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.

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

To: vince doran who wrote (89861)	10/12/1999 10:34:00 AM
From: Process Boy	Respond to of 186894

vince - <Of course I expect CUmine to be faster! What I want to TRULY know is how much faster, and the benchmark alteration prevents me from knowing that....What I want to TRULY know is how much faster, and the benchmark alteration prevents me from knowing that. > I submit the metric you are looking for is irrelevant, except if you are staging an engineering "who wins" contest. I also submit the question : Why should Intel spell it out for the competition? Or you, for that matter. The benchmarks will be what they will be on the Cumine, with any improvement over Katmai as a combination of design enhancements and prefetch optimizations. I'm sure there will be plenty of independent party reviews of Cumine systems with plenty of benchmarking results. Probably some with K7 and Cumine systems side by side, compared clock for clock. Nothing shady about the way Intel is presenting Cumine, IMO. PB

To: vince doran who wrote (89861)	10/12/1999 12:57:00 PM
From: Tenchusatsu	Read Replies (2) \| Respond to of 186894

Vince, even if Intel showed all four scores (Katmai with and without optimizations, Coppermine with and without optimizations), some people will still complain and say that SPECint and SPECfp aren't representative of real-world applications. Even Prof. Hennessy thinks that DOOMmarks are going to be more important than SPECmarks. (Just ask his son.) So no matter what Intel reveals, we'll still have to wait for the independent web sites to reveal just how well Coppermine does on "real-world applications" like Quake 3 or Half-Life. We do know two things, though. One, SSE is becoming increasingly popular. And two, SSE cache-streaming enhancements can easily be added by a compiler with the click of a mouse. AMD can duplicate the second benefit with the new Athlon instructions, but without the first benefit, getting a critical mass of developers to support Athlon's extensions is going to be tough. Tenchusatsu