> > I think it has already been reported and, if I'm not wrong, an easy > > solution would be to sort the moves by cubeful equity and then by > > cubeless equity: when all of them have cubeful equity +1.000 (or > > -1.000), this should prevent odd-looking moves. Not a big deal > > however, the match is already won/lost ... > > No that won't help. If all moves corresponds to > "game(gammon/backgammon) won/lost" they cannot be sorted by > evaluation, but need some "human" quality added to them. For > example maximum number of cross-overs. >
Humans will make the move that would be best if the race were closer, because we like to play "correctly". To simulate this, you could maybe feed into the evaluator the same board position for both opponents. But is it worth the effort to code, and the evaluation time? The only reason for it would be for learning good technique. -- Ian
_______________________________________________ Bug-gnubg mailing list [email protected] http://lists.gnu.org/mailman/listinfo/bug-gnubg
