Advanced Micro Devices - Moderated (AMD) Message Board

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)

AMD 192.49

-4.0%

Feb 5 3:59 PM EST

Public Reply Prvt Reply Mark as Last Read File

Previous 10 Next 10 Previous Next

To: Elmer Phud who wrote (254811)	7/25/2008 9:28:36 PM
From: pgerassi	Read Replies (1) of 275872

Ephud: Intel's defect densities are scaled to a standard set used to compare all fabs at all companies. That eliminates the effects of DEDP and differences between ultra high performance logic and plain DRAM. When those are factored in again, Intel has lower yields than AMD will at 45nm. And throughput is never figured in any such numbers since its orthogonal to yield. Furthermore the published under (something conveniently ignored by you and others) 0.5d/cm2 figures were from a slide from a 6/07 presentation. By a 12/07 presentation they were below 0.2d/cm2 and still falling. Again both of these are scaled. They do not take into account the higher step counts and things like DEDP. Both AMD's 65nm and Intel's 45nm process are higher pre scaling with Intel's much higher given that it is DEDP on the critical layers while AMD's is SESP on all layers. In the past Intel had higher throughput from any one of its fabs than AMD had in theirs because the cleanroom sizes were much larger at Intel's fabs. And smaller dies sizes only helps to a point. Once the fabs start running at less than full capacity, a smaller die size doesn't help very much as the same number of dies cost nearly the same because of the high fixed costs. That is why Intel may be ok with a lower yielding lower throughput process as long as the smaller dies also get more revenue. It uses up the spare capacity (most think that they have more fab capacity than they currently need). Each individual chip costs a little more, but more than makes it up in extra revenue. AMD does get both the benefits of smaller dies, higher throughput and much more good dies per month (they are at the other side having not enough fab capacity). Having them be worth more too helps as well. Those are good reasons to go with the 45nm 193i scanner based process. It will be interesting to see how this plays out now that they will be at the same process node. Both companies seem to think they will be switching to 32nm sometime around 2010. So they may be at the same node for a while. Pete

Report TOU Violation

Share This Post

Public Reply Prvt Reply Mark as Last Read File

Previous 10 Next 10 Previous Next