gnugo-devel
[Top][All Lists]
Advanced

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

Re: [gnugo-devel] regress endgame:603


From: Evan Berggren Daniel
Subject: Re: [gnugo-devel] regress endgame:603
Date: Fri, 11 Apr 2003 16:01:39 -0400 (EDT)

One solution to this whole problem would be to add a GTP command
environmental_genmove.  The command would take the color to move and the
ambient temperature as parameters, and play PASS if no move can be found
more valuable than the temperature.  endgame:603 could then become a
test with only the one move on the board, written as

environmental_genmove black 5
#? [C1]

and optionally

environmental_genmove black 7
#? [PASS]

This would provide a way to tune endgame values individually.  In
positions with multiple moves, the environmental parameter could be used
to mean either don't run the (future) CGT engine, or run it for the
corresponding chilled game.

Evan Daniel


On Fri, 11 Apr 2003, Arend Bayer wrote:

>
>
> Dan wrote:
>
> > > But the drawback is that e.g. endgame:603 will do nothing but producing
> > > accidental FAILs and PASSes for (I guess) at least another year or so.
> > > (Until someone starts to write a CGT-like or lookahead endgame module.)
> >
> > That doesn't seem too serious. It only takes about a second to do this
> > problem since it's at the end of the game.
>
> It"s not the second, but about the PASSes and FAILs that are annoying.
>
> Arend
>
>
>
> _______________________________________________
> gnugo-devel mailing list
> address@hidden
> http://mail.gnu.org/mailman/listinfo/gnugo-devel
>
>




reply via email to

[Prev in Thread] Current Thread [Next in Thread]