Re: Ye olde Moore's Law paper

From: Eugene Leitl (Eugene.Leitl@lrz.uni-muenchen.de)
Date: Sat Jan 12 2002 - 03:30:06 MST


On Sat, 12 Jan 2002, Damien Broderick wrote:

> At a conference in Kyoto, Japan, Intel displayed transistors, or
> circuits, only 70 to 80 atoms wide. This nanometer (a nanometer being
> one-billionth of a meter) technology should lead to low-power chips
> containing 1 billion transistors running at speeds of 20 GHz. (Today's
> fastest Pentium 4 models have 42 million transistors and run at 1.7
> GHz.)

Well, we have transistors which switch at ~3 THz, and will be up to 6 THz
shortly. Given defect rate and clock skew problem, this indicates we might
be getting large scale cellular architectures after all.

> An eyeballing of Moore's graph seems to imply that we'd be at a trillion
> [2^40] by now (but my eyes aren't too good).

It's not about the transistors, it's about performance. And benchmarks
don't hold with Moore's law. Need repeating.



This archive was generated by hypermail 2.1.5 : Fri Nov 01 2002 - 13:37:34 MST