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

BGonline.org Forums

Re ..etc opening 43 RO repost .etc etc

Posted By: neilkaz
Date: Saturday, 27 November 2010, at 5:29 p.m.

In Response To: RO ..etc etc etc (christian munk-christensen)

CMC asks whether I consider the opening 43 to be solved. While the result that D is superior to either split doesn't have 100% stat sig, it does agree with other high level bot RO's especially the long GNU RO posted here. These long times to complete are with my work computer which isn't as fast as what I have at home and really won't be viable for opening RO's using XGR+.

Anyhow, I'll likely RO XG2 3 ply 100k or so for the opening 43 anyhow as that won't take all that long, and if I have something to suspect that a split (most likely Z) is superior or even equal to D, then I can at some point try a RO using XGR+.

What if my XGR+ RO with the XG2 NN ends up showing that it is very correct to slot? Well that will mean (IMHO) that I should try a RO of the opening 41 using XGR+. It will also likely mean that splitting the opening 43 is worse since these splits result in getting hit loose with the opp slotting a key point and if slotting becomes more correct than it has been RO'd to be now, playing D with 43 will likely be more correct and playing D with 32 to be considered.

However, I am not convinced that slotting 21 will become more correct (per bot RO's than it is now). While higher strength bots become better at playing backgame-type positions they also become better at defending vs them. This being said, I now think when I start the process for opening 21, I'll RO 100k using XG2 3 ply to see if the $ to S differential is clearly different than when I spend the few weeks to RO with XGR+.

is Player 2

score: 0
pip: 167
Money session
Jacoby Beaver
pip: 167
score: 0

is Player 1
XGID=-b----E-C---eE---c-e----B-:0:0:1:43:0:0:3:0:10
to play 43

1.Rollout113/10 13/9eq: +0.010
Player:
Opponent:		 50.04% (G:15.47% B:0.91%)
49.96% (G:13.94% B:0.85%)		Conf: ± 0.003 (+0.007...+0.013)
Duration: 7 days 23 hours 18 minutes
2.Rollout124/21 13/9eq: +0.005 (-0.005)
Player:
Opponent:		 50.09% (G:14.38% B:0.81%)
49.91% (G:13.53% B:0.71%)		Conf: ± 0.003 (+0.002...+0.008)
Duration: 8 days 06 hours 49 minutes
3.Rollout224/20 13/10eq: -0.002 (-0.012)
Player:
Opponent:		 50.12% (G:13.84% B:0.77%)
49.88% (G:13.55% B:0.70%)		Conf: ± 0.004 (-0.006...+0.002)
Duration: 3 days 15 hours 49 minutes
4.Rollout324/21 24/20eq: -0.017 (-0.027)
Player:
Opponent:		 49.91% (G:12.55% B:0.70%)
50.09% (G:13.34% B:0.61%)		Conf: ± 0.006 (-0.023...-0.011)
Duration: 1 day 20 hours 20 minutes
 
1 46656 Games rolled with Variance Reduction.
Moves and cube decisions: 4 ply

2 23328 Games rolled with Variance Reduction.
Moves and cube decisions: 4 ply

3 10368 Games rolled with Variance Reduction.
Moves and cube decisions: 4 ply

eXtreme Gammon Version: 1.21

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.