Author: Robert Hyatt
Date: 10:20:54 01/21/99
Go up one level in this thread
On January 20, 1999 at 17:06:59, blass uri wrote: > >I believe that all of the moves are right because I read that darkthought found >all of them, but there are some cases when there is more than one practical >solution. >I believe that Junior5's move Nf6+ in number 29 is also winning(The solution >Nxh6 is the simplest). > the fact that Dark Thought (or any other program) gets the suggested best move doesn't convince me those are right best moves. IE I could easily pick a set of 100 endgame positions that Crafty solves perfectly, without being sure that the 'best move' is "the only best move that doesn't turn the game from a win to a draw or from a draw to a loss." That's not to say that Dark Thought is "bad" or that the key moves in the positions are "wrong". But I went thru a lot of this in the various suites I run. IE even in WAC Crafty found a couple of 'cooks' where it was thought that in a couple of positions only one move would win. But occasionally there is a 'waiting' move that doesn't give up the win, but it might be hard to see at first. That's why I asked my question... for each best move in the suite is it clear that the best move is really the only move to play? Or is it just slightly better? etc... > >I also know that Junior5 is bad at very simple positions relative to other >programs. >For example it could not solve position number 14 in 10 minutes. > >Uri
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.