R.L. said:
>While your simple scaling of fab plant cost is bogus at
>best, I will admit it will be VERY costly-
I'll admit that my naive scaling law was largely bogus but there is a fab plant costing extrapolation called "Rock's Law".
First refer to the following URL about the problem of fab plant expense:
http://news.com.com/A+fab+construction+job/2100-1001_3-981060.html
The important quote was:
"In the end, the final cost came to over $2.5 billion. By 2007, the price of building a fab is expected to reach $6 billion."
The company paying the $6 billion is IBM. Surprisingly, IBM's pockets are not as deep as Intel's, i.e. IBM has $ 8.66 billion in total cash versus $ 14.82 billion for Intel. IBM will be betting the company on this new fab plant (one of the reasons why IBM's stock has not been all that terrific as an investment).
Now link to the Wikipedia article on Rock's Law:
http://en.wikipedia.org/wiki/Rock's_Law
The key quote here is:
"a semiconductor chip fabrication plant doubles every four years"
Based on Rock's Law and the $6 billion price tag for the 2007 IBM fab plant, a hypothetical fab plant in 2011 will cost $ 24 billion. That's more than the total cash of IBM and Intel combined. It's clear that sometime between now and 2011, Rock's Law will be an impassable barrier.
Gary
Geek Speak » in reply to :) I was involved in the project
Fab plants and Rock's Law
The whole thread (40 posts)
- Moore's Law has finally failed ?
- Most computers today are limited by DISK SPEED
- Maybe not!
- Intel press release on failure to increase processor speed.
- I think the next big push will be multi-processor chips.
- Why any faster?
- Diminishing returns
- re: Moore's Law is not a continuous curve
- It's more of a guideline... (n/t)
- Not the physics, but the markets
