SI
SI
discoversearch

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: kash johal who wrote (26902)12/18/1997 7:04:00 PM
From: Petz  Read Replies (1) | Respond to of 1573902
 
Kash: Brian's formula 140/380*$70 was based on a $70 (incremental) cost of production from Business Week, but anyhow, Brian's estimate ($25) and yours ($20) are pretty close. I've seen $2000 cost per wafer used by Intel addicts. Even using $2500 cost per wafer to account for extra processing steps in AMD process + $10 packaging cost you get 10 + (2500/(0.50 yield * 380) = $23.

I doubt very much Intel can get > 70% yield from their 0.25 process in '98. Once AMD gets to two-thirds Intel's yield, they will be the low cost producer and further price cuts wouldn't make any sense.

Petz



To: kash johal who wrote (26902)12/18/1997 7:07:00 PM
From: FJB  Read Replies (1) | Respond to of 1573902
 
Kash,

I think the cost per 0.25/200mm wafer is more in the $2500 range. Of course this varies greatly as a function of volume. Maybe, somebody has a better estimate out there?

Bob



To: kash johal who wrote (26902)12/18/1997 7:36:00 PM
From: Brian Hutcheson  Read Replies (1) | Respond to of 1573902
 
Kash , re. math
it was a simple approach to get the big picture . If estimated cost of K6 on .35 was $70 (according to analysts) , then ignoring fixed costs which become less of a factor with volume production . The wafer costs are the same for .25 then the costs are proportional to the different yield between the 2 processes . That is why I used 140/380 .
BTW the total cost of 140 wafers at $70 is $9800 , divide that by 380
and you arrive at $25.78 which of course is the theoretical maximum yield . Of course someone (read Intel police) will pick holes in that
calculation , but it does give an idea of sorts .



To: kash johal who wrote (26902)12/18/1997 11:33:00 PM
From: Paul Engel  Read Replies (1) | Respond to of 1573902
 
Kash - Re: "The manufacturing cost of an 8 inch wafer in 0.25 micron is about $800-1000."

I think $2000 to $2500 would be closer to the real cost of a processing an 8 inch wafer on a 0.25 micron process.

Paul



To: kash johal who wrote (26902)12/19/1997 11:59:00 AM
From: greg nus  Respond to of 1573902
 
Kash, I think your right. To project out furtherLow processor prices of around $50.00 is also consisent with maoving the fully functional sub $1,000.00 market down to $500.00 as I have perdicted in a post to Paul Engle. That is one of the necessary requirements for the rest of the country (60% of the population) unwired to act to get wired. Also the beauty of the $500.00 system senerio occuring takes all the manufacturing production overcapacity out of the system even Dram.