BGonline.org Forums

51S-32, data

Posted By: Nack Ballard
Date: Friday, 28 May 2010, at 8:29 p.m.

In Response To: 51S-32 XG 4 ply RO (neilkaz)

51S-32

Thanks for the RO, Neil.

Here is data for 51S-32:

Money
Snowie .. [S D2 Z17 T20 U25 $66 %96] 15k 5k$%
GnuBG .. [D S2.4 T15 Z19 U38] 46k 15kT 5k
XG 3-ply [D S1] 15k
XG 4-ply [S D3] 46k

Other scores
GnuBG .. d[D S2.3 Z5.2] 5k
GnuBG .. g[D $12.3 S22] 5k

Key: S = Split (24/21 13/11), D = Down (13/11 13/10), Z = reverse split (24/22 13/10), T = sTack/Tower (13/8), U = Up (24/22 24/21), $ = Slot (13/10 6/4), % = alt slot (13/11 8/5). Error sizes are in thousandths.

A lower-case letter before the brackets denotes score: d = dmp, g = gammon go; else assume money. After the brackets is the number of trials (in thousands, rounded down). For example, "d[D S2.3 Z5.2] 5k" means that at dmp, D is best, S is -.0023 and Z is -.0052 after 5000+ trials.

In short, the bot aggregate seems to have S and D tied (perhaps S better by a whisker). At gammon go, D is definitely best, and dmp probably sides with D as well. There is no rollout for gammon save, but logically S must be best, and I'd guess D and T to be about .02 back.

Nack

Messages In This Thread

• 51S-32 XG 4 ply RO
  neilkaz -- Friday, 28 May 2010, at 12:28 p.m.
  • 51S-32 XG 4 ply RO
    David Rockwell -- Friday, 28 May 2010, at 3:01 p.m.
    • x1S-32 and x1$-32, Gnu's tendencies towards D and $
      Nack Ballard -- Friday, 28 May 2010, at 9:10 p.m.
  • 51S-32, data
    Nack Ballard -- Friday, 28 May 2010, at 8:29 p.m.
    • 51S-32 at GS
      David Rockwell -- Friday, 28 May 2010, at 9:16 p.m.
      • 51S-32 at GS
        Nack Ballard -- Saturday, 29 May 2010, at 12:03 a.m.
        • 51S-32 at GS
          David Rockwell -- Saturday, 29 May 2010, at 1:18 a.m.
          • 51S-32 at GS
            Nack Ballard -- Saturday, 29 May 2010, at 1:51 a.m.