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.

Technology Stocks : AMD:News, Press Releases and Information Only! -- Ignore unavailable to you. Want to Upgrade?

To: Paul Engel who wrote (4122)	2/1/1998 12:52:00 AM
From: Dave	Read Replies (1) \| Respond to of 6843

Paul: I am unsure if anyone has asked you this question, but are AMD's problems due to their Process in Manufacturing the K6, or in the original Design of the K6. Now, given that the K6 is basically a Nx686 with added MMX, pin outs for Socket 7, and the removed Cache controller. Nexgen was to have their chip fabbed by IBM. Do you think IBM would have the same problems as AMD? Thanks, dave

To: Paul Engel who wrote (4122)	2/1/1998 12:54:00 AM
From: Elmer	Read Replies (2) \| Respond to of 6843

Great analysis Paul, as usual. So seeing as they can't produce them at anything approaching production level yields at their development fab, it has to strike terror in every AMD investors heart to think what must be happening at Fab25. Add to this the news from CyberMAX and it's getting really really ugly. In the mean time Intel is +3 speed grades ahead and pulling away. Speed grades +4,+5 and +who knows what else are rumored to be ready to ship. Prices are dropping rapidly in the 233-266mhz arena so AMD will wrap even more hundred dollar bills around their K6's. How long can this go on? If this was a fight they'd stop it or someone would throw in the towel. Jerry though just keeps pushing the shareholders back in there to take another round of beating. It's getting hard to watch... EP

To: Paul Engel who wrote (4122)	2/1/1998 1:03:00 AM
From: Robert Walter	Read Replies (1) \| Respond to of 6843

Paul, RE:500 wafer outs over 8 weeks produces 4000 wafers. IF 10,000 good die were ultimately shipped, AMD must be running 3 1/8 net good die per wafer in their DEVELOPMENT FAB. IF I remember correctly I read a brokerage report just after the earnings conferences call that stated yields at the SMDC were in the 30-40% range on 0.25. Ashok Kumar comments on the 1% yeilds were in reference to Fab 25 not SMDC. Since only AMD knows the real number of wafer starts and yeild at SMDC it is rather senceless to guess with these projections. According to the report the problem lies in that the SMDC has different equipment than Fab 25. AMD was expecting even higher yeilds at Fab 25 because all of the 0.25 wafer starts being done there are on brand new equipment which is more modern than the equipment at SMDC. SMDC obviously had to purchase there equipment back some time in 1996 or earlier for its 0.25 wafer production. AMD has been producing 0.25 wafers at SMDC since Febuary/March 1997. I posted a URL relating this last year in late Febuary or early March if you want to go look for it. Robert PS, Here is the exact qoute and the URL. Good idea, but according to Loewenbaum & Co. chip analyst Ashok Kumar, the current yield on the 0.25 micron at Fab 25 is an alarmingly low 1%. forbes.com

To: Paul Engel who wrote (4122)	2/1/1998 9:03:00 AM
From: Kevin K. Spurway	Read Replies (1) \| Respond to of 6843

Paul, if you read my post carefully, you'd notice that I'm agreeing with you on the fundamentals here--it looks like AMD has a .25u yield problem. Yes, I've run the numbers on SDC .25u production, and I've come to the conclusion that I can't conclude anything, because I have no idea how many wafer starts have been dedicated to K6 since they started working on the process back in spring of '97. Tens of thousands was my mistake, in fact it was a repetition of what Stockman said in the post I was responding to: exchange2000.com As far as certainty about contractual obligations, obviously I have none. I am merely speculating that Compaq and IBM will have made absolutely sure, via contract, that they will receive X number of chips, and I'm speculating that AMD is willing to make them that guarantee. The evidence seems to reflect this in both cases: rumors of a deal far in advance of an announcement, larger deal sizes, and, in the case of Compaq, an unfortunate prior history of supply problems with AMD. I think in the case of CyberMax, it is considerably less likely that 1)CyberMax felt the need to guarantee themselves a sure supply and 2) AMD was willing to grant them a contractual guarantee. Maybe you can enlighten us as to what kind of contract is industry standard. I'm guessing that semi manufacturers don't guarantee supply, except in rare cases. The evidence suggests to me that the Compaq deal IS one of these cases. Sometimes selective deductions based on independent facts get you someplace. Like I said, I think the whole CyberMax thing reflects very poorly on AMD. It's bad business for one thing, and it's certainly very suggestive of continuing yield problems. I just find it unbelievable that you accuse me of "selective deductions based on facts that you would prefer to be in existance, independent if they even exist." and then have the nerve to engage in complete speculation about AMD's .25u yields, literally conjuring numbers out of thin air! Absurd! Typical! Personally, I believe that sometimes speculation can be a valuable tool. If you disagree with me, at least have the dignity to practice what you preach. Kevin