BGonline.org Forums

Rollouts of this unusual 51

Posted By: neilkaz
Date: Monday, 21 September 2009, at 11:57 a.m.

In Response To: Comparing XG Roller to GNU 4 ply match analysis (neilkaz)

I cut my r/o sizes to 1296 so that it will be quicker to finish this project. GNU 4 ply's clear pref to hit both will cost it dearly in the competition vs XG Roller.

The score (after 7 games) is: Maddy 8, neilkaz 5 (match to 15 points)

Move number 54: neilkaz to play 51

Maddy	102
neilkaz	53
Position ID: t1sEgCJ8DBgAAA Match ID: Q4nmAYAAKAAA

• neilkaz moves 13/12 13/8

#		Ply	Move	Equity
•	1	R	13/12 13/8	+0.1335
			0.5336 0.1798 0.0543 - 0.4664 0.0000 0.0000 +0.1335 +0.1335 0.0022 0.0020 0.0043 - 0.0022 0.0000 0.0000 0.0046 0.0046
			Full cubeful rollout with var.redn.
			1296 games, Mersenne Twister dice gen. with seed 865549781 and quasi-random dice
			Play: world class 2-ply cubeful prune [world class]
			keep the first 0 0-ply moves and up to 8 more moves within equity 0.16
			Skip pruning for 1-ply moves.
			Cube: 2-ply cubeful prune [world class]
	2	R	6/5* 6/1*	+0.1049 ( -0.0286)
			0.5192 0.1804 0.0278 - 0.4808 0.0000 0.0000 +0.1049 +0.1049 0.0029 0.0020 0.0027 - 0.0029 0.0000 0.0000 0.0060 0.0060
			Full cubeful rollout with var.redn.
			1296 games, Mersenne Twister dice gen. with seed 865549781 and quasi-random dice
			Play: world class 2-ply cubeful prune [world class]
			keep the first 0 0-ply moves and up to 8 more moves within equity 0.16
			Skip pruning for 1-ply moves.
			Cube: 2-ply cubeful prune [world class]
	3	0	13/7	+0.0757 ( -0.0578)
			0.5060 0.1726 0.0124 - 0.4940 0.0000 0.0000
	4	0	6/1* 3/2	-0.0168 ( -0.1503)
			0.4617 0.1621 0.0089 - 0.5383 0.0000 0.0000
	5	0	13/8 3/2	-0.0223 ( -0.1558)
			0.4559 0.1788 0.0133 - 0.5441 0.0000 0.0000

is Maddy score: 8 pip: 102

15 point match

6

pip: 53 score: 5 is neilkaz

XGID=-a-EaaB------B---a--abcbc-:3:-1:1:51:5:8:0:15:10

to play 51

1.	Rollout1	13/12 13/8	eq: +0.142
	Player: Opponent: 53.56% (G:19.20% B:4.89%) 46.44% (G:0.00% B:0.00%) Conf: ± 0.013 (+0.129...+0.155) Duration: 10 minutes 09 seconds
2.	Rollout1	6/5* 6/1*	eq: +0.110 (-0.032)
	Player: Opponent: 51.99% (G:19.03% B:2.93%) 48.01% (G:0.00% B:0.00%) Conf: ± 0.015 (+0.095...+0.125) Duration: 13 minutes 14 seconds
3.	2 ply	13/7	eq: -0.021 (-0.163)
	Player: Opponent: 46.68% (G:12.20% B:0.91%) 53.32% (G:0.00% B:0.00%)
4.	2 ply	13/8 3/2	eq: -0.139 (-0.280)
	Player: Opponent: 40.94% (G:11.55% B:0.96%) 59.06% (G:0.00% B:0.00%)
5.	2 ply	6/1* 3/2	eq: -0.180 (-0.322)
	Player: Opponent: 39.12% (G:10.13% B:0.36%) 60.88% (G:0.00% B:0.00%)
1 1296 Games rolled with Variance Reduction. Moves and cube decisions: 3 ply

Version: 1.03, MET: GnuBG 11 point

Messages In This Thread

• Comparing XG Roller to GNU 4 ply match analysis
  neilkaz -- Monday, 21 September 2009, at 12:22 a.m.
  • Comparing XG Roller to GNU 4 ply match analysis
    Rich Munitz -- Monday, 21 September 2009, at 1:42 a.m.
    • Comparing XG Roller to GNU 4 ply match analysis
      neilkaz -- Monday, 21 September 2009, at 1:59 a.m.
  • Cube play against a weaker Opponent
    Michael Petch -- Monday, 21 September 2009, at 1:54 a.m.
    • Cube play against a weaker Opponent
      neilkaz -- Monday, 21 September 2009, at 2:09 a.m.
      • Oops/// 8 cube was 0.16 error to take *NM*
        neilkaz -- Monday, 21 September 2009, at 2:11 a.m.
      • Cube play against a weaker Opponent
        Michael Petch -- Monday, 21 September 2009, at 2:19 a.m.
        • My cube to 8 using J/T 100 elo met
          neilkaz -- Monday, 21 September 2009, at 3:07 a.m.
      • Cube play against a weaker Opponent
        Phil Simborg -- Monday, 21 September 2009, at 4:06 p.m.
  • XG export shows MET used at bottom
    neilkaz -- Monday, 21 September 2009, at 2:14 a.m.
  • Rollouts of this unusual 51
    neilkaz -- Monday, 21 September 2009, at 11:57 a.m.
  • Maybe GNU-bg has learned about QF *NM*
    Chuck Bower -- Monday, 21 September 2009, at 4:54 p.m.