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)
INTC 37.52-0.8%3:59 PM EST

 Public ReplyPrvt ReplyMark as Last ReadFilePrevious 10Next 10PreviousNext  
To: Tenchusatsu who wrote (118699)11/21/2000 8:18:07 AM
From: Dan3  Read Replies (1) of 186894
 
Re: Funny how SPEC became less and less relevant over the past 14 months

SPEC is a very important benchmark. And you are right that it is a good indicator of potential performance on hand optimized code.

But it isn't the only benchmark, and when it comes to the P4, results from SPEC seem to contradict the results of almost every other benchmark. In terms of analysis, it's an outlier, and while statistical outliers aren't necessarily erroneous, they are suspect.

Given that SPEC, more than most other benchmarks, can be "helped" by creating compiler optimizations for it, it should not be considered to the exculsion of all other benchmarks, especially when just about all other benchmarks contradict the results of SPEC.

There is at least one benchmark (Super PI)where the P4 1.5GHZ has a tough time beating a K6-2 500. It's also an outlier, and I give it a lot less credence than SPEC. But in some ways it's as significant an indicator of the strengths and weaknesses of the P4 as is SPEC run from a compiler custom built for the P4.

Dan
Report TOU ViolationShare This Post
 Public ReplyPrvt ReplyMark as Last ReadFilePrevious 10Next 10PreviousNext