Author: martin fierz
Date: 04:11:13 10/08/03
Go up one level in this thread
On October 08, 2003 at 06:22:52, Daniel Clausen wrote: >On October 08, 2003 at 04:02:33, martin fierz wrote: > >> >>>The program was conceived by a Dutch engineer, namely myself, >>>who thought maybe chess could be solved in a similar way >>>as 4-in-a-row. >> >>4 in a row was solved by many different people by a simple brute force search. >>do you really intend to do the same with chess? > >Just a small addition: There are also people (I think Viktor Allis is one of >them) who solved 4-in-a-row with a lot of game-specific knowledge which solves >_many_ positions very elegant. All he had to add was a search, which solved the >rest. (which was _far_ from brute force) He wrote an interesting paper about >that work, but I wouldn't be surprised if you (martin) read that already. :) i think victor allis is the only one who did this (there is this program "velena", but i think it's basically his engine rewritten, not quite sure though) - yes, i am aware of his paper. the interesting question is whether it's better to use this kind of knowledge-based approach or to use the brute-force approach. AFAIK the two first independent solutions to C4 were by victor allis and james d allen; both used roughly the same amount of computing time to solve the game. >Surely chess is a whole other story and whether similar (in nature) >game-specific knowledge exists is at least questionable. It's not completely >impossible though in my opinion - at least for certain end game types or similar >things. right, for certain endings it might be possible. but the focus of this bookbuilder project is clearly not on that - in the helpfile it says somewhere that it's really intended for openings only. i downloaded the trial version, and from what i see i can imagine what vincent would say about it ;-) cheers martin
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.