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: fyodor_ who wrote (37936)	5/1/2001 12:33:16 AM
From: Jim McMannis	Read Replies (1) \| Respond to of 275872

Rambus news, sorry if previously posted. ebnonline.com

To: fyodor_ who wrote (37936)	5/1/2001 2:10:50 AM
From: dale_laroy	Read Replies (2) \| Respond to of 275872

>Jerry promised that they would be done with the transition by end of next year. He didn't say this in the context of Dresden and he didn't qualify his statement with any "Dresden only" remark. The more I listen to it (the shareholders' meeting), the more I interpret it as meaning that Austin will be converted to 0.13µ as well.< Considering that Appaloosa is unlikely to be produced at Dresden, it seems logical that Fab25 will indeed be converted to 0.13-micron. >I just don't see AMD developing 3 separate processes.< Well, they are already developing two separate processes, one for Dresden and one for Austin, and have been since 1999 when they had a much smaller market share. I don't think three separate processes is much of a stretch. What I look for is: SOI plus low-k dielectrics with copper - About as high of a speed grade improvement over bulk silicon with copper interconnects but without low-k dielectrics as copper is over aluminum. This 0.13-micron process should scale by about 50% versus 0.18-micron bulk silicon with copper interconnects, eventually taking AMD to between 2.7 GHz and 3.0 GHz for Barton. Bulk silicon with copper interconnects and no low-k dielectrics - SOI and low-k dielectrics could be about 20% faster than this process, so Thoroughbred will likely be limited to between 2.2 GHz to 2.5 GHz. Bulk silicon with aluminum interconnects and no low-k dielectrics - Could be about as much slower than bulk silicon with copper interconnects but without low-k dielectrics as aluminum is slower than copper. This could limit the peak speed grade of Appaloosa to anywhere between 1.8 GHz to possibly as high as 2.4 GHz. Note that even at 1.8 GHz peak speed grade, Intel would have nothing to compete against Appaloosa until value segment P4 comes out. But it is very unlikely that Appaloosa would be able to compete against a value segment P4. OTOH, if AMD were to introduce a value segment Hammer using the Austin process, such a processor might be able to compete against a value segment P4 throughout 2003.