[gnugo-devel] Fuseki strategy discussion

[SP LEE]

Hi,

The next situation is from the recent game on NNGS
(http://www.lysator.liu.se/~gunnar/gnugo/nngs/games/gnugo-3.5.7-teriath-
200406062058.sgf). Gnugo (white) is on the move.


   A B C D E F G H J K L M N O P Q R S T
19 . . . . . . . . . . . . . . . . . . . 19
18 . . . . . . . . . . . . . . . . . . . 18
17 . . . O X . X . O . . . . . . . . . . 17
16 . . O + O X . . . + . . . . . O . . . 16
15 . . . . . . . . . . . . . . . . . . . 15
14 . . . . . . . . . . . . . . . . . . . 14
13 . . . . . . . . . . . . . . . . . . . 13
12 . . . O . . . . . . . . . . . . . . . 12
11 . . . . . . . . . . . . . . . . X . . 11
10 . . . + . . . . . + . . . . . + . . . 10
 9 . . O . . . . . . . . . . . . . . . . 9
 8 . . . . . . . . . . . . . . . . . . . 8
 7 . . . . . . . . . . . . . . . . . . . 7
 6 . . O . . O . . . . . . . . . X . . . 6
 5 . . . . . . . . . . . . . . . . . . . 5
 4 . . X X . . . . . + . . . . . + . . . 4
 3 . . . . . . . . X . . . X . . X . . . 3
 2 . . . . . . . . . . . . . . . . . . . 2
 1 . . . . . . . . . . . . . . . . . . . 1
   A B C D E F G H J K L M N O P Q R S T

Top moves:
1. M17 24.11
2. J5  22.31
3. H6  20.74
4. J6  19.93
5. K7  19.59
6. H5  19.35
7. O15 17.99
8. J7  17.14
9. H7  17.09
10. Q14 16.74
Checking safety of a white move at M17
Move generation likes M17 with value 24.11
genmove() recommends M17 with value 24.11
= M17

Apparently the upper left three black stone is weak and J17 is
relatively safe. I would consider positions like H15, J15, E18 or E15,
even F18. However, gnugo played tenuki and went to some other area. From
the top moves we could also see all gnugo likes is the place which seem
to have big influence but does not have real big value.

I'm not sure if the attacking of the three stones is overlooked by
gnugo, or the virtual influence points are overvalued in gnugo. It seems
to me that gnugo doesn't spend much time in this kind of situation and
we might add some strategies without triggering performance issue. I had
a feeling that I could use the weakness calculations here to find a good
move, but I'm not sure. I'll be grateful if anyone can make some
comments about it.


SP Lee