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: wanna_bmw who wrote (50796)	8/12/2001 1:10:24 AM
From: Bill Jackson	Read Replies (1) \| Respond to of 275872

wanna, This is all about incremental gains and that is done by things like Si28, SOI, notched gates, and other numerour process tweaks that add to speed, yield etc. There are a sea of variable and for Intel to dismiss SOI so soon seems imprudent to me. Now IBM is certainly ahead of Intel in process tech and so I go with them. Bill

To: wanna_bmw who wrote (50796)	8/12/2001 9:42:45 AM
From: combjelly	Read Replies (1) \| Respond to of 275872

"Intel did extensive testing of SOI, and they felt they could do better without it." Intel blew their credibility on this with their pronouncements on copper at 0.18 micron. They repeatedly stated that copper offered no significant benefits at 0.18 micron. I think that AMD has proved pretty conclusively that wasn't true. Currently the fastest Al Athlons might be 1.13GHz and may only be 1GHz, the fastest Cu Athlons are at 1.4GHz and might reach as high as 1.5 or 1.6GHz. So either Cu gives a rather significant advantage at 0.18 micron, AMD has really poor process control with Al but does a lot better with Cu, or AMD has decided to cripple the production from Fab25 for some strange, inexplicable reason. Or maybe their definition of "significant" is something like "more than 80% improvement"... I hope you don't trot out Paul's argument that the P4 clock rates prove that Intel doesn't need Cu at 0.18 micron. He conveniently ignores the fact that the P4 has twice (or more, depending on what is being done) the number of pipeline stages of the Athlon. The fact that AMD has on their roadmap a processor that can clock almost as high as the P4 with half the number of stages should cause some very red faces. It doesn't, but it should. The brutal fact of the matter is that if SOI is only used for it's potential to lower the power requirements, the Barton should run pretty close to the P4 in absolute clock rate. If the Hammers have only two additional pipeline stages, then the Hammers should run faster. And both will beat up on the P4s as far as absolute computational power. Now the joker in the deck is Jackson technology, which Intel shows every sign of having trouble with. But unless it can increase the computational throughput by at least 50%, Intel will be behind next year.