Re: [Bug-gnubg] Problems with latest Windows Build

[Joseph Heled]

This is the way to report problems - with an explicit example.

Good news found a bug (order of crashed and race nets was reversed) - making 
crashed /race prune results complete nonsense.

I am submitting a fix now.

-Joseph


Theo Hermsen wrote:
> Hi, 
> her is an example for a big pruning mistake
>
> regards 
>
> Theo
>
>
> ------------------------------------------------------------------------
>
> The score (after 4 games) is: Piktout 1, RedTop 5 (match to 15 points)
> Match Information:
>
> Piktout's rating: 1930.81 (Exp 2645)
> RedTop's rating: 1879.84 (Exp 5597)
> Date: Oktober 31, 2004
> Place: GamesGrid Games Server
>
> Move number 35: X to play 51
>
>  GNU Backgammon  Position ID: mK+FBQL7uwEAQA
>                  Match ID   : QYnmARAAKAAA
>  +13-14-15-16-17-18------19-20-21-22-23-24-+     O: Piktout (Cube: 2)
>  |          O  O  O |   | O     O          |     1 point
>  |             O    |   | O     O          |     
>  |                  |   | O                |     
>  |                  |   | O                |     
>  |                  |   | O                |    
> v|                  |BAR|                  |     15 point match
>  |                  |   |             7    |    
>  |                  |   |             X    |     
>  |                  |   |          X  X    |     
>  | O                |   |       X  X  X  X |     Rolled 51
>  | O  O             | X |    O  X  X  X  X |     5 points
>  +12-11-10--9--8--7-------6--5--4--3--2--1-+     X: RedTop
> Pip counts: O 130, X 58
>
> * RedTop moves bar/20 2/1
>
> Rolled 51 (-0.282):
> *    1. Cubeful 2-ply    bar/20 2/1                   Eq.:  -0.236
>        0.452 0.182 0.001 - 0.548 0.034 0.000
>      2. Cubeful 2-ply    bar/20 3/2                   Eq.:  -0.317 ( -0.081)
>        0.423 0.187 0.001 - 0.577 0.042 0.000
>      3. Cubeful 0-ply    bar/20 4/3                   Eq.:  -0.456 ( -0.220)
>        0.367 0.154 0.000 - 0.633 0.033 0.000
>
>
> Annotation:
> KitWoolsey: not close to a recube -- not clear who is favorite
>
> Output generated Thu Nov 04 14:28:07 2004
> by GNU Backgammon 0.14-devel (Text Export version 1.68)
>
>
>
> ------------------------------------------------------------------------
>
> The score (after 4 games) is: Piktout 1, RedTop 5 (match to 15 points)
> Move number 35: X on roll, cube decision?
>
>  GNU Backgammon  Position ID: mK+FBQL7uwEAQA
>                  Match ID   : QQngARAAKAAA
>  +13-14-15-16-17-18------19-20-21-22-23-24-+     O: Piktout (Cube: 2)
>  |          O  O  O |   | O     O          |     1 point
>  |             O    |   | O     O          |     
>  |                  |   | O                |     
>  |                  |   | O                |     
>  |                  |   | O                |    
> v|                  |BAR|                  |     15 point match
>  |                  |   |             7    |    
>  |                  |   |             X    |     
>  |                  |   |          X  X    |     
>  | O                |   |       X  X  X  X |     On roll
>  | O  O             | X |    O  X  X  X  X |     5 points
>  +12-11-10--9--8--7-------6--5--4--3--2--1-+     X: RedTop
> Pip counts: O 130, X 58
>
> * RedTop moves bar/20 2/1
> Alert: very bad move ( -0.621)
>
> Rolled 51 (-0.282):
>      1. Cubeful 2-ply    bar/20 3/2                   Eq.:  +0.384
>        0.661 0.253 0.000 - 0.339 0.025 0.000
> *    2. Cubeful 2-ply    bar/20 2/1                   Eq.:  -0.236 ( -0.621)
>        0.452 0.182 0.001 - 0.548 0.034 0.000
>      3. Cubeful 2-ply    bar/20 4/3                   Eq.:  -0.475 ( -0.859)
>        0.382 0.145 0.001 - 0.618 0.047 0.000
>
>
> Annotation:
> KitWoolsey: not close to a recube -- not clear who is favorite
>
> Output generated Thu Nov 04 13:28:53 2004
> by GNU Backgammon 0.14-devel (Text Export version 1.69)