Well, that's what I thought. But IMHO if that's the answer this study is not as good as all that. I hate studies where the answer takes lengthy analysis. Especially since the position before e8=N is mutual zugzwang, and it's so irritating that I can't make it Black's move. But perhaps I can't.

Studies where Black can almost do something clever but can't quite are aesthetically a bit of a problem, as I see it.

Well, Gunter?!

I'm tempted to give my own offering (as in a study I like, not one I've composed). W Kh4, B e7, Ps g5, g6, c7. B Kc6, Rh8 Ph7, h4. W to play and win. Sorry, too stupid to do the diagram thing.

Many will have seen this before - it's in Nunn's Endgame Challenge - but I thought it was very impressive in an impossible-till-you-see-it way, which is what I like.

I think I've got it! In the spirit of this thread, I won't post the solution yet, but I can reveal Black's defensive idea:

1.e6 g3 2.Bf7 Ke3 3.e7 Kf2 4.e8=Q Bf3 5.Bd5 Bxg2+ 6.Bxg2 stalemate! So, somewhere along this line White has to improve.

Once again, a brilliant defensive idea! Took me ten minutes to see why pushing the e-pawn wasn't a simple win. In other words, I would have done just the same thing as your friend. Man, this is hard!

Gunter, 

That's unfair!

If I saw this in a game between two 1700s I would push the e-pawn and try to win by brute force.  Only afterwards would I realise there's a drawing mechanism available to Black.   

This is gonna be hard.

Wow, that's pretty cool .

Oh my. The reason I didn't think 4.a6 was working, was the line 5.a7 Kh5 6.a8=Q g6 7.Qa2 d1=Q 8.Qh2+ Nxh2 9.Bxd1+ Ng4 10.Bxg4+ Kxg4 11.Kc7 h5 12.e5 h4 13.gxh4 gxh4 14.e6 h3 15.e7 h2 16.e8=Q h1=Q 17.Qxg6+ and we have the position I mentioned with Q v Q. But.. is it as simple as 6.a8=R ? Black will no longer be zugzwanged, and while he has some dangerous pawns, the e pawn seems to assure White's victory.

Indeed, a brilliant defensive idea. However, unless I've missed something, that's the main brilliancy of the study. That said, I may indeed have missed something. In my head, the main line continues 2.Bd1 Ne3 3.Bb3+ Kg6 4.g4 (looks like 4.a6 almost works, but leads to Q v Q in the end!), when I can't really see a good continuation for Black. He can get a pawn to the 2nd rank in a couple of ways, but White stops it in time. Are there more brilliancies that I've overlooked?