Advanced Micro Devices - Moderated (AMD) Message Board

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)

AMD 203.14

-0.8%

Jan 9 9:30 AM EST

Public Reply Prvt Reply Mark as Last Read File

Previous 10 Next 10 Previous Next

To: mas_ who wrote (252870)	6/5/2008 2:38:43 PM
From: dougSF30	Read Replies (1) of 275872

It is what the numbers say it is, worse than Penryn/Merom. If you have an app whose working size is 6MB or less the performance will be inferior. Based on this claim, you don't understand pre-fetching at all, mas. Here's a hint: the size of the working set is insufficient data. The access pattern matters a great deal. If the access pattern is sufficiently regular, the data is already available in the closest cache by the time the core wants it, and the latency (from memory, L3, or wherever) is "hidden". Only extremely "random" access patterns too large for the L2, yet not much bigger than the L3 have a chance, and you won't find many (any?) real applications that exhibit this pattern.

Report TOU Violation

Share This Post

Public Reply Prvt Reply Mark as Last Read File

Previous 10 Next 10 Previous Next