Focardi and Luccio
DannyPurvis
Luccio. My guess is that I made some big errors. My critique is at
http://www.geocities.com/chessdp/focardi.htm. Unfortunately, the link
within the article to the Focardi / Luccio analysis is broken.
Dan Hoey
I saw your critique, and it seemed accurate. They arrived at the right
answer, but one of the steps was wrong.
Dan
Dan Hoey
Google found their paper at
http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.21.212
The problem is on page 13, "Strategy for x_0 = 7", case 1. They
claim that after 7+ 1(8)2 1(9)1[2,3] 4(10)9 the second player
should play 9(11)9[5,6], ensuring two survivors in the region
with spot 3 and three survivors outside that region. They claim
the region with spot 3 is a slight variant of a region with two
spots "but the strategy is exactly the same." However, in the
two-spot game, two survivors can be guaranteed by the second
player. In this game, the first player can play 2(12)2[3,8],
forcing one survivor.
Dan
Dan Hoey
It might be useful to look at just what happens after 7+ 1(8)2
1(9)[2,3] 4(10)9. If I recall correctly, the second player can
still win (by forcing an odd number of survivors). However, it
may be that the second player cannot force a _particular_ number
of survivors. I'd be interested in a simple example (ideally a
starting position) in which a player can win, but cannot force
the game to last a particular number of moves.
This would be a counterexample to the observation that the game
usually boils down to a battle over whether the number of survivors
is N or N+1.
Dan
lap...@tuxfamily.org
you should have a look to these positions (in Glop-notation) :
2AB.}AB.}]2AB.}AB.}]! for a normal game
1.2.2.}]2AB.}AB.}]! for a misere game.
--
Julien Lemoine
Dan Hoey
> Hi Dan,
>
> you should have a look to these positions (in Glop-notation) :
>
> 2AB.}AB.}]2AB.}AB.}]! for a normal game
> 1.2.2.}]2AB.}AB.}]! for a misere game.
>
addressed this sort of thing in their theory, while it is
fundamental to the strategy of nimbers *2 or greater.
Dan