Spend your weekend on this chess problem

[Paul Bonham]

During playtesting for my new online chess variant, this position came up and on further study, it strikes us as one of the more incredible chess problems we've ever seen.



It is White to move, and without using a computer, you must find a mate in 11 moves for White, who must immediately counter Black's threat of ...Qg1#.

There are also longer mates for White, but the 11 move one is the shortest one. I will post the line in a few days, unless someone finds it and posts it in the interim.

I should tell you that if you discover the line, you might be amazed at the profoundness of White's 7th and 8th moves. Again, other mates are there to be had, but these two non-obvious moves lead to the SHORTEST mate. What is especially profound is that White must calculate that his 8th move gives Black a free chance to do something besides get out of check -- with which Black can do nothing! A very hard-to-fathom idea, one that most players would dismiss without the proper exploration.

But then again, I'm not a regular chess player. If Tom O'Donnell finds the line and says White's 7th and 8th moves are remarkable, that will really mean something.

A further note: this is obviously a contrived position, like most chess problems. It could never occur in standard chess. Black has promoted a pawn to a Queen and is up a Queen, yet Black's two Queens have somehow removed themselves from the vicinity of White's lonely exposed King. Not only that, but both White and Black have pawns on the 7th rank, on the same file!

[Mathieu Cloutier]

Re: Spend your weekend on this chess problem

Ha ha! Another day, another trolling attempt?

Had a look with my old brain. Found a pretty straightforward mate starting with 1.c8=Q+. From there it's all checks until the black king is squeezed. However, it took me something like 13 moves.

So I thought, OK, maybe there is something shorter. But given Paul's history, not sure I want to spend too much time on that. So I fired up the engine. It does find a mate in 11, but...

SPOILERS ALERT!!!

It's all checks to the black king until mate. Nothing to see here, to be honest. Moves 7 and 8 are nothing remarkable. So maybe Paul hallucinated another nice solution, but it's probably cooked. I won't even bother to try and find this amazing variation with 'remarkable' moves.

[Paul Bonham]

Re: Spend your weekend on this chess problem

Ha ha! Another day, another trolling attempt?

Had a look with my old brain. Found a pretty straightforward mate starting with 1.c8=Q+. From there it's all checks until the black king is squeezed. However, it took me something like 13 moves.

So I thought, OK, maybe there is something shorter. But given Paul's history, not sure I want to spend too much time on that. So I fired up the engine. It does find a mate in 11, but...

SPOILERS ALERT!!!

It's all checks to the black king until mate. Nothing to see here, to be honest. Moves 7 and 8 are nothing remarkable. So maybe Paul hallucinated another nice solution, but it's probably cooked. I won't even bother to try and find this amazing variation with 'remarkable' moves.

Ok, there is an 11-move mate that does involve constant checks of the Black King. So let's just reword the problem:

Find a mate in 11 for White in which one of White's moves (I've already given that away, it's his 8th move) is not a check of the Black King.

Cloutier had to give up and fire up his engine... FAIL! And that was just to find the mate consisting of all checks. He cops out with the excuse "given Paul's history...". Exactly what history is that with respect to chess problems?

Let's see if better minds can solve the problem as it is now worded.

[Mathieu Cloutier]

Re: Spend your weekend on this chess problem

Cloutier had to give up and fire up his engine... FAIL!

I always feel bad when I fire up an engine to solve a problem. But given that I had you to reword your problem, I wouldn't say its a complete FAIL.

Oh, and your 'new' problem is completely uninteresting, IMO. I won't even try.

[Paul Bonham]

Re: Spend your weekend on this chess problem

I always feel bad when I fire up an engine to solve a problem. But given that I had you to reword your problem, I wouldn't say its a complete FAIL.

Oh, and your 'new' problem is completely uninteresting, IMO. I won't even try.

I thank you for having me reword my problem. The problem is not diminished, it is enhanced. And by the way, White's 9th move is also not a check of the Black King.

Saying this as someone who is far from an expert at chess problems: Perhaps there should be a category of chess problems in which at least one of the moves leading to mate should NOT be a check of the opposing King. Maybe the majority of chess problems already fall into that category? I'll leave it to someone far more knowledgeable of chess problems than I to comment on that.

As to Mathieu's lack of interest:

"Rejoice in your uniqueness
And consider every weakness
Something special of your own"

[Mathieu Cloutier]

Re: Spend your weekend on this chess problem

My lack of interest stems from the fact that your original problem was poorly written, innacurate and actually contained an alternate solution. So why should I spend (waste) some time on the 'new' version. For all we know, you hallucinated that solution and it's also cooked.

Your solution has to be unique, and should also represent best play by both sides.

[Paul Bonham]

Re: Spend your weekend on this chess problem

My lack of interest stems from the fact that your original problem was poorly written, innacurate and actually contained an alternate solution. So why should I spend (waste) some time on the 'new' version. For all we know, you hallucinated that solution and it's also cooked.

Your solution has to be unique, and should also represent best play by both sides.

