From: Colin Paul Adams <colin <at> COLINA <dot> DEMON <dot> CO <dot> UK>
Date: 23 sep 1996
Subject: CHU SHOGI: problem with mating problem

Problem number 17 in the 1675 document (A17 in MSM)
gave me some difficulty. But when I finally looked at the
solution in MSM, I thought that White's defence was not
the best possible.

For those of you without access to MSM, the problem in
modified Forsyth Notation is as follows:

3,S,1,+g,2,+dk,3 / 3,C,1,DE,6 / 3,G,VM,K,2,+DK,SM,2 /
+dh,1,FK,2,P,3,BT,ln,1 / 1,P,P,P,1,B,4,P,1 / SM,2,+RC,P,1,C,P,4 /
8,GB,3 / 1,fk,9,P / +R,5,+R,5 / 5,ph,5,r / 12 / 12 /

The main line of the solution given in MSM runs as follows:

1. +Gx7b, Kx7b (1. ..., K8d is given as a side line);
2. FKx1h, K-8b (2. ..., BT-4e also looks possible,
but transposes, more or less, after 3. FK-1b);
3. FK-1b, K-7c; (this is the move with which I quarrel,
the continuation given is fairly straight-forward:
 4. +DK-5a, K-6d; 5. FK-6b, +DK-6c; 6. Lnx3d-4e mate.)

However, it seems to me that White can put up a much
stronger defence by 3. ..., SM-3b ;

I can find nothing else but:

4. +DKx3b, +DKx3b; 5. FKx3b, K7c;

Now 6. FK-6a appears necessary to stop the king escaping
via 8d. Then 6. ..., Kx6a ; and for a long time I couldn't
find the Lion mate. But eventually I found:

7. Lnx3d-4e, K-7c; 8. Ln-5e (or 5d), K-8d;
9. Ln-7f, K-7c; 10. R-1c+ (or =), K retreats;
11. Lnx7e-6d, and Lion mate next move.

Could people please check my analysis, and say if you agree
that this should indeed be considered the strongest defence?
If people agree, I will send it to George Hodges for inclusion
in the next supplement to MSM.

Colin Adams