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: Petz who wrote (19103)	11/14/2000 7:54:30 PM
From: Pravin Kamdar	Read Replies (1) \| Respond to of 275872

John, I noticed the same thing when I first looked at the numbers. It sure would be nice to have one more data point (1.3 or 1.6 Ghz). Thanks, Pravin.

To: Petz who wrote (19103)	11/14/2000 8:31:09 PM
From: rsi_boy	Read Replies (2) \| Respond to of 275872

are those numbers really so bad? I actually thought they were kinda okay considering all the other bottlenecks at work. the real question is How do they compare to Athlon scaling from 1.1 to 1.2? I seem to remember the tbird scaling poorly past 1 GHz especially with the via platform.of course that's what DDR is supposed to fix. Still the P4 scaling numbers are meaningless until we can compare them to the slope of the scaling curve for AMD's high-end offerings. Only then do we know if a performance Delta between the two architectures will narrow or widen with time. t.

To: Petz who wrote (19103)	11/14/2000 9:57:43 PM
From: Joe NYC	Respond to of 275872

Petz, Re: Scalability I don't think it differs a great deal from Athlon. I did some calculations in the past comparing Athlon 1 GHz and 1.1 GHz. You would expect maximum performance increase of 10%. I got numbers mostly between 5 and 6%. You could theoretically get a 10% increase if the code is entirely inside the CPU (L1, L2). Joe

To: Petz who wrote (19103)	11/15/2000 7:33:12 AM
From: fyodor_	Read Replies (1) \| Respond to of 275872

<John: P4 scales horribly> I would actually expect the P4 on Rambus to scale the best. Here are my expectations (and it's quite evident what I base my evaluation on) - from worst to best: PIII on SDR TBird on SDR ... TBird on DDR ... Palomino on DDR P4 on DDR ... P4 on Rambus ------------- The dots are there to indicate a jump in scalability. The reason I rate P4 on Rambus so high is, of course, that the dual Rambus channels offer the highest memory bandwidth. The unknown here is what role the higher latency might play, compared to the P4+DDR solution. If NVIDIA ever actually introduces its 128bit DDR chipset, that would take the cake. -fyo