| |
BGonline.org Forums
Rollout
Posted By: Dave In Response To: Bear-In Problem (Dave)
Date: Wednesday, 21 May 2008, at 4:32 a.m.
Only 1024 but with that it's close. If anyone wants the rollout to go further let me know. It's interesting that 18/11 still wins a few more games but 9/2 is at least as good a play. What's further interesting is that nothing occurred to me except 9/2 so clearly I did miss out on this other possibility. Any ideas on what it would take to keep the basic idea but where the 9/2 play would be inferior? I got "lucky" here but it's obvious to me now that there are other things I need to consider when playing these situations.
The score (after 0 games) is: gnubg 1, user 0 (match to 3 points)
Move number 7: user to play 43
gnubg 208
user 97 Position ID: Jh8AgG/atlWAAA Match ID: QQluABAAAAAA
• user moves 9/2
Alert: doubtful move ( +0.000)
# Ply Move Equity • 1 R 9/2 +1.342
0.856 0.692 0.547 - 0.144 0.000 0.000 +1.342 +1.342 0.003 0.004 0.008 - 0.003 0.000 0.000 0.009 0.009 Full cubeful rollout with var.redn. 1025 games, Mersenne Twister dice gen. with seed 717121342 and quasi-random dice Play: 2-ply cubeful prune [world class] keep the first 0 0-ply moves and up to 12 more moves within equity 0.16 Skip pruning for 1-ply moves. Cube: 2-ply cubeful prune [world class] 2 R 18/11 +1.335 ( -0.006)
0.867 0.663 0.506 - 0.133 0.001 0.000 +1.335 +1.335 0.004 0.005 0.009 - 0.004 0.001 0.002 0.012 0.012 Full cubeful rollout with var.redn. 1024 games, Mersenne Twister dice gen. with seed 717121342 and quasi-random dice Play: 2-ply cubeful prune [world class] keep the first 0 0-ply moves and up to 12 more moves within equity 0.16 Skip pruning for 1-ply moves. Cube: 2-ply cubeful prune [world class]
| |
BGonline.org Forums is maintained by Stick with WebBBS 5.12.