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: pgerassi who wrote (93411)	2/20/2003 4:44:50 PM
From: Charles Gryba	Read Replies (1) \| Respond to of 275872

Pete, you missed my point. The 2 companies diverged when Intc introduced the P4 in lieu of further refinement of the P3. Intel went for new cutting edge architecture stuff in the P4 but stuck with the safe process. AMD went with safe architecture ( palomino->TBred->Hammer ) but a bit more adventurous on the process side copper 1st, more layers, SOI now, etc. Intel is getting ready to move up their process side to strained silicon and at the same time they are improving the P4 core to become Prescott. AMD on the other hand is still putzing around with basically with the same core and the same process. All I am saying is that Intel seems to be executing on both sides now but initially concentrated on the architecture side so while the early P4s were dogs their architecture is maturing to the point of dominance with Prescott. Unless AMD has a trump card on the process side (or the architecture side ) they will be falling behind. C

To: pgerassi who wrote (93411)	2/20/2003 5:12:05 PM
From: fyodor_	Read Replies (1) \| Respond to of 275872

Pete: How could merely doubling the cache and doing a few little things make it (Prescott) so much newer? Any (positive) change Intel makes to their processor core can be used to great advantage, because they have both the clout and the resources to affect changes in benchmarks. Anything Intel has that AMD doesn't will be used - you can count on it. Whether this be larger cache, greater bandwidth, special instructions… anything that can be used, will be used. Oh, and don't think there aren't a lot of improvements to be made to the P4 core. There are still plenty of cases where it is decode bound and HyperThreading can still be greatly improved as well. Doubling the trace cache and the decode resources (and possibly L1 data cache) would increase performance in a wide range of applications with minimal increase in die space. Not at all unlikely for a reworked core at 90nm, IMHO. -fyo