Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Computer Go Report

Author: Rémi Coulom

Date: 07:56:10 09/12/05

Go up one level in this thread


On September 12, 2005 at 02:05:40, Peter McKenzie wrote:

>Most (but not all) computer Go tournaments use Chinese rules because these don't
>punish you from playing extra stones at the end of the game (instead of
>passing).  However, you'll want your program to pass 'early' (rather than play
>all the stones that it possibly could) because otherwise humans won't want to
>play it.

Hi Pete

I have found a very efficient way to solve this problem: pass as soon as your
opponent passes. It works and avoids a lot of headaches.

Also, I confirm to other readers that Go is a great game to play, and writing a
computer-go program is incredibly more difficult than writing a chess program.
The most exciting part is that there is no really established state of the art,
and it seems that huge ideas still have to be invented. I think I have become
addicted to go programming, and I will probably not go back to chess.

Curious readers will find my go program there:
http://remi.coulom.free.fr/CrazyStone/

I learnt to play go with igowin:
http://www.smart-games.com/igowin.html
After having played a few hundred games against igowin, I tried go servers. KGS
seems to be the best for beginners. I am not sure that what I did is the best
way to get started, but it worked well for me.

A few programmers have already made this move away from chess. I hope many
others will follow. Go is really tremendously more fun than chess, both as a
game to play, and as a game to program.

Rémi



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.