Author: Reinhard Scharnagl
Date: 08:00:15 01/18/04
Go up one level in this thread
On January 18, 2004 at 10:25:09, Ed Trice wrote: >This game ended as a win for Vortex "regular", but it could have been drawn in >practice. The endgame began as Archbishop + 7 pawns versus Knight + Rook + 7 >Pawns, but black had a set of tripled pawns. Most of the ending was Archbishop + >5 Pawns vs. Rook + Knight + 3 Pawns. > >As a result of this one game, it is apparent that there is a wide open field for >setting the piece weights in Gothic Chess. Vortex consistently sowed a score of >+165 when it could have been a dead draw, while the Reinhardt version showed -45 >near the end, which was closer to the mark. > >Regular Gothic Vortex vs. Reinhardt Value Vortex > >[Event "Gothic Chess Match Game #1"] >[Site "Philadelphia, Pennsylvania"] >[Date "2004.01.17"] >[Round "1"] >[White "Gothic Vortex 1.0.1"] >[Black "Gothic Vortex 1.0.1R (Reinhardt Piece Values)"] >[Result "1-0"] Hello Ed, I had a short view on the opening of that game, which is to huge to enter it by hand completely. There I found again that piece values alone could not secure a well played opening. Here is a wide field for testing appropriate positional detail evaluations. (I am - as you will know - not a fan of huge opening libraries, which only would hide that problem.) Interesting to read your conclusions on the game though black could not secure a remis. Are you willing to explain the weights of your fine evaluation you use in your program compared to the average material bilance? I think it has to be about 25% more important than within traditional chess. Sorry that I still have not reactivated chess (GC) programming yet, but I am still not able to concentrate except for a very short time. I would like to generate qualified perft values, because there should be a pool of such figures to become able to verify new move generation for all the rising GC engines. Regards, Reinhard.
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.