Puzzle with 2 solutions

This is the puzzle in question:
https://lichess.org/training/fDqkD

If you analyze it and get deep enough you will see that it clearly has 2 solutions for the 1st move. I didn't know which one was the correct one so I picked the 2nd option because it seemed stronger to have opponent's king on 3rd rank and not on the 2nd rank in the final position.

There was another forum post where IM @lovlas told to report puzzles with multiple solutions.
https://lichess.org/forum/general-chess-discussion/puzzles-with-more-than-one-good-solution

Stockfish can't detect that 2nd move with low depth.
+50 and +16 are both completely winning, they should both be accepted.

This puzzle ended my 66 deep puzzle streak. I feel very sad now. :(

This puzzle still exists, so I will keep this thread alive. Is this wrong place to complain about these?

I know that lichess will always be generating these kind of puzzles because stockfish can't analyze infinitely deep on every puzzle they generate.
It would be technically impossible to not have this kind of puzzles.

I think the rating of the puzzle has went down. Currently it has 956 plays.

So you mean that there are two equally good responses to c7? I'm not sure I agree. I did throw Stockfish at it with pretty deep analysis, so I know what you are saying, but...

Even after the line ...Rd4+, which loses a pawn for Black, both lines converge on ...Re8 to try to stop the c7 pawn.

So Black would end up in the same situation either way. Perhaps ...Rd4+ prolongs the agony by a move or two (while costing material), but it doesn't change the quality of the puzzle; we're not being asked to find the best moves for Black, but the best moves for White. White's best moves don't change.

Either way White ends up with a Queen and a Rook to a Queen.

I don't know why you feel like I talked about two equally good responses?

I am saying that white will have a clearly winning position if they play one of these moves:

• 60. Rb7+
• 60. c7

I saw that both moves were completely winning and my decision to play Rb7+ was based on the fact that lichess puzzles have only one correct solution (almost always). If you find multiple good candidates and you can't decide after a long think, then you have to guess. That is is how it works in real tournament chess and also in puzzles.

There are two exceptions to this rule in online chess puzzles (these apply to chess.com, lichess, chesstempo)

• You shouldn't repeat the position in a puzzle if you can avoid it.
• There are puzzles where you have multiple mating moves. Those are all accepted like they should be.

I've been doing puzzles on chess.com and lichess for many years now.
If these types of puzzles are intentional, then I am happy that I have learned something new here.

I believe that this puzzle is an accident and should be removed. Please check the forum link I posted, there a moderator tells us that these are broken puzzles.

Well, now that you've finally said what you mean....!

You posted four times in this thread without saying anything about the puzzle you're talking about, or what problem you had with it. I responded with what I had to offer. On reflection...

You're right that c7 and Rb7+ are both winning, but c7 appears to be more direct.

All lines seem to converge on ...Re8 trying to stop the c7 pawn from promoting. Playing c7 first gets to the point.

This may still be an inferior puzzle, or my thoughts on it may be inferior.