Author: Howard Exner
Date: 06:30:51 10/29/98
Go up one level in this thread
On October 29, 1998 at 00:55:48, Mark Young wrote: >On October 28, 1998 at 21:01:36, Howard Exner wrote: > >>On October 28, 1998 at 11:41:10, Mark Young wrote: >> >>>Game 7 >>> >>>Fritz 5 1/2 1/2 1 1 1/2 1/2 1 >>> >>>Junior5 1/2 1/2 0 0 1/2 1/2 0 >>> >>>Can someone tell me at what point in this game is Junior 5 lost? >> >>Here is the position at move 62. >>8/6k1/3q2p1/8/4QN2/pP6/Pb2K3/8 w - - id Fritz 5.00/16Bit - Junior 5.0; bm Qxg6; >> >>It is unlikely that this position would be solved by any program without a >>special knowledge code. Just as programs now recognize the draw patterns in >>wrong colored bishops with the rook pawn so too would they need the knowledge of >>this draw. It is as common a theme as the wrong colored bishop one (twice >>I've drawn games to chess programs on chessnet with the operators probably >>wondering why I was playing on). >> >>The sequence 62. Qxg6+ Qxg6 63. Nxg6 Kxg6 is drawn. The white king reaches >>the a1 corner and cannot be ousted. I nice swindle. > >What ????????, But white won this game? So if Black is winning here, because >white needs to swindle a draw. What does black have to do to win, if white does >not play the swindle? What ??? indeed. This reminds me of the time while playing my friend in the pub. I picked up a piece and checked my opponent thinking to myself, "I've got you now". Only problem was that I picked up his piece and put my own king in check. Oh well sorry about the mixup in who had what colors. At least I salvaged a nice test position in the following: 8/6k1/3q2p1/8/4QN2/p7/Pb2K3/8 w - - id "Qxg6 - draw"; bm Qxg6; This is the same position without the white pawn on b3. Now white should play for the draw. By the way is there a way to solve this position? Do tablebases catch this as after the exchanges it goes to KP vs KBP? Which programs would play Qxg6? Now back to the game. Can Junior(black) draw this somewhere? I think on move 68 there is a line that looks like a draw. Here is the epd. 8/6k1/8/1P1Nb1p1/1Q6/p2K4/Pq6/8 b - - id "Fritz 5.00/16Bit - Junior 5.0"; bm Qxb4; One line runs like this:68... Qxb4 69. Nxb4 g4 70. Ke4 Bd6 71. Na6 Kf6 The white king is tied to the g pawn so black is free to march his king to pick up the white a pawn. The bishop will sacrifice itself for white's b pawn 72. b6 g3 73. Kf3 Ke5 74. b7 Kd4 ...draw Is there better play for white in this line? It reminds me a little of the KKUP2 ending with Crafty vs Rebel. Do programs have a kind of code that gives priorities to such exchanges that enable them to search deeper? Would any program exchange Queens here on move 68? Uri I also read your post to mine. As you can see it was a mental lapse on my part in mixing up the colors.
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.