Secondly, I see a very nice written report by David Fotland who shows clearly that go programs still are strugling to search basic tactics. Not counting forced moves how many ply is the mistake seen from brute force perspective when using a reasonable evaluation function?
David, is your report available for download anywhere? Or if not, is it possible to post here the position mentioned above? Darren _______________________________________________ computer-go mailing list computer-go@xxxxxxxxxxxxxxxxx http://www.computer-go.org/mailman/listinfo/computer-go/