Author: Peter Berger
Date: 07:46:09 05/03/02
Go up one level in this thread
On May 02, 2002 at 04:05:00, Uri Blass wrote: >On May 01, 2002 at 19:18:15, Peter Berger wrote: > >>On May 01, 2002 at 10:51:41, Uri Blass wrote: >> >>>I do not think that the word sacrifice is correct here. >>> >>>4 pawns is considered to be more than a piece so the player that sacrificed was >>>yace but it seemed that yace was outsearched because yace evaluats the position >>>at move 26 also as better for hiarcs. >>> >>>There are 2 possibilities: >>> >>>1)Yace was lucky to be outsearched and got a winning position that it did not >>>understand. >>> >>>2)Hiarcs blundered some moves after getting a big advantage by winning 4 pawns >>>for a piece. >>> >> >>As both engines agreed for some time on evaluation 1.) is somehow closer to the >>truth . I definitely disagree with your assessment about piece value though. In >>middlegame ceteris paribus I'd take the piece over the pawns any time. >> >>Peter > >piece is usually better than 3 pawns in the middle game but here we are talking >about 4 pawns. > >I do not think that piece is usually better than 4 pawns in the middle game. > >It is possible to check it by comp-comp game when one program plays without a >piece and another program plays without 4 pawns. > >Suppose that white plays without a2,b2,c2,d2 and black plays without >the bishop c8. > >Do you expect white to win the game? > >I remember that I did some comp-comp games in similiar conditions(it is possible >that I removed different pawns and different piece) and the side without the 4 >pawns scored less than 50%. > >Uri I am not sure; I think the problem with testing this is the ceteris paribus here. If you take a2,b2,c2,d2 you also create three connected passers on a,b and c for black (significant additional positional advantage). Also the start position isn't exactly a middlegame one. Look at the Hiarcs-Yace game. Progs evaluated about +1 for the side with the pawns. Any human would prefer the piece here I believe. It might be a better idea to test which side is better in material to create a few typical middlegame positions without any additional advantages for one side where one has the piece, and the other four pawns for it - I don't think I am going to do it though :-). Peter
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.