You are responding as if I am representing myself as some sort of expert chess problem composer, and you are having your "aha!" moment because you found an alternate solution. But the only things I represented as fact was that (1) this problem came up randomly, the chances of it already existing are too small to be considered (you do remember statistics and probability, don't you Mathieu? :p), and (2) the shortest mate possible for White was 11 moves. That means with best play from Black -- there are shorter mates if Black plays less-than-optimal moves.

If there was only supposed to be ONE 11-move mate, sorry, I didn't actually compose this problem or "cook" it, whatever that means. So there's a second 11-move mate, not as interesting as the one we found. That enabled me to enhance the problem by stipulating that the solution had to contain at least one move in which White does not check the Black King. Perhaps a whole new class of problems for composers to concentrate on, so now a much simpler mate-in-4 can be something like a "mate in 4 MAC" (Missing A Check).

So anyone who finds an 11-move mate for White that is all checks, that is the alternate solution. There is an 11-move mate where Black is given one (actually, two) chances to make a move where his King is not in check.

Mathieu, you do not have to spend (waste) any time at all, you are free to do as you please, and frankly, Scarlet, I don't give a damn. But this problem IS unique (having been generated randomly) and the solution DOES represent best play by both sides. I do believe I have seen problems in which the final one or two moves can create more than one mating position.

But yes, the problem was lacking the wording about the solution having one non-checking move, and I thank Mathieu for discovering that.

I'll post the solution as a pgn file tomorrow night.

[Paul Bonham]

Re: Spend your weekend on this chess problem

Someone just said I should mention that although there are two 11-move mates, they diverge on White's 6th move. So you still have to get the first 5 moves right, and then look for the branch in which one of White's moves is not a check.

[Mathieu Cloutier]

Re: Spend your weekend on this chess problem

Someone just said I should mention that although there are two 11-move mates, they diverge on White's 6th move. So you still have to get the first 5 moves right, and then look for the branch in which one of White's moves is not a check.

See there? I knew you would have to add some more 'instructions'.

What will it be next? The 7th move by white has to be a check by a knight landing on a light squre on an odd file in the lower right part of the board?

Yay! A whole new class of chess problems where the goal is to find WTH was going in the head of the composer!

[Paul Bonham]

Re: Spend your weekend on this chess problem

See there? I knew you would have to add some more 'instructions'.

What will it be next? The 7th move by white has to be a check by a knight landing on a light squre on an odd file in the lower right part of the board?

Yay! A whole new class of chess problems where the goal is to find WTH was going in the head of the composer!

Wow, you can't get anything right. I know you're desperate to get back at me for putting egg on your face on that probability thread, but anyone can see that here, you are just grasping at straws that only you can see.

What I just added was not an "instruction". It was a clarification, because in telling everyone that there's an alternate 11-move mate, you didn't make it clear that it still starts with the same 5 moves. So anyone could have thought they were two totally distinct lines, starting with the first move. That's not the case.

And that's the last response you're getting from me, and the only reason I even bothered with it was so readers could be clear there were no more "instructions".

It's just find an 11-move mate for White in which at least one move for White is not a check of the Black King. That's it, plain and simple. Go ahead, try and make more of it. Those with minds to reason will not listen to you.

[Mathieu Cloutier]

Re: Spend your weekend on this chess problem

I'm quite sure there's still more than one solution to your 'new and improved' problem. But we'd have to see your solution first...

EDIT: in fact, I think there's probably a hole in your solution.

[Louis Morin]

Re : Re: Spend your weekend on this chess problem

As far as problemists are concerned, two importants rules are:

1) a good problem should have one and only one solution (if there are two solutions or more, the problem is "cooked" and of little value)

2) a good problem should have a key (i.e. the first move of the solution) that is NOT a check

[Paul Bonham]

Re: Re : Re: Spend your weekend on this chess problem

As far as problemists are concerned, two importants rules are:

1) a good problem should have one and only one solution (if there are two solutions or more, the problem is "cooked" and of little value)

2) a good problem should have a key (i.e. the first move of the solution) that is NOT a check

Thanks, Louis. The fact that this isn't a true problem is fine by me, because my reason for posting it wasn't to say, "Look, here's a problem I composed." There was no composing at all, it was randomly generated.

So while Mathieu seems to glorify that he exposed it as not a true problem, that's of no consequence to me. My reason for posting it was that it was generated randomly, and yet it has so many pieces / pawns in critical positions in such a way that each one seems to play a role in making the solution hard to find. That is to say, there's plenty of ways for White to go wrong, not just taking longer to mate, but even to get mated himself. In fact, Black has a mate-in-1 threat, and White embarking on a long series of checks could easily see the string of checks run dry and suddenly having to make a defensive move that turns the initiative over to Black.

And if White decides to instead remove the immediate Black mating threat with Rxb6, White gets mated nevertheless in 4 moves with best play.

