| |
BGonline.org Forums
31P-11, data
Posted By: Nack Ballard In Response To: 31P-11 3-ply (David Rockwell)
Date: Tuesday, 20 July 2010, at 1:00 a.m.
Perhaps someone can inform me if I am overly concerned about parallellism, but I've been operating under the assumption it is preferable not to use the same seed for two (same-bot) rollouts of the same position even when a different ply setting is used. (In any case, I don't see how it can possibly gain.)
31P-11
That said, thanks for the long rollouts, David.
Here is data for 31P-11:
Money
Snowie .. [e N9] 38k
GnuBG ..[e N6] 46k
XG 3-ply [e=N] 62k
XG 4-ply [e N3] 46kOther scores (GnuBG):
d[N e5] 18k
s[e N14] 5k
g[N e20] 5kKey: e = each (24/22 6/5(2)), N = Near (8/7(2) 6/5(2). Error sizes are in thousandths. Before brackets, d = dmp, s = (gammon) save, g = (gammon) go; else assume money. After brackets is the number of trials, in thousands (k) rounded down. For example, "d[N e5] 18k" means that at dmp, N is best and e is -.005 after 18000+ trials.
Nack
| |
BGonline.org Forums is maintained by Stick with WebBBS 5.12.