Author: Ferdinand S. Mosca
Date: 05:17:03 11/01/01
Go up one level in this thread
On October 30, 2001 at 17:56:10, Robert Hyatt wrote: >On October 30, 2001 at 17:02:16, Simon Finn wrote: > >>On October 30, 2001 at 15:40:54, Robert Hyatt wrote: >> >>>Nope. You didn't understand. In a KRB vs KRPP ending, the position is >>>evaluated as "white can't win, black can". If the pawns are not advanced >>>then the score will be latched to zero. But if the pawns can advance (and >>>two connected passers don't have to be advanced very far to offset that >>>piece) then the score can swing negative. >> >>I understand that - if the pawns have real winning chances Crafty >>will find them. >> >>The problem is that if the pawns don't have real winning chances >>(if they're split and far back), Crafty is blind. There's no >>incentive to even defend them both since (I presume) KRBKRP >>is still "drawn". If it drops a pawn, it won't get enough >>positional compensation to outweigh the material disadvantage, >>so it's going to stay blind for the rest of the game. >> > >My suspicion, so far, is that a KRB vs KRPP is a draw. It is not easy to >prevent the KRB side from sacrificing the bishop for one or both pawns, >leaving the position dead drawn if the king is not horribly misplaced. Hello Bob, Some ideas: 1. Both sides have minimal chances to win 2. Both sides have many chances to draw 3. The key for the KRB side is to block the pawns or one of the pawns by the king. 4. The key for the KRPP side is to put the opponent's king away from the pawns. Which is easy to do, (3)? or (4)? Regards, Dinan > >I'm less worried about trying to win the KRPP side of this than I am >about not reaching the KRB side because I do have an extra piece there, >and that is not hard to reach if my program doesn't know that reaching it >is a bad idea.. > > > >>>It is _very_ difficult to walk into a mate in a KRB or KRN vs KRPP ending. >>>Those "mating nets" are nearly impossible to create... >> >>Against sensible play, yes. Against random play? >> >>>>How long can Crafty play like this until it falls into a mate that >>>>was just over its horizon? >>>> >>>In KRB/N vs KRPP it can play a _long_ time as there are hardly any mates >>>in the position. >> >>If you play sensibly, certainly. But if you drop a pawn, then >>continue to play "random" moves, who knows what might happen? >>Probably the KRBKR database will save you. >> >>Or perhaps a GM will discover that the "random" moves aren't >>actually random (there might be a tendency to evacuate the king >>to the top-left hand corner of the board, for example) and find >>some way to take advantage of this? >> >>> And if the pawns are advanced and the usual endgame >>>heuristics are being used, mates are impossible to create with the king >>>in the center of the board and the rook on a reasonable square... >> >>No argument there - if the pawns are sufficiently far advanced that >>Crafty plays using its usual evaluation function (without the cut-off) >>then you won't lose. >> >>Simon
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.