Re: [gnugo-devel] 9x9 testcases

[Arend Bayer]

> 4 more test cases I had lying around.  I'm unsure about 9x9:430;
> perhaps someone else could offer an opinion.

This is
   A B C D E F G H J
 9 . . . . . . . . . 9
 8 . . X . . . . . . 8
 7 . . + . X . X . . 7
 6 . . O . . . . . . 6
 5 . . . . + . . . . 5
 4 . . . . . . . . . 4
 3 . O + X . . O . . 3
 2 . . . . . . . . . 2     WHITE (O) has captured 0 stones
 1 . . . . . . . . . 1     BLACK (X) has captured 0 stones
   A B C D E F G H J

I see many reasonable moves for White, so maybe we should skip this
test.

> +loadsgf games/boardspace/GNUGo-GoFigure0.1-200503181316.sgf 11
> +410 reg_genmove black
> +#? [H5|B5]

B5 is so much better than H5 that I have disallowed the latter.

> +#solve the endgame problem with one move, not 3 like in the game
> +loadsgf games/boardspace/GNUGo-GoFigure0.1-200503302359.sgf 41
> +440 reg_genmove black
> +#? [G7]

H7 is also fine.

I have applied this with the changes, so that it won't get lost. If I
have overlooked something, we can still correct it.

Arend