SI
SI
discoversearch

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: graphicsguru who wrote (253056)6/8/2008 11:19:56 PM
From: wbmwRead Replies (1) | Respond to of 275872
 
Re: Of course, one can create code that will run slower on Nehalem than
Penryn. But that's not the question. Can you point to any example of
real code doing something useful that you're confident will run slower
per clock? That would be a substantive prediction.

You've been asked for this, and haven't replied.

The easy way to tell an expert from a charlatan is to ask them to go on record for their claims, and then answer for them later.

There's a reason why you haven't heard back from your request.



To: graphicsguru who wrote (253056)6/9/2008 3:58:44 AM
From: mas_Respond to of 275872
 
Have a look at slides 15,16 of this ppt which show the cpi spent in L1,L2,L3 and memory for TPC-C, SAP, SPECJBB and you can see that L2 is negligible compared to L1,L3 and memory.

eecg.toronto.edu

Reducing the cycle times for L1,L3 and memory will benefit more than reducing cycle time for L2. Nehalem has only done one out of those three ;-).


HomeMailHotSubjectMarksPeopleMarksKeepersSettings
Terms Of UseContact UsCopyright/IP PolicyPrivacy PolicyAbout UsFAQAdvertise on SI
© 2026 Knight Sac Media.  Data provided by Twelve Data, Alpha Vantage, and CityFALCON News