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)
AMD 215.11+0.1%Dec 24 12:59 PM EST
 Public ReplyPrvt ReplyMark as Last ReadFilePrevious 10Next 10PreviousNext  
To: Scumbria who wrote (5824)8/21/2000 3:08:50 AM
From: Charles RRead Replies (1) of 275872
 
<I wouldn't think that it would make a huge difference (<10%). The small L1 data cache however is a serious
problem>

<10% would probabl be right for normal code but I was guessing that it would be a significantly greater number for SpecFP and any code that has lot of FP or any code that uses prefetch extensively.

I agree with you on the small L1 being bad but given L2 is not that far on the pipeline the penalty may not be that much for a well optimized code (read benchmarks). Trace cache misses and Branch mispredicts in real life could kill it though.
Report TOU ViolationShare This Post
 Public ReplyPrvt ReplyMark as Last ReadFilePrevious 10Next 10PreviousNext  

HomeMailHotSubjectMarksPeopleMarksKeepersSettings
Terms Of UseContact UsCopyright/IP PolicyPrivacy PolicyAbout UsFAQAdvertise on SI
© 2025 Knight Sac Media.  Data provided by Twelve Data, Alpha Vantage, and CityFALCON News