Trying to compose a study...

[Arpad Rusz]

[D]8/p1p4p/p1B2Rp1/P7/P5b1/6P1/2K2P2/k2r4 w - - 0 1

The dream:
1. Bf3! Rd4 (1... Bxf3 2. Rxf3 +- Saavedra's double attack) 2. Kc3 Rxa4 3. Kb3 Bd7 (3... Rd4 4. Bxg4 Rxg4 5. Re6 +-) 4. Bc6! Rd4 (4... Bxc6 5. Rxc6 +- Saavedra's double attack) 5. Kc3 Rd1 6. Kc2 Bg4 (6... Rd4 7. Bxd7 Rxd7 8. Rf4 +-) Now we are back to the initial position but without the a4 pawn! 7. Rf4 +-

The nightmare:
2...c5! is a draw?
2.Bxg4!? is a dual?
3.Bxg4!? is a dual?

I have tried other settings too but I don't think I have found one with a clean solution. Here is one of them:

[D]7n/2p5/p1B2Rp1/P7/P5b1/6Pp/2K2P1P/k2r4 w - - 0 1

It would be very nice if somebody could find a position which works. Help me!

[peter]

[D]8/p1p4p/p1B2Rp1/P7/P5b1/6P1/2K2P2/k2r4 w - - 0 1

The dream:
1. Bf3! Rd4 (1... Bxf3 2. Rxf3 +- Saavedra's double attack) 2. Kc3 Rxa4 3. Kb3 Bd7 (3... Rd4 4. Bxg4 Rxg4 5. Re6 +-) 4. Bc6! Rd4 (4... Bxc6 5. Rxc6 +- Saavedra's double attack) 5. Kc3 Rd1 6. Kc2 Bg4 (6... Rd4 7. Bxd7 Rxd7 8. Rf4 +-) Now we are back to the initial position but without the a4 pawn! 7. Rf4 +-

The nightmare:
2...c5! is a draw?
2.Bxg4!? is a dual?
3.Bxg4!? is a dual?

Sorry, Arpad, I didn't see your posting till yesterday, looking for my own posting about Rubinstein-Tarrasch once again.
Probably it won't help, but as for your first nightmare, I think I can make you wake up, 2...c5 isn't a draw.

[Event "CCC"]
[Site "Dort"]
[Date "2012.03.28"]
[Round "?"]
[White "A, R."]
[Black "1"]
[Result "1-0"]
[Annotator "Martan,Peter"]
[SetUp "1"]
[FEN "8/p1p4p/p1B2Rp1/P7/P5b1/6P1/2K2P2/k2r4 w - - 0 1"]
[PlyCount "61"]
[EventDate "2012.??.??"]
[TimeControl "5400+30"]

1. Bf3 Rd4 2. Kc3 c5 3. Bxg4 Rxg4 4. Rxa6 Rxa4 5. Rxa7 Kb1 (5... g5 6. Kb3 Rb4+
7. Ka3 Rb5 8. a6 c4 9. Ka4 Rb6 10. Rxh7 Rxa6+ 11. Kb5 Rg6 12. Kxc4 g4 13. Kb3
Rb6+ 14. Kc3 Rg6 15. Rh4 Ka2 16. Kd4 Ka3 17. Ke3 Rg8 18. Kf4 Rf8+ 19. Kxg4 {#32
}) (5... h5 6. a6 Ra3+ 7. Kc4 Ra2 8. Ra8 Ra5 9. a7 Ka2 10. f4 Ka3 11. Rg8 Rxa7
12. Rxg6 Kb2 13. Rg5 Rh7 14. Kxc5 Kc1 15. f5 Rh6 16. Kd5 Kd2 17. Ke5 Rb6 18.
Rxh5 {#23}) 6. Rb7+ Kc1 7. Rb5 Kd1 8. Kb3 Ra1 9. Rxc5 Ke1 10. f4 Kf2 11. g4 Ke3
12. f5 gxf5 13. Rxf5 Kd4 14. Kb2 Re1 15. Rf7 Re4 16. Rg7 h6 17. a6 Kc5 18. Kc3
(18. Rg6 Rf4 19. Kc3 Ra4 20. Kd2 Kd5 21. Kd3 Rf4 22. Ke3 Rc4 23. Kf2 Rf4+ 24.
Kg3 Rf8 25. Rxh6 Rf7 26. g5 Rf1 27. Rb6 Ra1 28. g6 Kc5 29. Rf6 Kd5 30. Kf4 Rf1+
31. Kg5 Rg1+ 32. Kh6 Rh1+ 33. Kg7 Rc1 34. a7 Rc7+ 35. Rf7 Rc8 36. Kh7 Ke6 37.
Rb7 Ke5 38. g7 Rc3 39. a8=Q Rh3+ 40. Kg8 Re3 41. Kf8 Rf3+ 42. Rf7 Rf6 43. Qa1+
Ke4 44. Rxf6 Ke3 45. g8=Q Kd3 46. Qb3+ Kd2 47. Rf2#) 18... Kc6 19. a7 Ra4 20.
Kb3 Ra1 21. Kb4 Ra2 22. Rh7 Kd5 23. Kb5 Rb2+ 24. Ka6 Ra2+ 25. Kb7 Rb2+ 26. Kc7
Rc2+ 27. Kb8 Rb2+ 28. Rb7 Rf2 29. a8=Q Rf8+ 30. Ka7 Rxa8+ 31. Kxa8 {#16}

This 6MOB .pgn made with Rybka and Houdini won't help because of the Bxg4 duals, as for which you might be right anyhow, have to look at those even more deeply to be quite sure, but both seem to be white wins too to me.

