Author: leonid
Date: 05:31:09 02/21/01
Go up one level in this thread
On February 20, 2001 at 11:58:17, Frederic Friedel wrote: >On February 19, 2001 at 16:10:26, leonid wrote: > >>Very liked your position even if my selective search in 13 failed to find >>something in it. I hope that you will come with the next one very soon. > >The beauty of this problem by Gerhard Josten is that it starts off with: >[D]1q4r1/1bp2pQ1/3Q4/1Kp1pr2/1n1pkp1p/RQ1RN3/nbnP4/QN6 w - - 0 1 > >and ends with >[D]6r1/8/8/2p1pp2/2K1kr2/3Ppp2/1b6/8 b - - 0 13 > >White has given up all of his material and finishes it off with his last pawn: >13.d2-d3 mate. This position is for sure interesting and one reason for this you just said. I found this position attractive since it make me think about positions that I like to create and see their solition by different engins. I think that the main interest in mate containing position (beside what you indicated) are the next one: 1) Usual position from chess game where excellent human player found the mate. 2) Artificial mate position that is complet invention. The first version of positions are very few that are interesting now, since human usually don't find on the place any mate beyond 4 moves in one second. Time for solution of those position for the program is just too small to ajust computer mate solver in the best way possible. Short time in solution don't permit to recognize what way of solving those mates is the best. Second kind of position must not go too far in number of moves since solution by brute force will be almost impossible even in 5 hours time. Usually artificial positions are very good way to see the solidity of chess program and some mistake in its logic. Artificial positions are partially good (since the natural are absent) for ajusting selective search in mate solver. I hope that new "natural" and artificial mate positions will come to this Club very soon. Leonid.
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.