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: Brian Hutcheson who wrote (26847)	12/18/1997 12:42:00 AM
From: Yousef	Respond to of 1574212

Brian, Re: "The cost of the die is purely related to how many dies you can make of a wafer ..." Once again Brian, you are making a pronouncement that is NOT correct ... the cost of a die depends on more than just "how many dies you can make of a wafer". There is something about volume and wafer cost that I also believe is a factor. Whether you want to admit it or not, Intel is the low cost producer of microprocessors in the world. This is due to their high yield and high volume !! BTW Brian, please try not to continue to "burn yourself" with that soldering iron!! Make It So, Yousef

To: Brian Hutcheson who wrote (26847)	12/18/1997 12:45:00 AM
From: greg nus	Read Replies (1) \| Respond to of 1574212

Brain is that on a 200mm wafer. How many for a 300mm wafer?

To: Brian Hutcheson who wrote (26847)	12/18/1997 2:14:00 AM
From: Petz	Read Replies (2) \| Respond to of 1574212

Brian, have you noticed that everytime I mention die size, the Intelosers get very defensive. Whether the Intel Tillamook is 80 mm^2 or 95 mm^2 is pretty insignificant when the AMD equivalent is 68 mm^2 and faster besides. This is a big switch from '97 when the K6 die size was much bigger than the Intel P55C (162 mm^2 vs. 120 mm^2). BTW, can you guess how many AMD Fab 25's it would take to produce 100M K6's, with 60% of them K6-3D's, assuming 50% yield using the 0.25 process? 3000 wafers/week * (31400 area/wafer) / (81 mm area/K6-3D) 0.9 0.5 yield * 52 weeks/year = 27.2M K6-3D CPU's + 2000 wafers/week * (31400 area/wafer) / (68 mm area/K6) 0.95 0.5 yield * 52 weeks/year = 22.8M K6 CPU's So one Fab 25 at 50% yield could make 50 million CPU's. I think thats quite impressive, considering the entire x86 CPU market for '98 is only 100 million! Petz