Moore's Law is like driving in fog
IDF Spring 2008 We can see a way ahead, apparently
INTEL'S PAT GELSINGER is always good for quotability and he is very efficient at telling people the maximum amount of stuff in the minimum amount of time.
He is constantly shrinking the die.
But this quote from Pat beats anything we've ever heard from him.
"I compare Moore's Law to driving down the road on a foggy night, how far can you see? Does the road stop after 100 metres? How far can you go?
"That's what it's been like with Moore's Law. We thought there were physical limits and we casually speak about going to 10 nanometres. "We have work going on different transistor structures. Silicon has become scaffolding for the rest of the periodic table. We're putting these other structures into the materials. We see no end in sight and we've had 10 years of visibility for the last 30 years. "
What was also fun was when your own Charlie Demerjian picked up the pieces when another European journalist asked a question which slighly flummoxed Pat. Said Charlie: "I think I can answer that" and went on to answer the question. I looked at Pat on my left, and Charlie on my right, and thought to myself what the heck?
Comments
and?
What was the question then?Coming To END of Moe.
22 Nm is done deal, IBM has process underwar. Yet, thats about it, Theres NO Significant Way to reduce Size WITHOUT Changing Newtons Laws.Its' Not that bad, Seven Foot Screen is BIG. Tranistors should end upward Desktop count at ~4-6 Billion, O.K. Some NUT might want 12 billion, Gasp, Cough. If thats Countable Number.
Only Way to improve things would be to Start ALL Over, From Timex Sinclair Arthemetric unit. Why Bother, when it works its works & I'm Assured that Ultie_Tom invented it.
Stewart Drashek
Mad Mike
Good to see your name on the byline, you haven't completely abandoned the Inq for Hyderabad yet!Mageek is back?
Nice to see you writing here Mike!Please tell.....
Please tell teh question from the journalist and the answer from Charlie!Which brings us to the following:
What was the question and what was the answer.C'mon mike, don't leave us hangin' :(
Good to at least 16nm
'22 Nm is done deal, IBM has process underwar. Yet, thats about it, Theres NO Significant Way to reduce Size WITHOUT Changing Newtons Laws.'I guess you missed this earlier story; http://www.theinquirer.net/gb/inquirer/news/2007/12/13/nanometre-memory-tested - Toshiba have already built memory cells using a 15nm feature with.This is near the actual lower limit on transistor size (from leakage via quantum tunelling), which is why the 16nm node is the last one seriously defined in ITRS. It'll take some effort to get it into mass production but it'll come eventually. Below that limit we'll almost certainly need fairly radical technology changes. Note however that stacked dies is a technically viable technology for further increasing effective transistor density without requiring any lithography improvements; it's just waiting for commercial viability.
Question was obviously...
"Where's the bar?"