Another retrograde problem

There is one promoted piece for each side in this position. Which pieces are the promoted ones?

Originally posted by fetofs
There is one promoted piece for each side in this position. Which pieces are the promoted ones?

[fen]rn3bN1/1ppp2pB/4k1Np/8/5B2/Kp1RP1P1/3PP2P/8[/fen]

The last moves were:

0...a4xb3ep -1.b2-b4 Kd6xPe6 -2.d5xe6ep e7-e5 -3.Bg5-f4.



wBg5 is a promoted piece, because the original died on c1.

However, there is no need for a second promoted piece to be present in the diagram. Black's a-pawn could capture wQ and wR, letting wPa through to a7. Pa7xQb8 produces the Bf4. f2xNe3 opens the f-file and Pf7 promotes and goes to d5 for c4xd5. Thus, it can't be proven that there is a second promoted piece on the board and the problem is cooked.

Originally posted by BigDoggProblem

However, there is no need for a second promoted piece to be present in the diagram. Thus, it can't be proven that there is a second promoted piece on the board and the problem is cooked.

But does it need to be proven or needed?

Originally posted by fetofs
But does it need to be proven or needed?

Absolutely.

There should be no solution possible other than the composer's intended solution.

In this case, why even bother talking about which pieces are promoted? Only Bf4 was for sure. A better stipulation would be, "Is the position legal?". This would force the solver to find the two e.p. captures, which seems to be the main point of the problem.