote (80218)	11/17/1999 2:51:00 AM
From: Paul Engel	Respond to of 1571578

Petz - Re: "AMD can play this cheating game too, by implementing a pre-fetch compiler and I suspect they will. Such compilers are useless for real programs because they are specific not just to the CPU being used but also the chipset and memory system. For example, the prefetch instructions for an i820 mobo/600 MHz RDRAM combo could actually harm execution on an overclocked BX chipset or an i840 mobo with two channels of RDRAM" Ah.....excuses, excuses, excuses !!!!! If AMD adds 32 new instructions to its processor, is it PRECLUDED FROM USING these instructions - via new compiler instructions ? Excuses, excuses, excuses !!!! The NOOSE TIGHTENS ! Paul

To: Petz who wrote (80218)	11/17/1999 7:58:00 AM
From: Haim Barad	Respond to of 1571578

Message #80218 from John Petzinger at Nov 17 1999 2:02AM Engel, re:<FlopperMine's high SPECint scores> AMD can play this cheating game too, by implementing a pre-fetch compiler and I suspect they will. Such compilers are useless for real programs because they are specific not just to the CPU being used but also the chipset and memory system. For example, the prefetch instructions for an i820 mobo/600 MHz RDRAM combo could actually harm execution on an overclocked BX chipset or an i840 mobo with two channels of RDRAM. Yet another person who thinks that SW optimizations are cheating. Yet, you also think that AMD will resort to this cheating as well (i.e. create a prefetching compiler). You also said that this is useless. Too many contradictions in one paragraph... To be honest, AMD should create their own compiler for such features. It certainly wouldn't be cheating in my mind... (imagine... taking advantage of a feature that's now part of your ISA... would you call that cheating?) Haim PS: I also don't think that AMD taking advantage of 3DNOW on 3DWinbench and 3DMark is cheating... do you?