[ View Thread ] [ Post Response ] [ Return to Index ] [ Read Prev Msg ] [ Read Next Msg ]

BGonline.org Forums

XG rollout option

Posted By: eXtreme Gammon
Date: Friday, 31 December 2010, at 2:16 p.m.

In Response To: XG rollout option (Nack Ballard)


Nack: If you turn the option ON, then XG splits the rollout (from two candidates) into four candidates: (a) 61P-22N, (b) 61P-22E, (c) 61P-22N-C-T and (d) 61P-22E-C-T (where C = Cube and T = Take). The rollout continues until it becomes “clear” that (a) or (c) is doing better, and/or that (b) or (d) is doing better. For example, say that (a) and (d) are doing clearly better; in that case, (b) and (c) are dropped, and the rollout finishes with just (a) and (d), and XG reports only those two candidates in the results.


You are close Nack. a and b will always be rolled the same number of time, no matter what, same for c and d. At the end the system jsut pick what is the best choice: a or b

Here is how it will be displayed in the analyze dialog for XGID=---A-bD-C---cD---c-eA-b-B-:0:0:1:51:0:0:3:0:10

 


move Trial Winning chances Equity conf. Nack
24/23 8/3 1296 Joueur: 37.63% (G:8.15% B:0.32%) -0.567 0.022 a
Adversaire: 62.37% (G:23.43% B:1.20%)
opp. Double: No Joueur: 37.61% (G:8.26% B:0.31%) -0.534 0.028 b
Adversaire: 62.39% (G:23.54% B:1.14%)

13/7 1296 Joueur: 35.89% (G:7.38% B:0.31%) -0.677 0.022 c
Adversaire: 64.11% (G:25.81% B:2.56%)
opp. Double: Yes Joueur: 35.89% (G:7.47% B:0.34%) -0.707 0.028 d
Adversaire: 64.11% (G:26.09% B:2.71%)



The final display will report a (no double) and d (double) as they are the best cube option for the opponent


est Joueur 2

score: 0
course: 139
Partie illimitée
Jacoby Beaver
course: 171
score: 0

est Joueur 1
XGID=---A-bD-C---cD---c-eA-b-B-:0:0:1:51:0:0:3:0:10
à jouer 51
1.Simulation124/23 8/3éq: -0.567
Joueur:
Adversaire:
37.63% (G:8.15% B:0.32%)
62.37% (G:23.43% B:1.20%)
Conf.: ± 0.022 (-0.589...-0.545)
Durée: 23.8 secondes
2.Simulation113/7éq: -0.707 (-0.140)
Joueur:
Adversaire:
35.89% (G:7.47% B:0.34%)
64.11% (G:26.09% B:2.71%)
Conf.: ± 0.028 (-0.735...-0.678)
Durée: 27.0 secondes
1 1296 parties simulées avec réduction de variance.
Simulé avec ou sans double
Toutes les décisions en 1-pli

eXtreme Gammon Version: 2.00.153.alpha






Messages In This Thread

 

Post Response

Your Name:
Your E-Mail Address:
Subject:
Message:

If necessary, enter your password below:

Password:

 

 

[ View Thread ] [ Post Response ] [ Return to Index ] [ Read Prev Msg ] [ Read Next Msg ]

BGonline.org Forums is maintained by Stick with WebBBS 5.12.