| |
BGonline.org Forums
41S-21 RO
Posted By: neilkaz In Response To: 41S-21 RO (Nack Ballard)
Date: Friday, 1 January 2010, at 5:23 a.m.
When it comes to 41S-21 I think we have enough data to be sure that for normal money-like scores, S is correct.
Nack, thanks for your fine explanation comparing vs 51S-21. We all look forward to seeing more of your superb opening play books due to your amazingly insightful explanations.
Yeah...spending CPU time at 4 ply on 41S-21 isn't on my list since we have clear data from several bots and settings that agree with S. I suspect that I won't need to spend much time to come up with closer plays to RO 4 ply to attempt to mediate bot disagreements or ties.
For now...with about 9000 done, 51S-21 has U leading $ by .005.
At work with about 30,000 games done for 43 opener, D leads Z by .003 (S and U were dropped with much stat sig earlier)
The work comp will be freed up in a couple weeks as I only have a few scores left to RO for the 15 pt Supremo MET. It takes about 14 hours to RO a single score, so during a day, I let GNU RO one score, and GNU overpowers XG in priority, so the XG RO is basically stalled while GNU RO is in progress but when it ends, XG continues with its task. Both these are slowed a bit by my trading apps during the stock market day, as I set the BG apps on low priority so they don't affect my trading apps.
Once again, I want to use XG 4 ply RO to try to break ties and solve disagreements between the bots and in common positions... ie 51S-21 is lots more common than 41$-44. The dice sequence occurs twice as often, but in modern OTB play, most split 51 now at normal scores and few slot 41...well at least vs me.
.. neilkaz ..
| |
BGonline.org Forums is maintained by Stick with WebBBS 5.12.