Author: Bruce Moreland
Date: 10:47:28 08/14/98
Go up one level in this thread
On August 14, 1998 at 13:12:32, José de Jesús García Ruvalcaba wrote: >In "Computer Chess II" by David Welsh and Boris Baczynskj (Wm. C. Brown >Publishers, 1985), pp. 66-67, it is briefly described PAWNKING, a program >developed by Helmut Horacek at the Vienna University (it was a result of a >research about pattern recognition). It was able to play any kind of pawn >endgame. >Does anybody have more information about it? >I know that it is an old book, but I still hope somebody can help. >Thanks in advance. I don't have any information about the program, which is probably interesting. The position given as an illustration is also interesting: 8/5p2/2k5/K7/8/1P6/8/8 b - - 0 1 The key is 1. .. Kd5. On a P2/300 my general-purpose program finds this in under a second, but has some search instabilities and doesn't get a stable huge plus score (+7) for about four seconds, this with endgame databases turned off, of course. The book mentions that PAWNKING took ten minutes to get this answer on a CDC Cyber 170/720, and "would take about 30 seconds on a Cyber 176". I don't know how fast these machines are, so I can't compare effectively. It might be interesting to know what patterns they use. If you are interested in old stuff, you may want to check out PEASANT, which is another K+P program. It is discussed in "How Computers Play Chess", and "Chess Skill in Man and Machine" (Levy and Frey, respectively, probably both out of print, the Levy book more recently so). Monty Newborn wrote Peasant, and I know he's still around. bruce
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.