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

Re: computer-go: Most simple Go rules




The programs should indicate their evaluation of the group status. If they agree, use that status. If
they disagree, have a strong player adjudicate, with the correct status.

Ing solved this by specifying that programs must remove all dead stones before passing.

-David

At 11:45 AM 6/25/2001 +0200, you wrote:
> __ Chinese (area) scoring.

I also support chinese scoring, but some weird situations
can happen with any scoring in computer Go.

In many computer vs computer games (at least the ones Golois plays),
some situations are unclear even when both programs passes.
For example, in my game againt Star of Poland in Fost97, a
large group of SOP was still unsettled after both programs had
passed, so we had to call the referee to decide the game (it
really decided the winner).
The same thing happened recently against AYA, and Nick Wedd
reported the same situation during MSO2000 computer Go.

In my game against SOP, the referee (a 9dan pro) decided seki,
so SOP won the game.

I am sure this can happen even to the best programs, as we can
find quite difficult life and death positions they cannot handle
well.

What do you think is the appropriate handling of such situations ?

Given that the position can be unsettled, unsettled by ko, live or
seki, dead or seki, and so on... with both programs failing to see
it.


Tristan.
David Fotland