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: Elmer who wrote (106940)	4/20/2000 9:30:00 AM
From: pgerassi	Read Replies (1) \| Respond to of 1573887

Dear Elmer: Taking your points one at a time. #1 For yields to be 80-90%, the number of 0.18 micron wafers required to produce 10 Million Coppermines is 36K to 40K. Assuming linear ramp (absolute worst due to Coppermines being shipped in Q4), about 2.5K to 2.8K wafers per week average or 5K wafers per week 1. Total Wafers from all CPU fabs is about 50K to 60K a week. Thus Intel was at 10% conversion max at week 1. They said they were more than twice that at start of Q1. Thus usuable yields must have been less than that. #2 See #1. #3 That does not say how many or how many dies it takes to get one shippable 1 GHz CPU. From accounts, it appears that less than 1000 CPUs were shipped in any given week. #4 They do seem to be somewhat more Coppermines but not showing much incremental capacity. #5 They said the shortage would be over in Q4, then Q1, and now say Q2. The old saying applies, "I will believe them, when I see it". Their creditability is very low. One can not assume that they are telling the WHOLE story. The fact that they did not ship many more CPUs in Q1 than Q4 shows that the ramp is not going well. Somewhere there is a bottleneck and they should know where it is. One Katmai die plus two SRAM die at 0.25 micron are much larger than one Coppermine die at 0.18 micron. To get the same number of CPUs (roughly) from one wafer at 0.18 micron as from one wafer at 0.25 micron, implies that the yields at 0.18 are much lower than the yields at 0.25. At least half as much and probably less. To take more than 3 quarters to fix, they must be doing something terribly wrong. And you crowed when AMD had a breakdown of 1 quarter. They, at least, fixed it in 1 quarter. When I crunch the numbers, I continually show that somewhere Intel is being less than truthful. You say use different numbers. Intel does not provide meaningful numbers. THey only do when the numbers are good. So, when they do not provide numbers, most take it that the numbers will show something wrong. If Intel is capacity strained, they should at least provide how many PIII's, Celerons, Xeons, and Coppermines they sold in Q1. AMD does. Pete