Re: [Bug-gnubg] Snowie 4 vs. GNU 0.13

[Joseph Heled]

Another interesting "analysis" one can do is look at all cases where the 
bots disagree (this should be taken both from GNU and SN 4.1 analysis), 
and roll it out. Here is one example from match 39, flagged as a serious 
blunder by gnu 2ply analysis. Rollout indicated an error.

-Joseph

---

Game number 1

Snowie4 (white, 2 pts) v gnubg (red, 6 pts) (Match to 7)

---

Move number 1: White to play 42

Pipcounts: Red = 129, White = 118

• Snowie4 moves 13/9, 6/4

#	Games	Move	Equity
1	R2592	22/18*/16	24.31%
		0.7%  21.9%  79.0%   % 21.0%   4.4%   0.7%
2	R2592	13/9 6/4	23.58% (-0.73%)
		0.3%  16.9%  77.0%   % 23.0%   2.8%   0.2%

 GNU Position ID ZluDARjYbhcCCA Match sAHqACAAMAAA

---

fibs2html 0.5(0.18-c5123), Tue Jun 10 11:34:25 2003