[ View Thread ] [ Post Response ] [ Return to Index ] [ Read Prev Msg ] [ Read Next Msg ]

BGonline.org Forums

XG 4-ply RO 21S-63 just finished

Posted By: Nack Ballard
Date: Sunday, 2 May 2010, at 7:16 p.m.

In Response To: XG 4-ply RO 21S-63 just finished (neilkaz)

Thanks, Chase and Neil, for the new rollouts. Here is data for 21S-63:

Snowie .. [S R3 U29 X54] 31k 20kU 5k
GnuBG .. [S R5 X53] 46k 5kX, d[S R2.2] 10k, s[R S0.3] 15k
XG 3-ply [S R1] 82k, s[R=S] 10k
XG 4-ply [S R2] 46k

Nactation key: S = Split (24/18 13/10), R = Run (24/15), U = Up (24/21 24/18), and X = 24/21 8/2*). Error sizes of plays are in thousandths. If there is a "d" before the brackets, it means DMP, or if there is an "s" it means GS; otherwise assume money. The number of trials (in thousands, rounded down) appears after the brackets.

For example, "d[S R2.2] 10k" means that at DMP the best play is S, and R is -.0022 after 10000+ trials.


Hats off to Xavier for all his improvements to XG. I see that his latest version (1.14) is displaying the seed number in the output even when it is randomly generated.

I'm still unclear where these seed numbers are reproducible. That is, if you re-roll a position out with the identical seed number, will/should you get the identical result, as you will/should with GnuBG? Xavier?

Nack

Messages In This Thread

 

Post Response

Your Name:
Your E-Mail Address:
Subject:
Message:

If necessary, enter your password below:

Password:

 

 

[ View Thread ] [ Post Response ] [ Return to Index ] [ Read Prev Msg ] [ Read Next Msg ]

BGonline.org Forums is maintained by Stick with WebBBS 5.12.