easy really.Just do en ....
REALLY difficult problem
This is a study by the great Nikolai Grigoriev, a composer especially noted for his pawn endings, some extremely subtle, some, like this one, absolutely spectacular. It was published in the Soviet magazine Shakhmaty v SSSR in 1945.
Good luck to anybody who tries their hand at this--you may need it!

@vsarun
the solution is not to take the g5 pawn en passant. how do you know that black's last move was g5? This problem has nothing to do with this type of solutions.
@camembert
Thanks for you information.

I must be missing something, because there seems to be an easy solution..? So send me the answer. Also, I'm assuming you can't do en passant on the black pawn because it didn't just move?

If it was not en passant I see an easy solution in the g and h files .As far as I can see the solution is quite easily visible in the those files .post if Iam wrong

vsarun wrote:
If it was not en passant I see an easy solution in the g and h files .As far as I can see the solution is quite easily visible in the those files .post if Iam wrong
I've posted earlier that there is no way to know that black's last move was g5.

no en passant is possible if that fen is complete (complete fens show en passant squares: google to find pgn and fen standards)

lotus_elise wrote:
easy white win 1.kc1...ka4 2.kb2...___ 3.g3...____
No, it's not that simple.
I'll give everyone of you a hint.
1.Kc1 Ka4 2.Kb2 d5! 3.g3 b5 4.h4 gxh4 5. gxh4 a5 6.Kb1 (6.h5? stalemate) 6...Ka3 7.h5 b2 8.h6 b3 9.h7 b4 10.h8=Q a4 and no matter how white replies it is a stalemate. It's a big hint but the whole answer is really more complicated.
To make you undertsand how difficult it is I put Fritz11 to solve it and after 2 hours of analyzing it hadn't found the solution. I believe that if I had let it some more hours it would have solve it, but it still is really difficult.
So... Why not 2.kb1 instead? Because white doesn't quite make it with the h-pawn. Hmm. The first move might be d5 but I can't prove it.
1.Kc1 Ka4 2.Kb2 d5! 3.g3 b5 4.h4 gxh4 5. gxh4 a5 6.Kb1 (6.h5? stalemate) 6...Ka3 7.h5 b2 8.h6 b3 9.h7 b4 10.h8=Q a4 and no matter how white replies it is a stalemate. It's a big hint but the whole answer is really more complicated.

@Newagex
Your answer is right but if you haven't noticed I've asked that the answer is not posted here, so that it is not spoiled. As the answer is not complete even now I'm going to ask what is the solution after 1...Kb5. please do not post the answer here, just send me a pm.
@madcastle82
After 2.Kb1 the line goes. 2...Ka3 3.g3 b2 4.h4 gxh4 5.gxh4 Kb3 6.h5 a5 7.h6 a4 8.h7 a3 and 9...a2#

engines are bad at solving such studies. [they are not so good at endgames in general..].
I've noted it down for solving in future (after my game against Kosmeg is over - it is very drawish but I am neither offering nor accepting a draw at present :) )
Specially, it is a Grigoriev study, so it is expected to be difficult...
[iirc Grigoriev was related to Leon Trotsky and for this reason he was unfortunately killed in one of Stalin's purges...]

kosmeg wrote:
vsarun wrote:
If it was not en passant I see an easy solution in the g and h files .As far as I can see the solution is quite easily visible in the those files .post if Iam wrong I've posted earlier that there is no way to know that black's last move was g5.
In my second post ,I was not meaning en passant. I was just saying that We could promote a pawn in the g and h files just like newagex did.But I was not smart enough to see all that stalemate business.Good puzzle
P.S. If anyone knows the composer of this puzzle, please let me know.