From: scerir (scerir@libero.it)
Date: Sun May 19 2002 - 14:07:41 MDT
Hal Finney
> By 2012 gate lengths should be 10 nm and that's
> just not going to work with the current flavor of
> semiconductor technologies. We have to have
> something fundamentally new by then.
'Indeed, Moore's Law, interpreted as predicting the
number of electronic components on a computing chip,
predicts that the average number of elementary objects
participating in a computation approaches unity around
the year 2012. This number implies that single electrons
or single atoms would be involved in such computations,
the behavior of which, however, need to be described by
quantum mechanics, and are subject to the uncertainty principle.
Uncertainty in computation, it is perhaps superfluous to say,
is not welcome. A standard examination of this uncertainty
shows that it is due to the uncontrollability of quantum
degrees of freedom that interact with their environment,
as well as to the general concept of 'vacuum fluctuations'.
These factors appear to spell the end of Moore's Law.'
More at http://arxiv.org/abs/quant-ph/0202039
a good simple paper by Adami and Dowling
s.
This archive was generated by hypermail 2.1.5 : Sat Nov 02 2002 - 09:14:12 MST