Author: Ulrich Tuerke
Date: 10:39:33 05/01/02
Go up one level in this thread
On May 01, 2002 at 06:10:48, Uri Blass wrote: >On May 01, 2002 at 05:55:45, Ulrich Tuerke wrote: > >>This time, comet left its terrible book wirh a minus of 2 pawns. > >I do not understand it. >I do not understand what is the reason to play in 0 seconds lines that not you >and not your program analyzed them in the past. > >I will never tell my program to play these kind of moves in 0 seconds. Well, former versions of Comet used to play in the SSDF matches. You get killed by the book-learners there if you play with a small book. So, I definitely needed a large book. I don't have time to writemy self a larger book. Neither am I a chess expert. So, I 'm using database extracts of master games. I have also typed in lots of variations which I found in books and magazines. Lines which are bad will be deleted as soon as they have been found. > >I also do not understand Comet's evaluations > >Fritz1 has only 1 pawn advantage but Comet says 3.26 pawns against itself. It's a typical "Comet score", shouldn't be taken too serious. -:) But seriously, these are some passed pawn boni and some penalty for king being in corner in an ending. Usually this doesn't work too bad. After all, what is the evaluatiuon good for ? It's nothing but a way to guide the search in finding a good move. The most important thing for me is that the moves are okay. Don't care too much if -1.8 or -2.5. Uli >I can understand evaluation that say that white has positional advantage but >more than 2 pawns positional advntage seems too be too much. > >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.