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: TimF who wrote (229074)	3/28/2007 10:32:50 AM
From: Sarmad Y. Hermiz	Respond to of 275872

I agree with the idea that tweak ( whether by continuous adjustment or by copy exact from somewhere that is working) is not the answer to random variations. The answer is to accommodate the variations in the design by making it tolerant. The specific issue that I think AMD ran into with Brisbane shrink is that the design was originally for 90 nm (maybe inherited from 130 nm), and so it did not need to account for the degree of variation that is encountered at 65 nm. Since I have gone so far out of my depth, let me venture a little farther (to drowning level). I also think that achieving effective strain is more susceptible to randomness at 65 nm. Hence the low clocks of the shrunk parts. And now to completely expose myself as a know-nothing. I am convinced that APM is not nearly as effective at 65 nm as at larger geometries. And the specific reason is random vs systematic variation. The proof (or debunking) of this hare-brained idea will be seen when and if AMD produces faster Brisbane processors as Pete has estimated, and as seems to be the faith and orthodoxy of the people who like to see good things written about AMD here.