share
ChessWhat's the maximum amount of material that a puzzle with unique solution can have?
[+4] [3] alices_and_bobs
[2024-06-08 18:39:45]
[ puzzles records ]
[ https://chess.stackexchange.com/questions/45632/whats-the-maximum-amount-of-material-that-a-puzzle-with-unique-solution-can-hav ]

It's OK if the position is impossible to reach as long as it's (forward-rules wise) legal, like a board with nothing but 2 kings and queens.

(1) Can someone who has a problem with this post explain what's wrong with it and why the downvote? I searched and Googled for puzzle with most queens and puzzle with most material, no definite answer came up. This is also not the same as the question about most queens in an actual game. I also tried with stockfish many positions with many queens, and it's not clear to me how to compose a puzzle with only solution when there are many queens. - alices_and_bobs
(2) @alice_and_bobs Yes drive-by downvoting is a hobby for some in stackexchange. It would spoil their fun if they had to explain themselves. So here are some suggestions from me how to improve your post. Uniqueness by itself is fairly trivial for a problem. The point of a chess composition is to show an artistic effect elegantly. So for example all the units have to be necessary for the construction. If one could be removed without affecting the solution in any way, then that would be a defect. Also, "impossible" and "illegal" are synonymous. But prior play was legal but not sensible - Laska
@Laska Sorry, I meant obviously illegal positions like both sides in check or having no kings. As for artistic, maybe usually having unnecessary material is inelegant, but I think if a problem manages to be interesting while having the maximum amount of material that would be artistic too. It would be nice if there exists a puzzle with 62 queens that's also artistic in some way/not a straightforward mate, say if both sides have equal material and whoever is to move can win, but if there isn't one, that's still good to know. - alices_and_bobs
pdb.dieschwalbe.de/search.jsp; enter APIECES=64 to get examples of chess problems with 64 pieces. (My own is arguably most close to "real chess".) - Hauke Reddmann
Ill-formed positions are those where the forward rules aren’t applicable e.g. side to move already delivering check, or pawns on an end rank, or no kings. Ill-formed positions must be illegal because if they were legal, players would have stumbled on them and fixed the rules - Laska
How do you define unique solution? For many positions it will be close to impossible to show that there are no other forced mates, - Helena
@alices_and_bobs usually some own research is expected in StackExchange beyond finding answers in other sources. Maybe include your own best proposal in the question to indicate what you are going for? - FerventHippo
@FerventHippo I tried including a possible answer in questions before but always got downvoted, OK maybe the question's too long or hard to read, but now that I make the question short and don't show my own attempts, people still aren't happy. This question is not as obvious (at least to many people) as some might think. I tried using 16 queens, then 30, but no unique solution, so no "best". I tried 62 but didn't see the crucial idea (pins and blockages) at first and my engine choked on that so I didn't make progress. David's answer made me realize full of queens might be the right way to go. - alices_and_bobs
[+21] [2024-06-09 02:22:24] Noam D. Elkies [ACCEPTED]

This must be the maximum for an illegal position (adapted from David's attempt):

[Title "_David_, v. NDE: White to move and win"]
[FEN "qqqqqqqq/qkqQqqqq/qqqQqqqq/qQQQqqqq/qqqqQqqq/qqqqqqqq/qqqqqqQQ/qqqqqqQK w - - 0 1"]

1. Qd5xc6#

After any other move White gets mated after running out of Queens.


1
[+8] [2024-06-08 21:56:32] D M

There's no reason you can't have all the material on the board with all pawns promoted to queens, for maximum material.

For example: White to move, mate in 1, everything but the mating move loses.

[fen "3BN2k/5B1b/4N3/qq1q4/qqq1qn2/QQR2qbq/QRQrr1n1/QQQQ1QK1 w - - 0 1"]

Or, for a little more complexity: Here every White piece except the wrong-colored bishop can give check, but there's only one solution:

[fen "1B1NbRqk/4n2q/Q7/1qqq3R/1qQ1q2N/QQ1qr2q/Q1QQr1Bb/1QQ2n1K w - - 0 1"]

2
[+1] [2024-06-08 21:23:20] David

White: queens on g1, g2 and h2. King on h1 Black: bishops on g3 and h4, king on g4 and a queen on all the remaining squares of the board but h3.

White to move and mate in 1 [1]

[1] https://lichess.org/analysis/qqqqqqqq/qqqqqqqq/qqqqqqqq/qqqqqqqq/qqqqqqkb/qqqqqqb1/qqqqqqQQ/qqqqqqQK_w_-_-_0_1?color=white

(2) Might as well put bQ's on g3 (pinned) and h3 too. - Noam D. Elkies
@NoamD.Elkies Thanks! It seems a board literally full of queens is actually easier for composing. Is it possible to have a queen on every square except for the two kings? nvm, I thought I found one but it's cooked - alices_and_bobs
@NoamD.Elkies OK how about this? I think it's legit, although I only tested it manually but my local stockfish froze on it: lichess.org/analysis/qqqqqqqq/qQQQQQqq/qQqqqQqq/qQqkqQqq/… - alices_and_bobs
Not quite (assuming the stipulation is just "White wins" rather than "Mate in 1"), because there White can play QxQ+ first (any of about 20 choices!) and only then Qf3xe4#; indeed even 1 Qe3(f4)xe4+ seems to work (Qxe4 2 Qf3xe4+ Qxe4 3 Qxe4#, again with some further Zwischenzügen possible on the other side of the bK). But I'm sure there's some arrangement of the two Kings and some 62 Queens that gives White a unique winning line. - Noam D. Elkies
3