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 : Intel Corporation (INTC) -- Ignore unavailable to you. Want to Upgrade?

To: kapkan4u who wrote (150385)11/29/2001 3:40:18 PM
From: wanna_bmw  Read Replies (2) | Respond to of 186894
 
Kapkan, type the whole thing up and send it to me, and I'll find a Pentium 4 PC to run it on.

However, I can already guarantee you're wrong about performance. Even a lowly 2-bit smith algorithm can predict the kind of branches that you're using 99.9999% of the time. After (at most) two consecutive branches, your code has a chain of Strongly Not-Taken branches, and these will be easily predicted by the Pentium 4 Static Predictor. The Dynamic Predictor is even smarter than this, so perhaps the success rate would be even higher than I am predicting. I bet you on sheer clock speed alone, as well as the double-pumped ALUs, that your code would run faster on a 2.0GHz Pentium 4 than on the fastest Athlon available. :-)

wbmw



To: kapkan4u who wrote (150385)11/29/2001 3:55:45 PM
From: Charles Gryba  Read Replies (1) | Respond to of 186894
 
Kap, I wish you are right but no one would ever code that. They would make an array of 1000000 and they would just increment directly into the array without using this case statement.

C

p.s. Real business code runs 50% on the P4 than it does on the Tualatin.


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