Author: Ryan B.
Date: 23:37:07 10/14/05
Go up one level in this thread
It turns out you are right on the value of the pawn in fruit. I was mistaken on what fruit uses for its pawn value. Notice however that the lower value for pawn helps avoid the bad minor for 3 pawns trade when it really is bad and unlike the "BadTradeValue" idea it does not hold the program back from sacrificing the exchange for other forms of compensation. On October 14, 2005 at 23:46:14, Uri Blass wrote: >On October 14, 2005 at 18:07:53, Ryan B. wrote: > >>I think it would be safe to turn the value of pawn back down to 100. Pawn >>structure is undervalued in fruit and passed pawn is maybe a tiny bit >>undervalued if at all. > >I think that Jonas is probably right and there is no reason to turn the value >back down to 100. > >Fruit probably tends to underestimate pawns and it is not about pawn structure >and I think that in position when one side has some compensation for being pawn >down fruit tends to have higher evaluation for the side that is pawn down >relative to other top programs. > >Note that I am not sure about it and it is only my opinion and of course things >are dependent on the position but I guess that increasing the value of pawn can >be productive more often than counter productive. > >I do not expect big improvement from changing the value of pawn from 100 to 102 >and of course testing against one opponent may be misleading but my guess is >that the change is going to be productive for fruit. > >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.