bug-gnubg
[Top][All Lists]
Advanced

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

Re: [Bug-gnubg] Some positions in the benchmark database


From: David Montgomery
Subject: Re: [Bug-gnubg] Some positions in the benchmark database
Date: Wed, 20 Jun 2007 10:06:33 -0700

From: Achim Mueller <address@hidden>
>
>  +13-14-15-16-17-18------19-20-21-22-23-24-+
>  | O     X  O  O  O |   | O  X  X  X  O  O |
>  | O     X  O  O    |   | O  X  X  X     O |
>  |       X     O    |   |    X  X          |
>  |                  |   |    X  X          |
>  |                  |   |    X             |
> v|                  |BAR|                  |
>  |                  |   |                  |
>  |                  |   |                  |
>  |                  |   |                  |
>  | O                |   |                  |
>  | O                |   |    X             |
>  +12-11-10--9--8--7-------6--5--4--3--2--1-+
> Joseph-ID: ALOLMGAGAABAIAIDOPAN
> X to play 66.

[...]
My first choice would be something like 20/14(3), 21/15 to switch some
spares into o's outfield. In general I'd strive for a classical prime vs
prime position (as x), because of the three dead checker of o.

I would make the same play.  But I would be aiming at a distant
outside prime, from the 10-15 point, or further back if the 16 and 17
points become available.

>  +13-14-15-16-17-18------19-20-21-22-23-24-+
>  |             X    |   |          O  X    |
>  |                  |   |                  |
>  |                  |   |                  |
>  |                  |   |                  |
>  |                  |   |                  |
> v|                  |BAR|                  |
>  |                  |   |       6          |
>  |                  |   |    X  X  O  O    |
>  |                  |   |    X  X  O  O    |
>  |             O  O |   | X  X  X  O  O  O |
>  | X           O  O |   | X  X  X  O  O  O |
>  +12-11-10--9--8--7-------6--5--4--3--2--1-+
> Joseph-ID: AEAADGLMDHPILNIBCABA
> X to play 51.

x should not hit the checker on the 22-point. That checker is badly
placed at least for now. I'd either make the 12-point (17/12, 5/4) and
wait for the remaining checker coming nearer to home or move 12/6.

I'm really unsure about this.  My first thought was 12/6.
But then I decided that I actually did want to hit.  As O,
I would want to be diversifying my checkers and working
towards building up a blockade mostly in X's outer board.
By hitting as X, I prevent this, at least for now, which feels
like a good idea.  My goal is to come home quickly and
get as many checkers off as I can before O's offense (the
outside prime) is in place.  So I'll play 23/22* 17/12.  I don't
really want the 12 point, but I don't want to get hit, either.

I don't know if any bot plays these positions even plausibly well.
Up into the early 2000's JellyFish was still the strongest available
program for positions with a developed outside prime.  But 1) these
don't have the outside prime in place, so I don't know how good it
is; 2) JF wasn't that great, anyway -- the others were just horrible;
3) I haven't kept up with improvements to Snowie starting with v4,
so maybe it knows how to play these positions now.

David




reply via email to

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