On 2014-08-01 8:29 PM, Tim Leonard wrote: > When gnubg (version 1.02.000) had a certain win in one move, it chose to > waste a pip, causing it to win in two moves. > It had men on the 2 and 1 points, rolled a 5-1, chose to move the 2 to 1 and > cast off the 1, rather than casting off both. > I’ve attached the game log. >
What is going on is that both plays one off or 2 off still guarantee victory. The bot sees that it has a choice between 2 moves which guarantees a win. It happens to be that at the time of the analysis the move that was listed first happened to be the one that takes them off in 2 rolls. The bot didn't lose anything in doing so. I would call this a rather low priority bug/deficiency. The question would have to be asked why the bot didn't consult the bearoff database directly for this position. -- Michael Petch GNU Backgammon Maintainer / Developer OpenPGP FingerPrint=D81C 6A0D 987E 7DA5 3219 6715 466A 2ACE 5CAE 3304 _______________________________________________ Bug-gnubg mailing list Bug-gnubg@gnu.org https://lists.gnu.org/mailman/listinfo/bug-gnubg
