[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [computer-go] End Game



In message <402B5CE9.4090507@xxxxxxxxxxxxxxxxx>, Robert Jasiek <jasiek@xxxxxxxxxxxxxxxxx> writes

Calculating how much secured territory you
have then got (given the status of all groups) looks possible, but I am not aware of any software which does it perfectly.
This is easy and has the low time complexity of flood
filling once the life and death status are determined,
see the Japanese 2003 Rules. Notes: You can apply those
rules also for area-scoring or stone-scoring, if
principle determination of life and death is all you
want. OC, you can simplify life and death in the rules
by omitting the useless traditional exceptions.
Not so easy, if we are using what we might call "vernacular Japanese rules", as implemented on Go servers and played in Britain. These rules allow you to score territory in a live group even if it is adjacent to dame.

. . . . # O . Complete 7x7 board
# # # # # O . Both players have passes consecutively.
O O # . # O .
. O O # O O .
O . O # # O .
O O O . . O O
# # # # # # #
. . . . . . .

All these stones are (correctly) marked as alive. There is no move which gains anything. The question is whether D6 is Black's territory. A good player recognises that the bamboo join is a guaranteed connection so D6 is territory. A flood-fill algorithm might not know this.

Nick
--
Nick Wedd nick@xxxxxxxxxxxxxxxxx
_______________________________________________
computer-go mailing list
computer-go@xxxxxxxxxxxxxxxxx
http://computer-go.org/mailman/listinfo/computer-go