From: Steve Evans <trout <at> NETSPACE <dot> NET <dot> AU>
Date: 18 feb 1999
Subject: Chu Mating Problem Solution

I have now used my Zillions-of-Games based Chu Shogi mating problem
program to tackle a couple more of the ancient problems for which
solutions have never been found, and have met with some success!

The problems I looked at were a couple from the "D" document referred to
in George Hodges' Middle Shogi Manual. While the author of these
problems is unknown, David Murphy has a very plausible theory
attributing them to Ito Kanju, perhaps the greatest Tsume Shogi master
of them all. David's theory will be included in a future supplement to
Middle Shogi Manual.

With a lot of hard work George and David have recently found solutions
to a number of these "D" document problems. Until recently, efforts in
Japan and the West had failed to find a solution to a single one of
them.

The first success I've had with my program has been with the problem
given as D84 in the MSM.

In amended Forsyth notation it is as follows:

1,+sm,7,ph,2 / 8,G,B,K,1 / +dh,8,LN,C,+c / 11,KY / 7,P,1,S,1,l /
10,+dk,R / 11,FK / 12 / 9,+dk,2 / 12 / 11,dh  / 12 /

The solution the program found is:

1. +DHx3c, Gx3c;
2. +Cx2c , Gx2c  (if Kyx2c 3. Ph-1c+ mate) (if Bx2c 3. Ph-2a+, K-1c;
+DKx3e mate);
3. Phx3b+,  Kx3b (if K-1c  4. +SM-1a and mate next move);
4. +DKx3e-4d,  FKx3i;
5. +SM-4a mate.

If White plays 1. ....  Kx3c it goes:

2. +DK3ix3e,  K-4c (if K-2b, 3. +DK-4d and mate is swift);.
3. +DK-4e, K-5b;
4. DH-10b+ mate.

Many thanks to Colin Adams for assisting with verification and providing
some of the shorter lines (in brackets) above.

Having now achieved some success, I'm looking forward to tackling some
of the other unsolved problems in the "D" document. :-)


Steve Evans