Author: Marc van hal
Date: 17:49:29 09/22/98
Go up one level in this thread
On September 21, 1998 at 22:37:25, Danniel Corbit wrote: >On September 21, 1998 at 22:16:07, Jeff Anderson wrote: > >>This is not even an idea as much as a useless half-thought born in the mind of a >>bored highschool student. Still I thought it might be slightly relevant. >> >>Imagine a set of overlayed transparencies, each with a chessboard printed on it, >>completely blank becides one piece. For every piece, black and white, there are >>64 transparecies, each having a blank board it and one piece on a different >>square each time. Now total you would have 768 transparence sheets. If you >>overlay them in front of you, every possible position on a chessboard can be >>reached by removing certain ones. >> >>Could a computer some how internally sort its way through these 'transparecies' >>extremely fast? Actualy here is also something strange going on I once had a conversation with Herman Grooten(a duth grandmaster) about chess position and the position was from a game of Boby Fisher where Boby played rook a1! And a chessprogrammer had just made so much changes on his program that the program also made the same move rook a1! in the same position so ok in chesspositions a program has to find a move after a normal combination this is treu but on the other hand the program didnt play well anymore in other positions an other fact is that almost all programs dont see an combination if a piece will be lost for some time so after 12plys the program doesnt see the combination but after 2 plys played the program sees in 7 plys that the combination is correct i also had the same conversation with E Schroders informant and that is why the new grandmaster option was build in the program I did send him a game I played against chessmaster 5000 after normal openings theory wich was also in Rebel 9.0 it was the queens gambit accepted and after a forced combination white could win after theory but Rebel didnt find this nor Chessmaster the only saw a draw but i found the win it is only a pity that my computer was hacked and that in the same week on the net was that crafty did find this combination in the kkcup f 1997wich was not treu I also did send him a gameof Nimzowitsch where Nimzowitsch played after his famous french game 0-0-0 no program could find this move cause it would lose a knight but all programms saw that after the capture black made the capture of the knight that is that white was in a winning position so does the programm calculate the position then it is ok but if it is programmed to play a certain move in a position that is not realy making the program stronger >Tweaked a bit, this idea is called "bitboards" and is commonly used. >See, for instance: >http://www.xs4all.nl/~verhelst/chess/representations.html >http://wwwipd.ira.uka.de/Tichy/DarkThought/node7.html >http://home.fda.net/~wzrdking/bitboard.htm >http://home.fda.net/~wzrdking/index.htm
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.