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: fastpathguru who wrote (225605)	2/8/2007 5:28:59 PM
From: Tenchusatsu	Read Replies (1) \| Respond to of 275872

FPG, > You've got it backwards, IMHO Nope, actually you're the one who got a lot of things backwards. First of all, 1S systems are more sensitive to latency than 4S systems. In 4S systems, it's all about the bandwidth. Opteron achieves it by virtually combining the bandwidth of all four memory controllers via HyperTransport. Xeon does it by relying on the north bridge to provide the bandwidth. Neither approach is really superior as long as the bandwidth is available. (Power, on the other hand, is a different story.) Secondly, AMD needs to make its L3 a victim cache because the size of it is "merely" 2MB. That's the same as the sum of the L2 caches, so it's obvious that a traditional L3 cache would have been useless. In other words, the necessity of the victim cache approach is driven by AMD's inability to match Intel's manufacturing strength. And third, it's not Intel who needs to catch AMD at the moment. It's AMD who needs to catch up to Intel. Don't count your chickens before they are hatched. Tenchusatsu