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: dougSF30 who wrote (196983)	5/15/2006 11:15:19 PM
From: Joe NYC	Read Replies (2) \| Respond to of 275872

Doug, Latency to main memory is slightly longer for Woodcrest (in the random case-- for patterned accesses, prefetching kicks in for both) BUT the memory disambiguation helps there, basically by starting some memory requests before the K8 would, thus hiding some/all of the extra latency. Starting a memory access a few CPU cycles earlier does help, but memory is slow in comparison to CPU cycles, so it is hard to make up Opteron advantage entirely. I very much doubt that Woodcrest can significantly reduce the gap for random memory accesses. What is in Woodcrest's favor is that these completely random accesses are not that common, and that the average latency (combination of all the memory access) is likely to be lower on more common tasks. Joe