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: Ali Chen who wrote (48486)	7/20/2001 12:23:42 PM
From: eCo	Read Replies (1) \| Respond to of 275872

Ali: The AMD problem is that their current platform is less efficient than the P4 platform. With CPU core frequencies going up, this disparity will impact AMD system performance more and more. nForce should should take care of your worries here. :) eCo

To: Ali Chen who wrote (48486)	7/20/2001 8:25:51 PM
From: Petz	Read Replies (1) \| Respond to of 275872

Ali, what do you mean the AMD platform is "less efficient" than P4 platform? re:IPC increase does not help to improve AMD system performance, it is already 25-80% better than for P4. The AMD problem is that their current platform is less efficient than the P4 platform. With CPU core frequencies going up, this disparity will impact AMD system performance more and more. There are many benchmarks for which a 10% increase in clock rate gives more of a % increase in performance on an Athlon system than the same 10% increase in clock rate gives in a P4 system. I define the "platform efficiciency" of a system as 1. A function of the CPU, the platform (SDR, DDR, chipset, etc) 2. Also a function of MHz level 3. Also a function of the particular benchmark being used 4. Define it as the % increase in benchmark performance divided by the % increase in clock rate My assertion is that PE(TBird,DDR,1200,average benchmark) ~= PE(P4, i850 RDRAM, 1500, average benchmark) Platform efficiency normally declines as MHz goes up. For the P4, there is a quirk that a P4 1700 seems to have higher performance than it should based on the PE at 1500. In any event, the PE of the P4 using SDRAM will drop to zero above 1.5 GHz. So platform efficiency is not AMD's problem in the least, MHz is. IMO, it is essential that AMD exposes the poor performance of P4 on SDRAM. Publications and test reviews are not enough. But, what IF Intel supports DDR sooner, rather than later? One big question is: How much worse will P4/i845-DDR perform than P4/i850? If its not much difference, Intel will push DDR heavily and short-circuit any AMD publicity. I don't think we can count on Intel being stupid forever, but perhaps the earliest DDR/i845 could be ready is ~October. The second SDR P4 hits the streets, we need an AMD is faster campaign. Petz