Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Computers for Chess only.

Author: Dann Corbit

Date: 11:33:55 11/17/99

Go up one level in this thread


On November 17, 1999 at 05:10:04, Pekka Karjalainen wrote:
>On November 16, 1999 at 23:55:19, Dann Corbit wrote:
>>Has the formal solution to checkers put an end to draughts?  Just because a
>>machine can solve a problem does not mean that the problem is no longer
>>interesting.
>  Is there a formal solution to checkers?  I thought the game-tree was just
>too big for that.  Can you supply a reference, please?

http://www.cs.ualberta.ca/~jonathan/Papers/Papers/aimag96.ps

They had a goal to complete the endgame database (10^20th positions) for the
computer to play perfect checkers.  I thought that it has been accomplished, but
I was mistaken.  They have completed the 8 piece tables and are working on the 9
piece tables:
http://www.cs.ualberta.ca/~chinook/databases/databases.html

At any rate, it is probably the world's strongest player (but with the paucity
of matches, I don't think it is nearly as certain as the sponsors seem to
think).

Here is an interesting "Scientific American" article on computer games:
http://www.sciam.com/1998/1198intelligence/1198ginsberg.html

>>I suspect that Tic-Tac-Toe is played by as many children as ever before,
>> despite the fact that it is a cat's game from the get-go.
>
>  The fact that almost anyone can learn how to play perfect tic-tac-toe does
>mean the game has no scope for professional play.  If chess could be solved so
>that almost anyone could learn all correct openings and variations by heart
>easily, it would affect the game as it is played on the highest levels severely.
>
>  That if is only science fiction, however.  Solving the game is far different
>from producing a world champ -class player.



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.