If I have any constructive idea, let you know, pity I'm much better in proofing and cooking studies but composing them.

Just for my personal point of view: isn't the position way too complicated for human taste?

[Arpad Rusz]

Thanks for your analyses! Probably 2...c5 is not that dangerous like I have thought. That's good news.

Adding some pawns on the h-file changes things a little but probably won't eliminate all cooks:

[D]8/p1p4p/p1B2Rp1/P7/P5b1/6Pp/2K2P1P/k2r4 w - - 0 1

[D]8/p1p4p/p1B2Rp1/P7/P5bp/6P1/2K2P1P/k2r4 w - - 0 1

[peter]

Adding some pawns on the h-file changes things a little but probably won't eliminate all cooks:

At the first and at the second one we still have the 2.Kc3-2.Bxg4 dual yet, I guess, but I have to look deeper, maybe my first impression is wrong, tell you so, if this is the case unexpectedly

[peter]

Adding some pawns on the h-file changes things a little but probably won't eliminate all cooks:

[D]8/p1p4p/p1B2Rp1/P7/P5b1/6Pp/2K2P1P/k2r4 w - - 0 1

What a pity, Arpad!

2.Bxg4 really is a hard one now, Black resists over 50 moves and White has a real small edge, at one 6MOB- position White has to underpromote to a Knight instead of a Queen, just wanted to show it because it's really funny, just have a look at the longest variant in braces.


But however, it seems to be the same major dual:

[Event "Stellung ausspielen (DeepRybka41SSE42x6"]
[Site "?"]
[Date "2012.03.29"]
[Round "?"]
[White "AR"]
[Black "3"]
[Result "1-0"]
[Annotator "Martan,Peter"]
[SetUp "1"]
[FEN "8/p1p4p/p1B2Rp1/P7/P5b1/6Pp/2K2P1P/k2r4 w - - 0 1"]
[PlyCount "91"]

1. Bf3 Rd4 2. Bxg4 (2. Kc3 c5 3. Bxg4 Rxg4 4. Rxa6 Rxa4 5. Rxa7 h5 6. a6 g5 7.
Rc7 Ra3+ 8. Kd2 Ra2+ 9. Kc1 Rxa6 10. Rxc5 Ka2 11. Rxg5 Kb3 12. Kd2 Rh6 13. Ke3
Kc2 14. f4 Kd1 15. f5 Kc2 16. Rg6 Rh8 17. f6 Re8+ 18. Kf3 Kd2 19. f7 Rf8 20.
Rf6 Kd3 21. Kf4 Kc2 22. Kg5 Kd3 23. Kg6 Kd2 24. Kg7 Rc8 25. f8=Q Rxf8 26. Rxf8
Ke3 27. Kf6 Kd4 28. Rh8 Ke3 29. Kf5 Kf3 30. Re8 Kf2 31. Kf4 Kg2 32. Ke3 Kg1 33.
Kf3 Kh1 34. Ra8 Kxh2 35. Kf2 Kh1 36. Ra4 h4 37. gxh4 Kh2 38. Ra3 Kh1 39. Rxh3#)
2... Rxg4 3. Rxa6 Rxa4 4. Rxa7 Ra2+ 5. Kd3 h5 (5... Rxf2 6. Rxc7 Kb1 7. a6 Rxh2
8. Rxh7 Ra2 9. Rxh3 Ra3+ 10. Kd2 Rxa6 {#50} 11. Rh4) 6. a6 Ra4 7. Ra8 Ka2 8. a7
c5 (8... Ka3 9. Kc3 Ka2 10. Rg8 Ra3+ 11. Kc2 Rxa7 12. Rxg6 Ka1 13. Rf6 Ra5 14.
Rf4 c6 15. Rf3 Ka2 16. Rc3 c5 17. Rc4 Ka1 18. Rf4 Ra2+ 19. Kc1 Ra5 20. Rf5 Ka2
21. Kc2 Ka3 22. Kc3 Rb5 23. Rxh5 Rb1 24. Rxc5 Rc1+ 25. Kd4 Rh1 26. Rh5 Kb4 27.
g4 Rxh2 28. g5 Rh1 29. Ke4 Kc5 30. g6+ Kd6 31. Rxh3 {#23}) 9. f3 Ka1 10. g4
hxg4 11. fxg4 Ka2 12. g5 Ka1 13. Kc3 Ka2 14. Kc2 Ra3 15. Rh8 Rxa7 16. Rxh3 Re7
17. Rh6 Re2+ 18. Kc3 Re3+ 19. Kc4 Re4+ 20. Kd5 Rg4 21. Rxg6 c4 22. h3 Rh4 23.
Ra6+ Kb3 24. Rb6+ Kc3 25. Rh6 Rd4+ 26. Ke5 Rd1 27. h4 Kb4 28. Rh7 Rg1 29. Kf6
Rf1+ 30. Kg6 Ra1 31. h5 Ra6+ 32. Kf5 Ra5+ 33. Kf4 Rc5 34. Rb7+ Ka3 35. h6 c3
36. h7 c2 37. Ra7+ Kb3 38. Ra1 Rc4+ 39. Kf5 Rc5+ 40. Ke6 Rc8 41. g6 Re8+ 42.
Kf5 Kb2 43. Re1 Ka3 44. Rc1 (44. Rxe8 {#52} c1=Q 45. h8=N {#51 } (45. h8=Q $2
Qf1+ {1/2-1/2})) 44... Rf8+ 45. Kg5 Ra8 46. Rxc2 {#10} 1-0

[Arpad Rusz]

Thanks, Peter!