Author: Les Fernandez
Date: 20:19:26 01/14/04
Go up one level in this thread
On January 14, 2004 at 06:09:22, Richard Bean wrote: >On my webpage I've now linked to all the unique checkmate positions at ply >depths up to 8. These were used to generate the integer sequence >http://www.research.att.com/projects/OEIS?Anum=A079485 > >They aren't in EPD format, but you can see the EP and castling data in the last >two columns, so it is not difficult to convert. The first column is the number >of times that position occurs at that level. > >http://www.eskimo.com/~rwb/chess/p4.mates.bz2 >http://www.eskimo.com/~rwb/chess/p5.mates.bz2 >http://www.eskimo.com/~rwb/chess/p6.mates.bz2 (4k) >http://www.eskimo.com/~rwb/chess/p7.mates.bz2 (100k) >http://www.eskimo.com/~rwb/chess/p8.mates.bz2 (1.1Mb) > Hello Richard, I think the generation of these checkmates may have some interest. One example is Dann Corbit had done something similar to this a while back and thought having this info may allow us to prune trees so that those branches dont have to be involved in searches. I am interested in seeing the data but when I try to open the file it first launches ICC I think because of the extension. However even when I change the extension to txt the file is obviously not in text format. Is it possible for you to put the info into a text based file? Thanks, Les >Now, is it useful to keep generating these for plies 9 and 10? I can't think >why, because I'd have to start considering three-fold repetition to make it >meaningful. I'd be able to solve the famous problem about 5...NxR mate with >grep, but I wouldn't get anywhere near finding the shortest stalemate. I don't >know what sort of algorithms would be needed to find it. I'm sure something >quite imaginative would be needed.
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.