sssthepro wrote on 09/24/09 at 09:47:28:
Did you really find that variation by yourself Girkassa? If you did, you must be really strong. How did you see these kind of stuff?
Found it by myself, yes. I'm 2400-ish, one norm short of the IM title. How I see this kind of stuff? Well, I could try to reconstruct my thoughts when I solved the study (Günther hasn't confirmed that it's the solution, but since he hasn't commented on it, and since it looks winning, I believe it is):
1.e6 - This has to be the first move, as it's a simple draw if Black captures the e pawn.
1...g3 - Not as obvious, but White threatens 2.e7 Ba4 3.Bf7, so Black has to do something about it. The first line that came to mind was 1...Ba4 2.Bf7 Bb3 3.Bg6+ winning. Then I saw Black could get the same position with the pawn on g3, which could hardly be a disadvantage - Black gets closer to counterplay based on the weak White king position.
2.Bf7 - In case of 2.e7, Black plays 2...Bh5 followed by 3...Be8, and then Black can simly pick up the pawn with his king. Therefore, this move seemed to be the only way to achieve something
2...Ke3! - This is where it got hard. In a game, I might very well have calculated only 2...Bb3 3.Bg6+, concluding that White wins by "normal" means. However, since this is a study, you know there has to be something more, so I kept searching. And since White's e pawn can't be stopped by normal means, the reduced material only gives one more possibility: Counterplay against White's king. I saw the moves Ke3, Kf2, Bf3, but at first, I thought White could simply play Kg1 when needed. Then I saw 2...Ke3 also threatened 3...Bb3, when 4.Bg6 is no longer with check. Thus, the move 2...Ke3, with an amusing double threat, suggested itself even before I had established that there is no win for White after 3.e7 Kf2 4.e8=Q Bf3.
3.Be8!! - This is where it got really hard. First, I established that, after 3.e7 Kf2, promoting to a rook wouldn't prevent the stalemate, that a bishop would be useless, and that a knight would even lose. Then I quickly looked at 4.Bd5 Ba4, but since White has to prevent ...Bc6, he can't make any progress there. Therefore, I searched for other moves than 3.e7, trying to find something after 3.Kg1 Bb3 4.Bg8 Ba4, but there just wasn't anything. Having looked at 3.e7 and 3.Kg1, there weren't really a lot of other useful moves left. Then it struck me: If White could get to the a8-g2 diagonal before Black, there wouldn't be time for stalemate. The ideal square for the bishop must be c6, where it not only protects the diagonal, but also covers e8 - and that square is available in two moves! Again, it helped a lot to know this is a study: You know there is a win, so you don't stop searching, and also, 3.Be8 is the kind of study-like move that could very well be the solution. If this had been a real game, I would probably have played 3.e7 Kf2 4.e8=Q Bf3 5.Bd5, hoping Black would resign, and if he played 5...Bxg2+ 6.Bxg2, it would at least be a very amusing end to the game.
3...Kf2 - Quickly calculating 3...Bf3 4.Kg1 and 3...Bb3 4.e7 Kf2 5.Bc6, this move appeared to be the only sensible move left.
4.Bc6 - White has to prevent 4...Bf3, and this is the only way.
4...Be2 - Since White's e pawn is still unstoppable, Black has to continue playing against g2. Having already seen the stalemate theme, I knew White couldn't rely on g2 being defended by the bishop, and the manoeuvre ...Be2-f1 suggested itself as the quickest route to f1.
5.e7 Bf1 - Like Cox, I tried to find a way to get this position with Black to move, but I couldn't find any, and anyway, I just couldn't believe then position to be anything but won after 6.e8=N. Black can never exchange the bishop for either of White's pieces, meaning that White can always drive away Black's king with a knight check, covered by the bishop. If White threatens Ne4+, Black has to play ...Ke3 in order to save his pawn, and then White's king can escape from the corner. Once I saw that, I thought it just had to be winning somehow.
As you see, my thinking in this, and most other, studies, is pretty much a process of elimination. I often don't find the right continuation until I've seen that all other sensible moves are useless.
Pages: 
Endgamestudies (Read 99998 times)
