Win, Draw, or Loss? You Decide :)

This will be much easier to solve than it was to devise. In fact, the position featured below was my own solution to a rather diabolical construction challenge, though I won't reveal precisely what that challenge was just yet... that would give too much away about the following puzzle:

 1.) Decide whether the above position is

     A. a win for White

     B. a draw

     C. a loss for White

     D. Unclear, without certain additional information

2.) Suppose White had a queen on h5 [instead of the bishop]. Would this change the answer to the 1st question?

1) White wins. This is clear.

2) Black wins. Depending on your stance on conventions this is less certain, but the standard answer is that Black wins.

It's definitely one of those three.

1) Black mates in 1 KxG4.. Nf6#

2) Black mates in 2 Q(B)xG5.. BxG5+   KxG4.. H5#

Win

I would advise some Stockfish analysis on #1. After 1. fxg6+ e.p. Be6! 2. exf4 Bxf4 things are not so clear to me. White is on the brink of being mated and there are also repetitions in view but I found nothing definite.

In #2 black mates in two moves. En passant is not permitted as it cannot be proved.

I didn't take En passant into account for this reason, The puzzles previous move needs clarifying.

Not necessary as you can prove in #1 that blacks last move was g7-g5. Not so in #2 though. For you to figure out why that is, or to wait for the OP to explain his puzzle.

d

2.  d..

The Stockfish on this site gave +18 when I checked the given position, and of course I didn't bother checking with wQh5 instead because white's plain lost there. With Bh5: 1. fxg6 e.p.! (only move, fortunately provably legal) Be6 2. exf4 Bxf4 3. Bg4! (only move) and black just doesn't have enough pieces to cover all escape squares. 3...hxg6 4. Bxe6 Bg3+ 5. Kg4 for instance, no mates to be had.

I did try a similar composition challenge a while back and got a problem without words, but also with the check in the position. Making the idea work without the check is twice as hard.

Thanks, Remellion! That's enough to put my mind to rest!

Yep, most e.p. problems have either a king in check or a king on the square where it would have been in check, had the single pawn step been played.

The phenomenon is related to one of my fairy ideas. Permit a pawn to be captured e.p. not only immediately after the pawns leap but also until some later event, for instance until some other pawn move on the board is played. That would multiply the number of ways to profit from the e.p. rule in compositions, both of the retro-type and the non-retro type.

Congratulations to Remellion for being the first to correctly solve it. I was hoping to see the relevant retro analysis somewhere (and it's worth mentioning that White is in fact provably lost with queen in place of bishop), but oh well. I'll post that in another day or two if no one else does.

Btw, the construction challenge went like this: devise a position in which White has 2 rooks and 2 bishops, and is [provably] winning... but if either rook or either bishop was instead a queen, then he'd be [provably] losing.

solved it,wait,i don't even know what I'm doing,I'm sad an lonely,make me happy by adding me as a friend

You did the right thing. There are no positions in standard chess where a "bishop" wins while a "queen" loses ("draw" is another thing). So you sought refuge in the retrograde type where "positions" are not fully determined by what is visible in the diagram.

Btw, another challenge would be to do it without a white queen in the diagram (which you probably already thought about). I am not sure it is possible though.

^ I assume you mean "with no White queen on the board"... in which case you've probably noticed the trick with the h-pawn (it can't promote on g8 if the last move was g7-g5+) that it seems you and Remellion overlooked earlier. If it's only required that e.p. not be provably possible after replacing rook/bishop with queen, then the challenge would be relatively easy to achieve even with no White queen originally on the board.

As is, however, it must surely be impossible to satisfy such stringent criteria with a queenless position. Remember, you'd need ALL 4 positions with a White queen in place of rook/bishop to be legal when e.p. is illegal but illegal when e.p. is legal, while the original position having both rooks and both bishops remains legal when e.p. is legal but illegal when e.p. is illegal. It's hard enough to make all of that work, while simultaneously ensuring that all positions will actually have the desired win/loss outcome, when you can trivially prove the occurrence of a queen promotion in the alternative positions. If there's no obvious queen promotion, the whole thing simply devolves into a topic exclusively for professionals.

EDIT: Oh, come on! Your edit of "with only one" to "without" was not exactly a matter of life or death, was it? 

We'll look forward to that. Mostly the legality verification of a posiiton requires only (a) explaining pawn structure (b) counting captures (c) noting necessary promotions. Amongst problemists this is considered "easy" and is usually left to the solvers. What they publish are the dynamic factors like (d) fixed game fragments especially the fixed moves preceding the diagram (e) the future options as a consequence of prior analysis. In the current problem (d) and (e) are concerned with the yes/no of the en passant move.

On chess.com it is a good idea to present the complete story as few attendants here are familiar with retrograde analysis. It is much better than producing a "proof game" (which many ask for) since proof games do not explain the logic surrounding the construction of the problem. I am now persona non grata with the math boys :-)

I edited for clarity as I often do. "Without" referred to the diagram, while "only one" referred to the position with the queen subsitution. I try to write in a way that others reading the thread may understand as well.

I'd wager a second strategy for this particular composition challenge would be in the vein of a "Mate in 1 (who?)" stipulation, where replacing wR/wB with wQ results in no black last move, or a check to which black responds with mate. That would still be firmly in retroland, but probably be even heavier material wise and very finicky.

Also, yep, overlooked that e.p. status is definite in either position. That makes this position even nicer.

In fact, I only commented on the assignment given to us (and so did Remellion, I guess), not to the original construction challenge with all rooks and bishops substitutable.

With the queen on h5, you need not prove that blacks last move was not g7-g5 since it is enough that g6-g5 was possible. I merely quoted the convention because Remellion referred to a conventional issue which I thought wasn't there.

Obviously, in this position (with queen), the move g7-g5 was out of the question here as I would haved noted had I gone on to solve the other substitutions. But, as said, I only solved the assignment.

It looks like this may need retrograde analysis. How did black's pawn get to g5 and what was white's move before that?

Either it was a push from g6 in which case white's previous move can only have been e2-e3, in which case the white bishop on h5 must have got there by promotion and white still has 7 pawns.

However it can only have been the h-pawn and if that queened on g8, although it's a white square that bishop would never have got out of there.

Thus it is impossible for white to have only just played e2-e3.

The other options are therefore a capture on g5 or g7-g5. In the latter case, white can capture en-passant and his last move must have been bishop from g6 to h5.

With regards to a capture, white is missing just 2 knights and a pawn but the black pawn on b4 has in theory made 2 captures to get there so the pawn on g5 cannot have also made 2 captures which would be necessary.

Therefore, whilst i haven't worked out if the position can be legally reached, if it can it must have been via g7-g5 so en-passant is possible.

Thus said I would now have to analyse the existing position to see what difference that makes as to who can win.