BGonline.org Forums

Bot backgame comparison Redux

Posted By: Phil Simborg
Date: Monday, 14 December 2009, at 7:10 p.m.

In Response To: Bot backgame comparison Redux (Michael Depreli)

This is a fascinating study. It's amazing how few rolls they play all play the same.

The real question to me is this: how will we ever be able to judge which one is best? Even long rollouts on several bots, if they agree, may be "wrong".

It's an interesting exercise, and I, for one, plan to put several of the plays in XGRoller+ as that is all I use anymore before going to rollouts. I can't see a reason to use 3 ply when I can get something far more accurate in about 6 seconds.

Messages In This Thread

• Bot backgame comparison Redux
  Michael Depreli -- Monday, 14 December 2009, at 3:38 p.m.
  • Bot backgame comparison Redux
    Dan Pelton -- Monday, 14 December 2009, at 4:30 p.m.
    • Bot backgame GNU 41 Should Be 12/8 6/5 (No Hit) Typo
      Michael Depreli -- Monday, 14 December 2009, at 4:34 p.m.
  • Bot backgame comparison Redux
    Phil Simborg -- Monday, 14 December 2009, at 7:10 p.m.
    • Bot backgame comparison Redux
      Michael Depreli -- Monday, 14 December 2009, at 8:25 p.m.
      • Bot backgame comparison Redux
        Timothy Chow -- Monday, 14 December 2009, at 10:51 p.m.
  • Bot backgame comparison Redux
    Timothy Chow -- Thursday, 17 December 2009, at 2:14 a.m.
  • Bot backgame comparison Redux
    Daniel Murphy -- Thursday, 17 December 2009, at 3:18 a.m.
    • Bot backgame comparison Redux
      Timothy Chow -- Thursday, 17 December 2009, at 3:42 a.m.
      • Bot backgame comparison Redux
        Daniel Murphy -- Saturday, 19 December 2009, at 3:45 a.m.