Silicon Investor (SI) -- The First Internet Community

STOCKTALK

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.

Politics : Formerly About Advanced Micro Devices -- Ignore unavailable to you. Want to Upgrade?

To: Scumbria who wrote (44335)	12/29/1998 9:34:00 PM
From: Steve Porter	Read Replies (1) \| Respond to of 1584454

Scumbria, I know what spec was designed to test.. now I'm telling you what it does test.. I will dig out one of my Doctor Dobbs Journal's later and quote from an article in there.. compare the difference between the 21164 and the p6 on spec95 and spec92.. the p6 is ALOT closer on spec95.. why, the data set on SPEC 95 is MUCH MUCH larger than the caches... Remember we are talking SPECint here, I'm not so sure about SpecFP.. I believe SpecFp's data set is much more cachable. For example: (right from the site you posted), this is one component of the INT section. SPEC CINT95 Benchmark: 099.go ------------------------------------------------------------------ 099.go Artificial Intelligence; pattern matching, look-ahead, etc. Based upon an internationally ranked 'Go' playing-program called 'The Many Faces of Go' by David Fotland. As a benchmark, this is really just a base test of simple integer performance. Cache activity is small, but does require something larger than an 8MB cache to run well. Regards, Steve