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 Applied Materials -- Ignore unavailable to you. Want to Upgrade?


To: Math Junkie who wrote (48942)7/9/2001 8:49:06 PM
From: Hayduke  Read Replies (1) | Respond to of 70976
 
Richard, Moore's Law is an approximation that gives a nice straight line, but like most rules of thumb is only valid over a limited range. Within that range subsequent generations of silicon transistors were shrunk by an evolutionary process - the process tools may have changed and become far more sophisticated, but the basic idea was the same - slice the silicon's active region a little thinner and ditto for the insulator. For those conceptually simple changes you were guaranteed higher packing density, faster speeds, and lower power consumption.

But we are entering an era where that's no longer the case in part for the reasons that Katherine and Zeev have already alluded to... the interconnects are now a far bigger part of the picture in deternining circuit speed, thinner insulators no longer insulate because of quantum mechanical tunneling (and that's not to mention the difficulty of producing an insulator that's a few monolayers thick) and so on. INTC technologists (at one of the technical conferences - IEDM I believe) have said that future designs will NOT be denser, faster, AND lower power but must be designed for the tradeoffs between these (and other factors) for the specific application. I doubt that you will find this on their website - it is more the stuff of technical journals.

But I think that it is generally accepted among transistor designers (of whom I am one) that the Moore's Law era is ending at least in the sense of the simple evolutionary progression from one technology node to the next, and the situation is becoming a lot more complicated as we approach significant physical limits.

Hope this helps,
HD