To: chic_hearne who wrote (108429 ) 8/26/2000 9:33:10 AM From: semiconeng Read Replies (2) | Respond to Al: Going to .13um on Al is a pipe dream in my opinion. I don't think it can be done without a severe sacrifice to yields. If Intel is willing to go to .13 Al with bad yields, it should be taken as an obvious sign of desperation. If the yields are good, it contradicts all of the research I've done. The answer to whether or not this is possible should be obvious when the time comes. intel has stated several times that the 0.13u process will be Copper, not Aluminum, so this statement is irrelevant. Cu: As far as I know, Intel plans to start Cu at .13um. I think this is a huge mistake on their part. IBM started Cu years ago and it took them 18 months after their chip was qualified before it finally made it into a server. It took 24 months for a full rollout. Considering IBM did this on a .25um process, I find it highly unlikely for Intel to do this flawlessly on .13um. Also, it seemed like it took AMD forever before they finally got their Cu chips shipping. Of course it should be easier for Intel because IBM and AMD have done the learning, but neither has any shipping chips at .13um yet. Who said that intel is using the same Copper process that IBM and AMD is using? You seem to be assuming that the only way to do a copper process is the IBM/AMD/Motorola way. There are other ways to do it. Maybe intel has developed a different way. In my opinion, .13um in mid 2001 is not going to happen like Barrett claims. I'd think that Intel would be trying this in multiple fabs on both processes. My guess is that combined efforts from all fabs using .13um will not total 1 megafab running full bore anytime in 2001, maybe even 2002. A few hundred thousand parts in 2001 sounds more realistic I think (I'm being generous, ZERO is what I really think). Or maybe Craig Barrett has a little better insight into intel's copper process development than you do. Also, it would be foolish to mix Copper and Aluminum processes in the same Fab. The contamination issues would be a nightmare. All speculation on my part. I'm sure the pro-Intel guys will respond by saying I'm full of it, but because none of them actually responded to your post, you can tell it's a touchy subject, or they have absolutely no idea what is in store for .13um in 2001. I don't think that you're full of it, simply misinformed. Maybe nobody responded to Burt's post.... not because nobody knows, but maybe because they're not interested in getting fired for saying.
