Author: Sune Fischer
Date: 04:02:46 12/29/01
Go up one level in this thread
On December 29, 2001 at 06:34:12, Sune Fischer wrote:
>I think if you took 1 million GM games and did a history "heuristics" on them,
>we would get an "idea" of how often a GM places the knight on c3 or e5 or....
>Some basic analysis would then give you a good table score.
And if you want to further improve on these tables, you could divide the games
into groups: {(w&b did O-O),(w did O-O, black did O-O-O), (w didn't castle, b
O-O),...}, this will influence the history no doubt.
Also try and look at the move history when in the opening, midgame and endgame.
I'm quite sure that for instance the pawn score tables will be much different in
the endgames. Pushing the pawns forward too soon will make them hard to defend
but in the endgame they get closer to queening.
Those are quite a lot of tables, but they would be a little more adaptive and I
think this is exactly the kind of knowledge a GM has.
-S.
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.