Author: Dann Corbit
Date: 12:35:26 10/04/05
Go up one level in this thread
On October 03, 2005 at 16:47:03, Chris Conkie wrote: >On October 03, 2005 at 16:37:36, John Merlino wrote: > >>On October 03, 2005 at 15:00:39, Chris Conkie wrote: >> >>Do you honestly think that a programmer will spend time trying to fix the kinds >>of bugs that you are reporting, that come from situations that are either 100% >>unlikely to happen in a real game or are illegal, rather than fix/improve his >>engine's ability to simply "play chess"? >> >>Your comment "These kinds of position are extremely useful" seems to only be >>true for you. >> >>jm >> >>p.s. If an engine writer has actually bothered to fix any of these kinds of >>issues that Chris has reported, please feel free to reply. I would, once again, >>be intrigued.... > >Would an engine programmer be bothered to fix this John? > >[d] K2Q1Q2/1PQpppQ1/1NRpkp2/2QpppQ1/3Q1Q2/8/8/8 w - - 0 1 It is impossible for this position to occur in a real game (that pawn formation cannot be formed by any sequence of legal moves). So there is no value to fixing it for a chess playing engine. However, puzzle writers sometimes free themselves from the fetters of legality with the chess moves. For this audience, there may be some interest in making a correction. Since the position literally cannot possibly occur in a legal chess game, why should the programmer spend time and energy bothering to fix it, if best possible chess play is the goal of his efforts? IMO-YMMV.
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.