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 : Advanced Micro Devices - Moderated (AMD) -- Ignore unavailable to you. Want to Upgrade?

To: Elmer who wrote (61172)	10/30/2001 1:48:57 PM
From: Gopher Broke	Read Replies (1) \| Respond to of 275872

What is unclear to me is if this should be considered 1 cpu or 2. They come 2 cpus per module so I don't see how it can be 1. I think you are indirectly preparing the ground for your arguments against SledgeHammer :^) Seriously, the dividing line between UP and MP will have to blur, so the measurements will eventually become one of points/$. (Already what is done for the big system spec benchmarks). Otherwise where do you draw the line? Are two CPUs on a chip with a shared FSB still two processors? What if that sharing extends into the caching? Or into the fetching/decoding? Should a system just with multiple ALUs/FPUs be considered MP? What about SMT - the OS thinks it is two processors?

To: Elmer who wrote (61172)	10/30/2001 1:50:05 PM
From: wanna_bmw	Read Replies (2) \| Respond to of 275872

Elmer, Re: "IBM has just posted SPEC scores for their POWER4." Power4 is a very aggressive design. Power sacrifices had to be made in order to get these very impressive scores. According to Bradley McCredie at MPF, the 1.1GHz Power4 dissipates 150W of power, and that's on an SOI process. I believe those scores are for the 1.3GHz model. I may be too optimistic, but from what I saw and heard at MPF, I bet McKinley can get close to those fp scores at 1GHz and 125W of power. Fortunately for IBM, though, Power4 beats McKinley to market by two quarters. wanna_bmw P.S. SPEC is a single threaded benchmark, so it is accurate to consider the scores as representative of a single CPU core.

To: Elmer who wrote (61172)	10/30/2001 1:56:18 PM
From: Paul Engel	Read Replies (1) \| Respond to of 275872

Elmer - Re: "IBM has just posted SPEC scores for their POWER4. Although not available until December the FP scores are remarkable. SPECfp2000 = 1098 base, 1169 peak. No SPECint2000 scores published. What is unclear to me is if this should be considered 1 cpu or 2. They come 2 cpus per module so I don't see how it can be 1." Whether 1 or 2 CPUs, the results are very impressive. Gee...do you think these results will keep IBM from adopting the Hamster?