Author: Gerd Isenberg
Date: 14:24:57 11/01/02
Go up one level in this thread
On November 01, 2002 at 14:40:48, Dieter Buerssner wrote: >Hi Gerd, > >for Yace, I get no dependence on lazy eval. Rxf2 is found at depth 8 witgh or >without lazy eval - the later needs about double time. With very slightly >different extension tuning (I have no idea which is practially better), it is >found at depth 7 (again with or without lazy eval). > >Cheers, >Dieter Hi Dieter, that's interesting. Of course a lot of other things have influences, eg. my eval detects mates. I'm not quite sure about this point, but this dynamic threshold i used is a bit suspect, or the way i implemented it. It should be better, to use maxPositionalScore[2] for both sides. Thinking about a smarter way to distinguish between full and lazy eval, eg. making maxPositionalScore depending from some cheap to compute properties of the position, that are likely to get huge positional scores, like advanced passers, ability to give checks, attacks of opponent pieces to adjacent king squares, king-mobility, gamestate etc.. Regards, Gerd
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.