You are possibly right, nevertheless Black managed both times to achieve a nice blockade during the transition phase from middlegame to late middlegame/emdagme. I can't remember the exact games, but I do remember moving rather hastily for the final attack (a bad habit of mine)..

Yes, you both seem right. The defence may not be comfortable at all times, but it may as well be an easy one for a Berlin specialist. I play the Spanish exchange and this is why I was mainly interested in this endgame, since I have reached it a couple of times with Rooks and Knights instead of the Bishops. The result was ..ahem.. draws, but I thought I could have been missing sth clever.. Anyway, cheers for the analysis and the insights.

This endgame has been played more than once, and generally it is a draw. Black should not trade rooks (for example he mistakenly did in Howell-Parker, 4NCL), and correctly didn't here and in Someone-Hracek, from the 2006 Olympiad I think). The pure bishop ending is very close, certainly closer than this.

First of all: I want to look at the postion of your improvement line starting with 34.Be3
The fact that white has played Be3 gives black the posibility of an intermediate check:
after 38.gxf4,
http://www.france-echecs.com/diagramme/imgboard.phpfen=4k3/2p1b3/1p2p3/p1p1PP2/P...
4k3/2p1b3/1p2p3/p1p1PP2/P1P1K3/1P2BR2/8/7r b - -

Now black has the intermediate check [b]38.....Rh4+ which is of course impossible with the bishop on f2.[/b]
All the same, even after your sample line I still dont see a clear win.

Whites best seems
I.39.Bf4, Bg5 when I see no clear winn for white after either
A) 40.fxe6, Ke7 41.Kf5,Bxf4 42.Rxf4,Rh5+= (not 42...Rxf4?? 43.Kxf4,Kxe6 and whites wins)
B) 40.f6,Kf7! 41.Rf1,Bxf 42.Rxf,Rxf 43.Kxf, Kg6 =
While
II.39.Rf4,exf 40.Kxf5,Rh5+ 41.Ke6,Bg5! provideds another usefull simplicfication

In the game the position after blacks 40th move is:

http://www.france-echecs.com/diagramme/imgboard.phpfen=2k5/2p1b3/1p2P3/p1pK1p1r/...
2k5/2p1b3/1p2P3/p1pK1p1r/P1P5/1P2BR2/8/8 w - -

Here your suggestion 41.Ke5 is positional sound, but unfortunately has a tactical drawback:
41.....f4+! 42.Kxf4 (forced), Bd6+ 43.Kg4 (43.Ke4,Re5+ and 44...Rxe6), Re5 and white's e-pawn drops.

All I can say at the moment is that white certainly has a small advantage, but I've not found
a clear winning line yet. But I have to admit, that I probably should spend more time on it,
as I have the feeling I've not yet grasped all the important characteristics yet.
One of the problems is that rook endings, after an exchange of bishops seem to have a
wide drawing margin.


[img]http://www.france-echecs.com/diagramme/imgboard.php?fen=[/img

Yesterday I was following online the Russia-China match. One of the games that cought my attention was that of Timofeev vs. Ni (a Queenless Spanish Berlin), where White reached a rather favourable endgame (in the spirit of thematic Spanish Exchange endgames). I must admit I was expecting a white win and was a bit dissapointed by the resulting draw. Here's the game score:

[Event "IV Match of Russia – China"]
[Site "Nizhniy Novgorod"]
[Date "2007/8/29"]
[Round "09"]
[White "Timofeev, Artyom"]
[Black "Ni, Hua"]
[Result "1/2-1/2"]
[Board "03"]
[Input "DGT1028"]
[Owner "Association of chess federations, Moscow"]

1. e4 e5 2. Nf3 Nc6 3. Bb5 Nf6 4. O-O Nxe4 5. d4 Nd6 6. Bxc6 dxc6 7. dxe5 Nf5
8. Qxd8+ Kxd8 9. Nc3 Ne7 10. h3 h6 11. Be3 Ng6 12. Rad1+ Ke8 13. a3 Be7 14.
Rfe1 Bd7 15. Ne4 Nh4 16. Nxh4 Bxh4 17. Bd4 Be6 18. Nc5 b6 19. Nxe6 fxe6 20. g3
Be7 21. Kg2 Rd8 22. f4 g6 23. Kf3 a5 24. a4 h5 25. Ke4 Rh7 26. Be3 Rd5 27. c4
Rxd1 28. Rxd1 c5 29. b3 Rh8 30. Rd3 Rh7 31. Bf2 Rh8 32. Re3 Rh7 33. Rf3 Rh8 34.
g4 hxg4 35. hxg4 Rh1 36. f5 gxf5+ 37. gxf5 exf5+ 38. Kd5 Rh5 39. Be3 Kd7 40.
e6+ Kc8 41. Kc6 Bd6 42. Bf4 Bxf4 43. Rxf4 Kd8 44. e7+ Kxe7 45. Kxc7 Rh3 46.
Kxb6 Rxb3+ 47. Kxc5 Ke6 48. Rf1 Rb4 49. Re1+ Kd7 50. Kd5 Rxa4 51. Rf1 Ra2 52.
Rxf5 Rh2 53. Rf7+ Kc8 54. c5 Rh6 55. c6 1/2-1/2

I think that after 29. b3, White has many positional trumps (K-side majority and available pawn breaks there / Bishops of same colour / more space; especially regarding the Kings).

a) If White had arranged to reach the position after 36. f5 but with his Bishop already at e3 to stop an Rh6+ (e.g. 34. Be3 Rh7 35. g4 hxg4 36. hxg4 Rh1 37. f5 gxf5+ 38. gxf5 exf5+ 39. Kxf5), it looks like a clear win.

b) Even 41. Ke5 Bd6+ 42. Kf6 looks critical for Black.

What do you think?