| |
BGonline.org Forums
4 ply and rollout
Posted By: neilkaz In Response To: OLM Sat. Feb. 2nd (Stick)
Date: Sunday, 3 February 2008, at 9:46 p.m.
The lesson here is that an race at an even pip with 100 vs 100 is about 57.5% cubeless for the side on roll. So, we do better by increasing the varience a bit with some contact. I have an huge race db so no huge rollout needed.
The score (after 0 games) is: gnubg 1, user 0 (match to 3 points)
Move number 4: user to play 66
Pip counts: gnubg 104, user 128
Position ID: 5pwzBgDsNsEYCA Match ID: cAl7ABAAAAAA
# Ply Move Equity 1 R 22/10 13/7(2) -0.1646
0.4341 0.0040 0.0001 - 0.5659 0.0026 0.0000 -0.1299 -0.1646 0.0014 0.0006 0.0000 - 0.0014 0.0006 0.0003 0.0030 0.0072 Full cubeful rollout with var.redn. 311 games, Mersenne Twister dice gen. with seed 1049564416 and quasi-random dice Play: world class 2-ply cubeful prune [world class] keep the first 0 0-ply moves and up to 8 more moves within equity 0.16 Skip pruning for 1-ply moves. Cube: 2-ply cubeful prune [world class] 2 R 22/10 16/10(2) -0.2002 ( -0.0356)
0.4275 0.0001 0.0000 - 0.5725 0.0000 0.0000 -0.1448 -0.2002 0.0006 0.0001 0.0001 - 0.0006 0.0008 0.0002 0.0013 0.0049 Full cubeful rollout with var.redn. 311 games, Mersenne Twister dice gen. with seed 1049564416 and quasi-random dice Play: world class 2-ply cubeful prune [world class] keep the first 0 0-ply moves and up to 8 more moves within equity 0.16 Skip pruning for 1-ply moves. Cube: 2-ply cubeful prune [world class] 3 4 22/16 13/1 13/7 -0.1886 ( -0.0241) 0.4390 0.0049 0.0000 - 0.5610 0.0034 0.0000 4 4 22/16 13/7(2) 8/2 -0.1900 ( -0.0254) 0.4385 0.0050 0.0000 - 0.5615 0.0030 0.0000 5 4 22/4 8/2 -0.2011 ( -0.0365) 0.4349 0.0051 0.0000 - 0.5651 0.0046 0.0001
| |
BGonline.org Forums is maintained by Stick with WebBBS 5.12.