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: Shane Geary who wrote (54224)	4/6/1999 9:32:00 AM
From: Steve Porter	Read Replies (2) \| Respond to of 1570830

Shane, Okay, let me see if I can respond intelligently to a couple of your questions/points. Could a key benefit of going to Cu at 0.18um be the reduction in die size greater than that achieved by a linear shrink? I don't think there would be any direct benefit to the K7 on .18 copper vs. traditional methods, other than if they are really careful they may be able to shave another .1v off of the core voltage. The problems with Cu vs. Al at these relatively large process sizes (I know.. it's bad when you are calling top of the line processes large) usually weigh-in at just about the same as the benefits. On the positive side you have reduced cost, more 'ideal' electrical characteristics and you get the transition to the new material out of the way now. On the negative side you are now working with a process that you know even less about (let's be honest, even best process engineers get nervous when firing up a traditional process), a process where there is relatively little 'guiding' material or information and a process that only 1 or 2 companies have used successfully. I guess what I'm trying to say is that it's not that moving to Cu at .18 won't gain any benefit.. there certainly are some (some are more tangible than others). What I'm trying to say is that moving to Cu while going to the smallest geometry you have worked with yet, in your main fab plant CAN if not done properly be a HUGE disaster. Let me pose a question: Why has IBM introduced Copper on a 0.22um process? Does the PowerPC750 get no benefit? Was it just for the PR? In a word, Yes. I haven't seen any benefit to the PowerPC being on copper. Core voltage has not changed, clock speeds have no greatly increased and die size has not been reduced. Give all that, why did IBM do it? Well a couple of reasons. IBM likes being a 'leader' in process technology, this was just as much an EGO move so that IBM could hang another plaque as it was a business move. Another reason IBM did it is because IBM can afford to do it. If it doesn't work IBM just moves production to another fab while they fix the bugs. I mean just one of IBM's big fabs running at probably 30% can supply the entire world's demand for PPC chips. AMD doesn't have this luxury. Also IBM could afford to do it, because IBM is losing interest (rapidly) in the PowerPC. They don't touch the design of it anymore and it's probably only a matter of time before they stop making anything that resembles what we call the PowerPC. Anyway that's enough blabbering ;-) Regards, Steve