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: hmaly who wrote (113168)	5/28/2000 11:59:00 PM
From: Joe NYC	Read Replies (1) \| Respond to of 1573307

Harry, The article about using pure silicon had a schedule for mass rollout at the end of 2000, so even if the white paper is true, I don't think the current Tbirds are based on it. Joe

To: hmaly who wrote (113168)	5/29/2000 6:54:00 PM
From: chic_hearne	Read Replies (2) \| Respond to of 1573307

Re: 1) I am talking about the fact that when Chic Hearne started posting here in March he said that the buzz was that the IBM engineers ;who had seen both Willy and Tbird; said that Tbird was a better chip hmaly- There was no specific chip mentioned in the rumor, but it HAD to be the t-bird. It was a pure rumor, take it for what it's worth. 6) Yields at ibm's new copper process plant are rumored to be in the 70% range. It was reported in an EBN article on the new plant several weeks ago. IBM has claimed this over and over at many different times in many different sources. On top of that, they also claim that the copper process is cheaper than the Al process (not sure if this is only due to yields or if there's more to it). The whole copper process/dresdan thing was a big part of why I got into AMD in the first place. I read all of the IBM news on the internal site. They post the links (so I don't have to do any digging) and a few times a week offer perspectives. They really don't post anything that they don't say in public, so it's not like I have any info that you guys don't. Over and over, I keep getting the feeling from IBM that there are serious problems shrinking Al below .18um. I'm sure it can be done, but you must hit a point where yields are so bad it doesn't make sense. This is where copper comes in. The company view seems to be that starting with the S80, IBM has taken the performance lead and doesn't plan on giving it up for some time. There could be some short term leapfrogging by our competitors, but with .18um in the fall, and POWER4 by next year, we will be way ahead in terms of processing power. The theme used over and over is copper. You don't just go to copper over night. I know some people working on one of our next generation processors and it's a 3 year project. The processor technology used in the S80 was already 18 months old at the debut. From IBM's experience, I think it's safe to conclude that going to copper is not an easy task that happens over night. AMD being able to sell copper based processors on June 5th is HUGE NEWS. Intel should have it easier because they will be able to learn from AMD's and IBM's past and hopefully avoid some common problems, but it isn't going to happen over night. On the other hand, I think Intel is taking a far more risky strategy by going to copper so late. IBM started copper on .25um and AMD at .18um. I'm no engineer, but it seems to me both companies decided to start with the copper process where there really wouldn't be any performance gains but by starting so earlier they both would be far ahead of the competition in terms of learning experience. For both companies, I expect this "learning experience" will REALLY pay off with the .13um chips. If Intel stumbles with copper, they will have no way to improve performance. They will be stuck in neutral while the copper process stabilizes. Meanwhile, it seems like AMD has pushed Al to the limit while ironing out the bugs in their copper process. Now they are ready to move on to copper. Intel seems to be intent on pushing the theoretical limits of the Al process. Personally, I don't care if AMD's dresdan chips run at 700 Mhz or 1.5 Ghz on debut. What's important is they can produce the chips regardless of speed. It's hard to expect much out of what amounts to major change in the process of manufacturing processors. Just getting some workable parts out the door June 5th says enough to me. The best is yet to come when they've been delivering chips for a year and have had some time to tweak and toy with the process. This is a major technological leapfrog. chic