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: Dan3 who wrote (33614)	3/27/2001 9:46:24 AM
From: dale_laroy	Read Replies (3) \| Respond to of 275872

"Availability of Intel's .13 will be limited for most of the first year of production. The one part they have that is somewhat competitive even on .18 is P4." Intel's 0.13-micron capacity will not be able to significantly impact AMD until Ireland begins to ramp, which is looking more and more like it won't happen until the second half or 2002. "Were Intel to focus their limited .13 resources on P4, AMD would have a field day selling into the midrange and mobile markets, since its Dresden .18 process and/or its core designs have been more successful than Intel's." Well, they still need to establish a presence in the high end mobile market. But, without 0.13-micron in this market prior to H2 '02, Intel would be in serious trouble. "I write and/or because Austin vs. Intel is a battle Intel would win, while Dresden vs. Intel has been a loser for Intel, so far. Some sources claim Intel's .18 is more like .15 if compared to output from the older AMD lines at Austin - in which case a lot of credit has to go to the core, which has kept Austin competitive with PIII." Currently, Austin could produce higher speed grade Spitfire cores than Intel can produce Coppermine cores. With the AXIA revision, this would improve even more. Then we have Morgan. I believe that, while Fab25 will not scale nearly as well as Fab30 at 0.13-micron, AMD will eventually be able to push Appaloosa to 1.4 GHz. If Tualatin does not push beyond 1.4 GHz, even with copper interconnects at 0.13-micron, Austin will remain competitive as long as the P-III maintains a major market presence. "But it's pretty clear that the main success for AMD has come from the advanced copper .18 process at Dresden (which Intel swore was irrelevant for .18, just as they are now claiming SOI is irrelevant)." Well, if Intel had devoted resources to the conversion to copper interconnects, they would not have had nearly the resources available for tweaking their aluminum interconnects. We have no idea how much AMD could have squeezed out of 0.18-micron aluminum if they were not diverting resources to optimizing Fab30. "If Intel were to suddenly have vast amounts of .13 production running with copper interconnect, it could put a lot of pressure on AMD until the .13/copper/SOI AMD process is on line. How likely, do you think, is such a scenario? How likely is it that some aspect of implementing copper with those high aspect ratio traces will prove troublesome?" Yes, assuming similar percentage yield at 0.13-micron versus 0.18-micron, Intel could produce nearly as many P4 processors at 0.13-micron as P-III processors at 0.18-micron, even without the move to 12" wafers. Even if Intel could only convert over to 0.13-micron as quickly as AMD, AMD could be in deep doo-doo. By being able to convert over to SOI by the time Ireland fully ramps, AMD can probably neutralize this threat.

To: Dan3 who wrote (33614)	3/27/2001 12:01:30 PM
From: Pravin Kamdar	Read Replies (2) \| Respond to of 275872

Dan, but it's pretty clear that the main success for AMD has come from the advanced copper .18 process at Dresden (which Intel swore was irrelevant for .18, just as they are now claiming SOI is irrelevant). I have the sneaky suspicion that a lot of the success has come from the 0.10u drawn gate length that they are using in their 0.18u copper process. The plan to only shrink this to 0.09u with their 0.13u process worries me a bit. But, AMD's process engineers have done a great job the last couple of years, so I will give them the benefit of the doubt (as long as they don't start running up against resolution issues with the existing Dresden equipment -- I know that it is supposed to be able to support 0.13u and perhaps a bit lower, but it could be the limiting factor in not going to the 0.08u gate length that Intel plans). Pravin.