Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Intelligent software, please

Author: Sune Fischer

Date: 15:20:34 11/25/01

Go up one level in this thread


On November 25, 2001 at 12:27:10, Victor Fernandez wrote:

>Of course. I am not a programmer, but I believe that in chess it is
>continued programming like 50 years ago. New techniques have been
>introduced but the chessprograms continue with the rules A and B of
>Shannon. The chessprograms continue being blind, they specify to analyze
>of 40/40 moves and they don't plan, they continue smashing with the
>"horizon effect" the chessprograms have that to begin to break the
>"horizon effect", using "chunks", planning to long term and to use
>algorithms to examine only 1 or 2 positions in each move discarding
>the other ones very faster, in another case will never end up playing
>chess well.
>
>Regards,
>
>Victor.

It sounds to me like you have little respect for the complexity of the game.
Chess is a huge "mathematical" problem, growing exponentially in size at every
ply. I'm actually very impressed with how deep top programs search today. Last
week I had a position and fritz6 announched mate in 18 in less than two minutes!
Maybe it was a really a mate in 17 or 16 but still very impressive I think.

Other ways like neural nets have been tried, not very succesful.
If you expect chess programmers to develop AI you are expecting a lot.
Probably 2600-2700 elo is about the best anyone can do on todays hardware.

Another problem is that humans don't really know "how" they think when they
play. Take me for example my rating is about 1800, what did I learn from when I
was a 1600? I wish I knew but I don't, just in general better understand of
positions I guess. I can't put exact words to it, so you can imagine why it is
hard to program.

-S.



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.