Author: Charles Worthington
Date: 14:18:34 04/15/03
Go up one level in this thread
On April 15, 2003 at 15:46:33, Charles Worthington wrote: >On April 15, 2003 at 05:23:39, Ulf Flörsheimer wrote: > >>I used the two original books delivered by ChessBase. And this was an >>automatically played game in a match of 40 games, which was won 22.5 - 17.5 by >>Shredder 7.04. >> >>I've never taken a look on the playing-logic of ChessBase books, but maybe its >>an effect of booklearning. I only can say, that both original books were used >>and it was a really played game - not a constructed one. >> >>Ulf > > >Well the blunder by the engine isnt suprising but try to remember the books >chessbase uses are thrown together from human games...not program games. Don't >let anyone try to convince you that a human book is stronger in machine vs >machine than a program book because it just isn't so. The factory books will >often leave you with a lost game out of book. A book of your program games >imported into a book like maybe remis.ctg will prevent many of these bad lines. > >Charles All Programs have little peculiarities in certain positions where they will make such blunders. Most of the time a programmer can take some sample games and correct the problem by duplicating the position and finding the problem. I send most such games to the programmers as well but normally by e-mail to ensure that they get it. Often times they only post here sporadically and he may not see the post.
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.