Author: Steven Edwards
Date: 09:22:28 07/04/05
I've done some more work on Symbolic and it has produced a new search narrative for the BWTC.0031 position. The narration and the tree dump are linked from the updated web site (you might have to hit "refresh" on your browser): http://www.geocities.com/chessnotation/ See: http://www.geocities.com/chessnotation/Story.html Notes: 1. Symbolic thinks that BWTC.0031 is a mate in six, although the position is really a mate in ten. The program does get the first part of the analysis correct. 2. I've fixed a few bugs with analysis propagation, and that has significantly reduced the node count. 3. There is still a very long way to go. Perhaps the next step will be to throw the entire BWTC suite at Symbolic to see if the program crashes or hangs. Excerpted fom the narration: I've used five seconds of wall clock time in making the selection. I've used four seconds of processor time performing the selection. My processor utilization rate is eighty-three percent. The search tree has twenty-two interior nodes and six leaf nodes for a total of twenty-eight nodes. The maximum variation length is eleven ply. My selection rate is seven nodes per processor second, and my mean node processor time is one hundred forty-four milliseconds.
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.