Re: computer chess

From: Alejandro Dubrovsky (s328940@student.uq.edu.au)
Date: Mon May 14 2001 - 13:23:25 MDT


On Sun, 13 May 2001, Spike Jones wrote:

> OK, so it was 133 and 166 Mhz Pentium Pros. In
> any case, since then we have had three generations
> of processors and more than a factor of 10 improvement
> in processor speed. I saw a 1.7 Mhz P4 for sale at
> Fry's last week for 1400 bucks. I bet those machines
> would scalp some the best carbon based players.

the problem is that a factor of 10 improvement doesn't buy you much in
chess since the tree to be searched is exponential. Algorithms and
evaluation improvements make a much bigger difference.

>
> However, I now suspect some such tournament
> must be taking place somewhere, behind closed
> doors with the scores being kept secret for reasons
> obvious to those handful of players in this world
> who manage to make a living playing chess.
>

They are played and (some at least) are not behind closed doors. There
was a recent match between 10 IMs and GMs versus Deep Fritz and Deep
Junior (two very strong programs, Fritz and Junior, running on Quad PIII
1Ghz. It has become a custom to prepend Deep on any program when it runs
on multiprocessors), and the computer pair won 15-5 or something along
those lines. Deep Junior also took part in a super-GM tournament recently
(super-GMs mainly being top 20 people) and came fourth with 50% score.
There is a match coming up (or just happened, dunno, haven't been keeping
up) between Kramnik (current world champion) and either Junior or Fritz.
Kramnik is a computer specialist though, so he is supposed to win. Both
Junior and Fritz have FIDE ratings that would put them in the top twenty,
and they play quite a bit of tournaments. Shredder, the current computer
champion, doesn't play much though.

Alejandro



This archive was generated by hypermail 2b30 : Mon May 28 2001 - 10:00:05 MDT