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.

Politics : Formerly About Advanced Micro Devices -- Ignore unavailable to you. Want to Upgrade?

To: Process Boy who wrote (68430)	8/11/1999 2:34:00 AM
From: THE WATSONYOUTH	Read Replies (1) \| Respond to of 1579785

kash - <But with Cu they may be able to get a much bigger increase in speed while keeping power reasonable.> PB <This is the $64k question, referring to the operative may. It will be interesting to see the performance benefit from Cu at .18.> It depends entirely on how it is done. I don't believe AMD is using low k dielectric along with the Cu at .18um (can anyone confirm?) If so, the wiring capacitance will increase (thinner intralevel and possibly thinner interlevel dielectrics). Also I think AMD uses a nitride etch stop at both the via and line levels. So - even more capacitance. Intel, on the other hand, uses a low k dielectric and purposely uses high aspect ratio line levels. This results in lower intralevel and interlevel capacitance as well a lower line level sheet resistance (because of thicker Al lines) This is not easy. But they do it to avoid having to use Cu at .18um. The result will be be no worse than what AMD will achieve from Cu BEOL if their process is as I assume. The AMD folks better not depend on Cu BEOL to tip the scales in this race. Even if AMD combines Cu and low k dielectric, it would at best be a 5% performance advantage based on the different BEOLs. What goes on in the FEOL and the design differences will determine the winner in this race. THE WATSONYOUTH