Author: Robert Hyatt
Date: 11:29:29 09/24/98
Go up one level in this thread
On September 24, 1998 at 13:12:27, Ernst A. Heinz wrote: >On September 24, 1998 at 11:43:38, Robert Hyatt wrote: >> >>In the ICCA Journal, Vol. 16, No. 4, GM Timoshchenko is comparing various >>chess programs on how they evaluate bishops vs knights. >> >> [... rest of interesting post snipped ...] > >Bob, > >I strongly doubt that GM experience and practice provide a good measure >of the relative values of Bishops and Knights in *computer* endgame play. > >IMHO, humans tend to handle Knights miserably as compared with how well >they handle Bishops. Fast chess programs, on the other hand, discover >beautiful yet deep Knight maneuvers within fractions of a second. They >often prove the common wisdom of "great advantage of Bishop(s) over >Knight(s) in endgames" wrong. Especially with respect to advancing and >blocking passed Pawns, Knights are actually superior to Bishops because >they cover both black and white squares. > >Conclusion: the static advantage of Bishops over Knights is far less in > *computer* endgame play than in human endgame play because > deep-searching chess programs handle Knights much better > than most (or even all) humans. > >What do others think about this? > >=Ernst= I disagree for the special case I gave: pawns on both wings. The knight has a hopeless task to defend on both sides, while the bishop can attack on both sides without even moving. The reason I added this was that in playing against some GM players, this slowly became a "motif" of theirs to play against Crafty (on ICC) *if* they could survive the middlegame tactics, they could reach endgames where this resulted in wins. Not every time, but more often than not. I have gotten two comments about this so for after OTB games vs two GM players, and both remarked something like "ok, it is about time you fixed that hole, now let me find another one." :)
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.