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 221.02

+6.4%

3:59 PM EST

Public Reply Prvt Reply Mark as Last Read File

Previous 10 Next 10 Previous Next

From: dougSF30	6/5/2008 8:47:46 PM
	Read Replies (1) of 275872

Debunking mas' claims regarding the impact of any purported increase in Nehalem's L1 latency over Penryn: siliconinvestor.com You better have a look at table 3 here mas: courses.ece.uiuc.edu The performance loss of increasing L1 latency by one cycle (from 2 to 3) in the original P4 uarch with all its nasty long replay loops is 0.72% for SPECint95 and 2.56% for SPECint2k. The C2D family have far less brittle scheduling behavior than P4, a property carried into Nehalem (as disclosed at IDF). If the Nehalem's L1 latency really has gone up by one cycle vs NGMA (that would be 3 to 4 BTW, not 2 to 3) then the real performance impact for most apps is likely on the order of magnitude less than your "5-10%". Follow the thread. More schooling here: siliconinvestor.com siliconinvestor.com As I said earlier, another "insufficient data" claim from mas, another win for common sense.

Report TOU Violation

Share This Post

Public Reply Prvt Reply Mark as Last Read File

Previous 10 Next 10 Previous Next