Does moving the king from d8 to a2 help? Or is he needed on d8?
(I've not looked closely and I'm no good at problems anyway.)

Good thinking.
The king is certainly needed in its current place to avoid dual
solutions.
However, there is indeed a way to put a different white piece to a2
(not a pawn, not the King), provided you also introduce further changes
to the position, to save both lines of play as intended by the author.
There are other different ways to fix the problem which are "better",
though.

I hope this answers your question.
JP

wrote:
David Richerby wrote:
Does moving the king from d8 to a2 help? Or is he needed on d8?
(I've not looked closely and I'm no good at problems anyway.)

Good thinking.
The king is certainly needed in its current place to avoid dual
solutions.
However, there is indeed a way to put a different white piece to a2
(not a pawn, not the King), [...]

I hope this answers your question.

Yes, it does, and without giving anything much away to people who
might be interested in solving and/or fixing the problem themselves --
thanks.


Dave.

--
David Richerby Edible Electronic Hat (TM): it's like
www.chiark.greenend.org.uk/~davidr/ a hat but it uses electricity and you
can eat it!

There are actually several possible fixes, the first one I thought of
was legal but only retained one solution, the other retained both but
was an illegal position (yikes!!), and Jose found one that worked, but
was uneconomical. Finally, I found a simple, legal , economical fix.
Given how that fix is so simple - and given Howard's prowess as a
composer - I still wonder if the original problem was correct, and one
"bit" was simply left off the diagram in the book's reprint of the
problem. One of my UK friends is visiting the archive where all British
newspapers are kept, can't remember the name but its in Colindale.

That is one of the beauties of the computer - you can answer questions
quickly for yourself also like "what is that black bishop doing on h8"
by removing it and seeing the cooks that sprout up.

SBD wrote:
There are actually several possible fixes, the first one I thought of
was legal but only retained one solution, the other retained both but
was an illegal position (yikes!!), and Jose found one that worked, but
was uneconomical. Finally, I found a simple, legal , economical fix.
Given how that fix is so simple - and given Howard's prowess as a
composer - I still wonder if the original problem was correct, and one
"bit" was simply left off the diagram in the book's reprint of the
problem. One of my UK friends is visiting the archive where all British
newspapers are kept, can't remember the name but its in Colindale.

That would be the British Library. Their newspaper archive is in
Colindale.

That is one of the beauties of the computer - you can answer questions
quickly for yourself also like "what is that black bishop doing on h8"
by removing it and seeing the cooks that sprout up.

On 15 Nov 2006 07:37:50 -0800, "SBD" wrote:


There are actually several possible fixes, the first one I thought of
was legal but only retained one solution, the other retained both but
was an illegal position (yikes!!), and Jose found one that worked, but
was uneconomical. Finally, I found a simple, legal , economical fix.
Given how that fix is so simple - and given Howard's prowess as a
composer - I still wonder if the original problem was correct, and one
"bit" was simply left off the diagram in the book's reprint of the
problem.

I think the simpler explanation is that Howard wasn't very careful in
exploring possible cooks of his problems. Just one year later, he
published another also involving a rook masking a bishop following two
moves of a single black pawn:

8/7P/3pP3/1p1P3K/P2B2p1/5prp/5Npb/1R3nk1
White mates in five

Kenneth S. Howard
V. American Chess Bulletin
January, 1926

1. Ba7 bxa4 2. h8Q a3 3. Qa1 a2 4. Rb6 Kxf2 Rb2#
1...b4 2. h8R b3 3. Rb8 b2 4. Rb6 Kxf2 R6xb2#

Once again, it's cooked by a queen promotion, even after the original
had already been repaired due to some other serious flaw! After 1. h8Q
bxa4, 2. Ba7 transposes into the first variation above.

If 1...b4, the queen has three ways to reach the first rank for eventual
mate: 2. Q to h7, h6 or c8. 2...b3 3. Rb2 N~ 4. R/BxN Kf1 5. Q mates.

Source and analysis are the same is in the original post. This was #153
in the composer's The Enjoyment of Chess Problems. The preface of the
book contains the following sentence:

A "V" (version) before the name of the periodical indicates that since
its first publication the problem has been revised, either because it
was found to be unsound or to have some serious flaw.

Note that The Weekly Westminster was misspelled in my original post, and
the h-file was mis-designated.

Clifford Stern

8/7P/3pP3/1p1P3K/P2B2p1/5prp/5Npb/1R3nk1
White mates in five

Kenneth S. Howard
V. American Chess Bulletin
January, 1926

1. Ba7 bxa4 2. h8Q a3 3. Qa1 a2 4. Rb6 Kxf2 Rb2#
1...b4 2. h8R b3 3. Rb8 b2 4. Rb6 Kxf2 R6xb2#

Once again, it's cooked by a queen promotion, even after the original
had already been repaired due to some other serious flaw! After 1. h8Q
bxa4, 2. Ba7 transposes into the first variation above.

This one is also very easy to fix though, and again when you see the
new version you cannot avoid thinking of a typing error.
JP

Clifford Stern wrote:
I think the simpler explanation is that Howard wasn't very careful in
exploring possible cooks of his problems. Just one year later, he
published another also involving a rook masking a bishop following two
moves of a single black pawn:

8/7P/3pP3/1p1P3K/P2B2p1/5prp/5Npb/1R3nk1
White mates in five

Kenneth S. Howard
V. American Chess Bulletin
January, 1926

1. Ba7 bxa4 2. h8Q a3 3. Qa1 a2 4. Rb6 Kxf2 Rb2#
1...b4 2. h8R b3 3. Rb8 b2 4. Rb6 Kxf2 R6xb2#

Once again, it's cooked by a queen promotion, even after the original
had already been repaired due to some other serious flaw! After 1. h8Q
bxa4, 2. Ba7 transposes into the first variation above.

I can think of several different ways (at least 3) to fix the problem
and still save both lines of play as intended by the author. All of
them make the key move either less elegant ore more obvious. If I had
to choose one that would be:

1B6/7P/3p4/1p1K4/P5p1/5prp/5Npb/1R3nk1

Joaquim C.

Joaquim wrote:

1B6/7P/3p4/1p1K4/P5p1/5prp/5Npb/1R3nk1

Ugh. The immediate battery formation and removal of a flight
square/capture don't appeal. The key, especially for the time, was a
major portion of the problem.

(And you know, Joaquim, normally I am a fan!!!)

But I suspect you knew that the fix was a "patch job," anyway. It's
still interesting and fun, isn't it?

Ugh. The immediate battery formation and removal of a flight
square/capture don't appeal.

Yes, I share your view. An alternative possibility, in which the need
to move the bishop precisely to a7 is less obvious might be:
8/7P/3p4/1pBK4/P5p1/5prp/5Npb/1R3nk1
again an atrocious key move as the bishop escapes not only from threat
but also from liberating the black pawn.

the key, especially for the time, was a major portion of the problem.

and I must admit then that I cannot envisage any easy way to fix it!

Joaquim

I knew you knew that.

From that, I played around and made the following little chess *puzzle*
from the position. It's good to show those who would want to see the
difference between problem and puzzle - it's good chess, but not
especially aesthetic:

FEN: Q7/p7/3p4/1p4K1/P5p1/4Bprp/5Npb/1R3nk1 w - - 0 1

White to play and mate in 5.

Of course, anyone who has been following this will see the key to the
puzzle.