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: kapkan4u who wrote (65212)12/8/2001 1:36:27 AM
From: Dan3Read Replies (4) | Respond to of 275872
 
Re: So 1.4GHz P4 performs no faster than 175MHz PIII in this test.

That factor of almost 7.5 to one was not expected, right? Some of the earlier results you posted were less than 2 to 1 - any thoughts on what in P4 is taking so long? Does the trace cache take longer to process instructions than a straight-through decoder?

The trace cache was introduced on the coppermine version of PIII, to the 933mhz PIII had it, while the 450 didn't, for what ever that's worth.



To: kapkan4u who wrote (65212)12/8/2001 2:45:36 AM
From: Joe NYCRead Replies (3) | Respond to of 275872
 
Kap,

One problem I see with this is that if you assume that somehow the decoder is full speed, and can decode 1 istruction per clock cycle and generate 4uops per clock cycle, while the trace cache can feed (= accept?) 6 uops every other cycle, or 3 per cycle. I think the best instruction would be one that is decoded to no more than 3 uops.

Joe


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