So true problem or not, it seems to have both instructional and intrinsic value, and if even 1 person sees it that way as we did, great. Also, the odds of randomly generating such a position with so many pieces involved has to be.... well, maybe Mathieu, the probability expert, can tell us! :D

But Louis, I looked up the topic on Wikipedia and there's one other characteristic of a "good" problem you missed: there should be no promoted pawns. So that's another strike against this one, but again, I couldn't care less about that aspect of it.

Here's the full pgn of the solution we came up with, and again, I think White's 7th and 8th moves make a very remarkable sequence, even if there was another mate possible:

[Event "?"]
[Site "?"]
[Date "????.??.??"]
[Round "?"]
[White "?"]
[Black "?"]
[Result "*"]
[FEN "7Q/n1P5/1q6/q4k1N/1R2NB2/r2b4/2p1n3/7K w - - 0 1"]
{White must counter Black's threat of ...Qg1#.
But does White have his own mating net?}
1.Qh7+
( 1.Rxb6 Qe1+ 2.Kh2
( 2.Kg2 Bxe4+ 3.Kh2 Qh4# )
2...Qg1+ 3.Kh3 Bxe4+ 4.Kh4 Qg4# )
( 1.c8=Q+ Nxc8 2.Qxc8+ Kg6 3.Qg4+ Kf7 4.Qg7+ Ke6 5.Qg6+ Ke7 6.Bd6+ Kd7
7.Nhf6+ Kc6 8.Qe8+ Kb7 9.Qb8+ Ka6 10.Qa8+ Qa7 11.Qc8+ Qb7 12.Qxb7# )
1...Kg4
( 1...Qg6 2.Qd7+ Qe6 3.Nd6+ Kg4 4.Qxe6+ Bf5 5.Qxe2+ Rf3 6.Nf6+ Kh4 7.Bh6+
Qxb4 8.Nxf5+ Rxf5 9.Qh2# )
( 1...Ke6 2.Qg8+ Ke7 3.Bd6+ Qxd6 4.Qg7+ Ke6 5.Qf6+ Kd7 6.Qxd6+ Ke8 7.Rb8+
Nc8 8.Rxc8+ Kf7 9.Qf6# )
2.Nhf6+ Qxf6
( 2...Kxf4 3.Qh6+ Kf5 4.Qg5+ Ke6 5.Rxb6+ Qxb6 6.Qd5+ Ke7 7.Qd7+ Kf8 8.Qe8+
Kg7 9.Qg8+ Kh6 10.Qh7# )
3.Nxf6+ Kf3 4.Qh3+ Ng3+ 5.Qxg3+ Ke2 6.Qe3+
( {A less interesting line which also leads to mate in 11 is: } 6.Qg2+ Ke1
7.Bg3+ Kd1 8.Qg1+ Bf1 9.Qxf1+ Kd2 10.Ne4+ Ke3 11.Bf4# )
6...Kf1 7.Qc1+
( 7.Qg1+ Ke2 8.Kg2 c1=Q 9.Qxc1 Qd5+ 10.Nxd5 Be4+ 11.Rxe4+ Kd3 12.Qc4# )
( 7.Qf3+ Ke1 8.Bg3+ Kd2 9.Qf2+ Be2 10.Ne4+ Kc1 11.Qe1+ Bd1 12.Qd2# )
( 7.Bg3 Be4+ 8.Qxe4 Rxg3 9.Qf4+ Ke2 10.Re4+ Kd1 11.Qf1+ Kd2 12.Qe1+ Kd3
13.Qxg3+ Kd2 14.Qe1+ Kd3 15.Qe3# )
7...Ke2 8.Kg2!
{ Now the point of 7.Qc1+ becomes profoundly apparent. Black has no defense,
because there are twin threats of Qf1# and Qd2#. The key to the entire
position is White's Rook on b4, which has sat there unprotected for the
entire line. It now block's Black's Queen from defending d2, and it
defends against ...Be4+ after the Knight on f6 is used to defend against
...Qd5+. This same Rook could have, back at the beginning of this line,
taken the extra Black Queen on b6, but as the variations show, this would
result in WHITE getting mated! And so instead it sits en prise on b4,
waiting for the time when it becomes the key to the exquisite finish. }
8...Qd5+ 9.Nxd5 Be4+ 10.Rxe4+ Kd3 11.Qe3# *

[Wayne Komer]

Re: Re : Re: Spend your weekend on this chess problem

While I wouldn’t get involved in this discussion for worlds, I would point out that there is, in fact, a whole category of chess problems called Allumwandlung, where the pawn is promoted.

[Paul Bonham]

Re: Re : Re: Spend your weekend on this chess problem

While I wouldn’t get involved in this discussion for worlds, I would point out that there is, in fact, a whole category of chess problems called Allumwandlung, where the pawn is promoted.

OMG, yes... and according to Wikipedia, it refers especially to pawn UNDERpromotion.

But this kind of problem must be considered BAD by the "chess problem establishment". It's impure, it's unholy!

If only Monty Python had gotten hold of this...

NOBODY expects the German Allumwandlung!

Put Mr. Svidler in the COMFY chair!