Re: [Bug-gnubg] RE 22 OCT BUILD

[Joern Thyssen]

On Sun, Nov 03, 2002 at 07:00:41AM -0500, address@hidden wrote
> POSID a94MYAC7zYABBg
> MatchID  MIEJAAAAAAAA
> 
> I get this:
>     1. Cubeful 2-ply    21/16                        Eq.:  -0.613       0.364 
> 0.076 0.001 - 0.636 0.230 0.004        2-ply cubeful 100% speed, 8 cand., 
> 0.16 tol. [world class]    2. Cubeful 2-ply    6/4 6/3                      
> Eq.:  -0.665 ( -0.053)       0.336 0.024 0.000 - 0.664 0.141 0.001        
> 2-ply cubeful 100% speed, 8 cand., 0.16 tol. [world class]    3. Cubeful 
> 2-ply    6/1                          Eq.:  -0.677 ( -0.064)       0.334 
> 0.026 0.000 - 0.666 0.136 0.002        2-ply cubeful 100% speed, 0 cand., 0 
> tol.    4. Cubeful 2-ply    13/11 13/10                  Eq.:  -0.786 ( 
> -0.174)       0.310 0.067 0.001 - 0.690 0.182 0.004        2-ply cubeful 100% 
> speed, 0 cand., 0 tol.

This is result when the Jacoby rule is enabled.

> The other guy says this:
> I'm using build 021022 (Windows). 
> 
> If I set the evaluation chequer play settings to "world class," I get: 
> 
> 1. Cubeful 2-ply 21/16 Eq.: -0.725 
> 
> 0.362 0.076 0.001 - 0.638 0.229 0.004 
> 
> 2-ply cubeful 100% speed, 8 cand., 0.16 tol. [world class] 
> 
> 2. Cubeful 0-ply 6/1 Eq.: -0.787 ( -0.061) 
> 
> 0.299 0.026 0.000 - 0.701 0.118 0.001 
> 
> 0-ply cubeful [expert] 
> 
> 3. Cubeful 0-ply 13/11 13/10 Eq.: -0.792 ( -0.067) 
> 
> 0.311 0.045 0.000 - 0.689 0.165 0.004 
> 
> 0-ply cubeful [expert] 
> 
> 4. Cubeful 0-ply 6/4 6/3 Eq.: -0.868 ( -0.143) 
> 
> 0.286 0.021 0.000 - 0.714 0.124 0.001 
> 
> 0-ply cubeful [expert] 
> 
> Notice that only the top play is actually getting evaluated at 2-ply. 
> 
> If I use "world class+" (which increases the search space and tolerance to 
> 16/1.0) instead, I get: 
> 
> 1. Cubeful 2-ply 6/1 Eq.: -0.682 
> 
> 0.333 0.026 0.000 - 0.667 0.132 0.002 
> 
> 2-ply cubeful 100% speed, 16 cand., 1 tol. [world class++] 
> 
> 2. Cubeful 2-ply 6/4 6/3 Eq.: -0.689 ( -0.006) 
> 
> 0.334 0.023 0.000 - 0.666 0.136 0.001 
> 
> 2-ply cubeful 100% speed, 16 cand., 1 tol. [world class++] 
> 
> 3. Cubeful 2-ply 21/16 Eq.: -0.725 ( -0.043) 
> 
> 0.362 0.076 0.001 - 0.638 0.229 0.004 
> 
> 2-ply cubeful 100% speed, 16 cand., 1 tol. [world class++] 
> 
> 4. Cubeful 2-ply 13/11 13/10 Eq.: -0.779 ( -0.097) 
> 
> 0.310 0.067 0.001 - 0.690 0.179 0.004 
> 
> 2-ply cubeful 100% speed, 16 cand., 1 tol. [world class++] 

This is result when the Jacoby rule is enabled.

> 
> I noticed the W G BG breakdown is slightly different. Shouldn't they be the 
> same if the build hasn't changed? The equities are way off so I'm guessing 
> our settings somewhere are not the same?

Yes, one of you is playing with Jacoby rule and the other without.

Jørn

-- 
Joern Thyssen, PhD
Vendsysselgade 3, 3., DK-9000 Aalborg, Denmark
+45 9813 2791 (private) / +45 2818 0183 (mobile) / +45 9633 7036 (work)
Note: new mobile number!