| |
BGonline.org Forums
43Z-64, data
Posted By: Nack Ballard In Response To: 43Z-64^36 RO (Claude Landry)
Date: Thursday, 2 December 2010, at 8:52 p.m.
Thanks for the ^36 rollout, Claude. Here is data for 43Z-64:
43Z-64
[R P30 H46 X48] ~5
[R P10 X41 H54] /20*5
[R P10] <62
[R P15 X32] ^36*3d[R X5 P21] /15
g[H P18] /5Key: R = Run (24/14), P = Point (8/2 6/2), X = hit-and-split (24/18 8/4*), H = Hit-and-down (13/7 8/4*). Margins are in thousandths.
Before the brackets, d = dmp, g = (gammon) go; else assume money. After the brackets is a bot symbol (~ Snowie, / GnuBG, < XG 3-ply, ^XG 4-ply) and the number of trials in thousands, rounded down. (An asterisk is a repeater, not necessary after the final number).
For example, "[R P15 X32] ^36*3" means that R is best and P is -.015 in an XG 3-ply rollout of 36000+ trials, and X is -.032 after 5000+ trials. For another example, "g[H P18] /5" means that at (gammon) go, H is best and P is -.018 in a GnuBG rollout after 5000+ trials.
Most likely, Snowie's big margin of .030 is due to high variance from only 5k trials; it will be extended.
Nack
Blue played R, best at most scores
Blue played H, best at gammon go
| |
BGonline.org Forums is maintained by Stick with WebBBS 5.12.