| |
BGonline.org Forums
Data for Opening 51
Posted By: Nack Ballard In Response To: Opening 51 at DMP 4 ply RO (neilkaz)
Date: Tuesday, 14 June 2011, at 7:33 a.m.
Thanks, Neil, for the "^46 rollout. Here is data for opening 51 at dmp:
d[S=$ U10] /46*15 (Stick)
d[$ S3 U15] <51*10 (David 36k, Stick 15k) repaired or not?
d[S $1 U13] ^46*31 (David) repaired or not?
d[S $3] >1 (Stick)
d[S $3 U12] "<101*10 (Neil)
d[S $1] "^46 (Neil)Key: S = Split (24/23 13/8), $ = Slot (13/8 6/5), U = Up (24/18). Margins are in thousandths. Before brackets, d = dmp. After brackets is a bot symbol ( / GnuBG, < XG1 3-ply, ^ XG1 4-ply, > XG1 5-ply, "< XG2 3-ply, "^ XG2 4-ply) and the number of trials in thousands, rounded down (an asterisk repeats a number). For example, d[S=$ U10] /46*15 means that at dmp, S and $ are equally best in a GnuBG rollout of 46000+ trials, and U is -.010 after 15000+ trials.
The XG1 4-ply rollout shows opening 51S winning 50.3% (though it may not be repaired), whereas the XG2 3-ply rollout has 49.96% and the XG2 4-ply rollout 49.93%. (FWIW, the XG1 5-ply rollout was apparently extended from 1k to 5k and observed to hold above 50% before the data was lost.)
Nack
| |
BGonline.org Forums is maintained by Stick with WebBBS 5.12.