To: kapkan4u who wrote (65061 ) 7/13/1999 12:25:00 PM From: Charles R Read Replies (2) | Respond to of 1580109
Kap, <<Chuck - re: AMD faces a capacity mismatch (between fab and test/assembly) if they turn all their production to K7. > Chuck, Can you elaborate on the above? Also, what is your assessment of Saltich (Fab30 mgr) resignation?> First Saltich: Not a great news for AMD but I don't read anything too sinister - the guy moved on to become a CEO. The disconcerting part for me is that there have been mixed rumors on the yields and speeds at Dresden. I am looking forward to tomorrows conference call to get more info on that one. About mismatch: Capacity planning is big thing in this business. A company has to forecast resonably well the volumes and product mix of each product line it sells. Let's say AMD did their capacity planning based on selling 5Mu of K6s in Q4 and 1Mu of K7. They would then go around putting the appropriate amount of people, tester, packages, assembly/test capacity, fab capacity etc. Assume that they have already put a lot of this in place and then find that K6s are not making much profit and want to shift production to K7. The fab capacity is already nailed down and, say, they figue out that they can manage another 1Mu of K7 on the fab side. Now they have the 5Mu of excess assembly/test capacity (and the people to go with it) that was put in place for K6s and no good way of utilizing it. On the K7 side, they need to make sure that they get additional 1 Mu worth of SRAMs, chipsets, assembly capacity at SCI, etc. Some of the resources dedicated for K6 can be used to fill the K& needs but it is problematic. Some things they will have more than they need and some other things they will have less than they need. Does that help? Chuck