Author: Robert Hyatt
Date: 11:20:52 03/18/04
Go up one level in this thread
On March 17, 2004 at 19:16:55, ERIQ wrote: >On March 17, 2004 at 18:39:37, Uri Blass wrote: > >>On March 17, 2004 at 16:59:19, ERIQ wrote: >> >>>After reading what others thought about other unix compatable engines, I found >>>pepito and just for fun started a blitz match w/ crafty which ended 4-4-2 dead >>>even. I then set pepito to aggressive and started a 10min/40move match and wow >>>at the moment crafty is lossing yet another game and the score after present >>>game will be 7 wins 1 draw 0 losses in favor of pepito. I have noticed that w/o >>>a book crafty tends to lose matches but this is its worse beating to date. >> >> >>I do not see the points of matches with no book. >> >>Even if you want to test general knowledge it is better to use the nunn match >>or something similiar. >> >>Playing a match from the opening position with no book can cause programs to >>repeat almost the same game again and again and give you no information about >>the question which program is better. >> >>Even if the games are not almost the same repeating the same opening line can be >>enough for result that give almost no information about the strength of the >>programs. >> >>Uri > >I can see your point but it will show how well a program learns! How can a program "learn" when there is no book for "book learning" to work with???
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.