Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Superconductivity and it's relationship to Chess Computers?

Author: Pete R.

Date: 16:52:20 12/02/99

Go up one level in this thread


On December 01, 1999 at 22:07:35, Tom Amburn wrote:

>
> A few years ago  I was reading somewhere that scientist were trying to discover
>Superconductivity, it said the benifits of this technology would lead to a
>thousand times increase in computer speed. I am not sure if I have all the facts
>correctly, as the article is hazy in the back on my memory. Can someone explain
>exactly what this superconductivity is and how it would benifit computer chess
>if ever discovered?  thanks

Conductors resist the flow of electrons through them, which generates heat,
which is a major problem when you are trying to fit millions of transistors into
smaller and smaller spaces. Superconductivity is a state where this resistance
goes to zero.  Without any resistance you could conduct the power for an entire
city through one wire, which is incredible if you think about it.  You can also
transmit over long distances without loss, etc., all of which would have amazing
practical benefits, except so far it only works at very cold temperatures.  The
phenomenon has been known for decades, and in fact my freshman year physics
professor shared a Nobel prize for the first explanation of why it works (Robert
Schrieffer I believe, I think Bardeen, Cooper, and Schrieffer shared the prize
on that one).  No experiments, I think he said they just sat around and drank
lots of coffee while mulling it over. ;)  Anyway the quest now is to find
materials that achieve this state at higher and higher temperatures, the Holy
Grail of course being room-temperature superconductivity.  I tend to think this
will have less impact on computing than continued advances in miniaturization
and materials.  A single-molecule transistor has been demonstrated, and I think
a carbon nanotube-as-transistor has as well.  If computer power keeps steadily
increasing as it has, or even accelerates, it will be quite interesting. The
current issue of Scientific American has an interesting article (one of many in
their Millenium issue) where a robotics researcher predicts computers with
sufficient processing power to achieve human level intelligence might be
possible in 40 or 50 years.  A lot of assumptions there of course, but it's an
interesting read which basically compares processing power of computers and
various organic creatures, current high-end desktop machines being at about
insect level. Regardless, if and when computers get near that sort of power they
will have long ruled the chess world.  IBM has already built one that defeated
the world champ, and could easily build a stronger one if they felt like it, so
the technology exists already.  The interesting question is when standard
desktop machines will be world champ level. Then of course as time goes on, you
have to wonder what level a Deep Blue equivalent could play at 50 years from
now.  Let's hope we'll all be around to see. :)



This page took 0 seconds to execute

Last modified: Thu, 15 Apr 21 08:11:13 -0700

Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.