Author: Uri Blass
Date: 07:45:33 01/12/02
Go up one level in this thread
On January 12, 2002 at 05:00:39, Matthias Gemuh wrote: >On January 12, 2002 at 00:58:08, James T. Walker wrote: > >>I remember seeing some complaints about the book provided by Chessbase for >>Shredder 6. I don't know what started the complaints but I decided to do a >>little test when my Shredder 6 arrived. Since I have 2 Athlon 1400 Mhz machines >>I played the Chessbase version with it's book vs the Shredder 6 Classic version >>with it's own book. (Played in auto232 mode with both machines using 128M hash) >> The engines are supposed to be the same. I just finished 66 Blitz games >>(G/5minutes) with the Chessbase version winning by 40.5-25.5. Maybe someone can >>explain this result to me. I can't. It's a small sample but I rarely see the >>kind of turn around required after that many games to make these two come out >>even. It certainly does not look like there is anything wrong with the book >>provided with the Chessbase version. >> >>Jim > >Thanks for proving the old story: >that the autoplayer favours CB native engines. The story is not proved. There are other possible explanations: 1)The classic book is not good for blitz 2)The classic book is bad for a long match. Common sense say that big books are better for blitz and long matches. The quality of moves in blitz is not good so it is logical to trust book and not the engine. When the time control is slower the quality of the engine move becomes better than the quality of the book moves so smaller book is more productive. Big book is going to win against small book in long matches because the program with the small book may have a learning problem when it can see that all the lines are bad. A possible solution for long match may be to start the match with a small book and to increase the book only after losing games. 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.