Re: Limits and Capabilities.

From: James Rogers (jamesr@best.com)
Date: Thu Jun 13 2002 - 13:51:45 MDT


On Thu, 2002-06-13 at 07:38, Eugen Leitl wrote:
In practice, even with reversible logic you will run into power
> dissipation issues way before. (Of course, fractal cooling channels with a
> high flux of coolant dilute your circuitry, increasing average distances).

For superior cooling, you might want to consider building your logic on
a porous substrate and immersing the whole system in superfluid He II.
It is very inert, offers extremely high thermal conductivity, and the
phase change from superfluid to fluid creates a very efficient pump. The
limit placed on conventional coolants in this type of environment is how
fast and efficiently you can move coolant through channels of varying
sizes, something which becomes far less of a concern with superfluids.

I don't know if this is practical in all cases, but over the last few
years there has been some interest and study of the superfluid immersion
cooling using porous insulation. The military is interested in it right
now, but eventually I could see it having application in logic cooling.
If I'm not mistaken, superfluid cooling technologies very substantially
extend the thermal dissipation capabilities of conventional cooling
technologies.

-James Rogers
 jamesr@best.com



This archive was generated by hypermail 2.1.5 : Wed Jul 17 2013 - 04:00:39 MDT