Due to an error with my ChessBase I can't attach a PGN file, so I'll attach my analyses in this post.
Quote:6...a6 immediately comes to mind given 7. c5 Nc6 8. e3 Ne4! (Dautov's exclam.) is a line Dautov marks as unclear. 7. e3 would tranpose into 6. e3 a6, but without white playing the best line against it (namely 7. Qc2, instead of a3), in which case black might be okay. I'll have to check...
[Event "Urumia op 2nd"]
[Date "2008.08.30"]
[Round "8"]
[White "Ibrahimov, Rasul"]
[Black "Ghane Gardeh, Shojaat"]
[Result "1-0"]
[WhiteElo "2537"]
[BlackElo "2415"]
1. Nf3 e6 2. c4 Nf6 3. Nc3 d5 4. d4 Be7 5. Bf4 O-O 6. e3 (6. a3 a6 7. e3 dxc4 {
This move has some merit as Black can follow up with a later ...b5, but White
should keep a comfortable edge due to his superior central control.} (7... Nbd7
8. c5 {- game}) (7... b6 $6 8. cxd5 exd5 9. Bd3 {is a much improved Carlsbad
position for White.}) (7... c5 {is thematic, but insufficient after} 8. dxc5 $1
Bxc5 (8... dxc4 9. Bxc4 Bxc5 10. b4 Be7 11. Qc2 $1 {gives White a powerful
initiative as he is miles ahead in development.})) 8. Bxc4 b5 (8... Nh5 9. Be5
Nc6 10. O-O f6 11. Bg3 Nxg3 12. hxg3 {leaves Black in a very passive position.}
) 9. Bd3 Bb7 10. Qc2 $1 {favours White, who will follow up with 11.b4 and
achieve a stable positional edge. If Black tries to stop this with} c5 $5 {,
then} 11. dxc5 Bxc5 12. Rd1 Qb6 13. O-O {gives White the initiative due to his
lead in development.}) 6... Nbd7 {6.a3 prevents this move due to 7.Nb5! which
gives White a comfortable edge.} 7. a3 (7. c5 {is more common.}) 7... a6 8. c5
{This sort of position is very comfortable for White - he has a space
advantage on the queenside, more active pieces and Black has no active
counterplay because White has too many pieces preventing the ...e5 break.} Nh5
(8... Ne4 9. Bd3 f5 10. Ne5 {is a great Stonewall for White.}) (8... c6 {has
been played, but} 9. Bd3 b6 10. b4 a5 11. O-O (11. h3 $5 {makes sense but in
my opinion White should not feat ...Nh5}) 11... Nh5 12. Qc2 Nxf4 13. exf4 g6
14. g3 {and Black's position leaves a lot to be desired - he risks getting
squashed.}) 9. Bd3 Nxf4 10. exf4 Re8 11. Qc2 Nf8 12. O-O Bd7 13. b4 {White is
already clearly better - Black has no active plan and White can build up on
the queenside to his leisure.} Qb8 14. Ne5 Rd8 ({In hindsight, Black should
have stopped White from opening the centre with} 14... g6 {although his
position remains suspect.}) 15. f5 $1 $16 c6 16. Rae1 Bf6 17. fxe6 Bxe6 18. f4
Qc7 19. Nb1 Re8 20. Nd2 Rad8 21. Ndf3 Qc8 22. h3 Be7 23. Ng5 Bxg5 24. fxg5 Ng6
25. Nxg6 hxg6 26. Bxg6 fxg6 27. Qxg6 Qd7 28. Qh5 Bf7 29. g6 Bxg6 30. Qxg6 Re4
31. Rxe4 dxe4 32. Qxe4 Qxd4+ 33. Qxd4 Rxd4 34. Rf3 a5 35. Rb3 axb4 36. Rxb4 Rd3
37. Rxb7 Rc3 38. a4 Rxc5 39. Rb1 Kf7 40. Ra1 Ra5 41. Kf2 Ke6 42. Ke3 Kd5 43.
Kd3 Ra7 44. h4 Ra5 45. g3 Ra6 46. a5 Kc5 47. Kc3 g6 48. g4 Kb5 49. Kd4 c5+ 50.
Kd5 Kb4 51. Rb1+ Ka4 52. Kxc5 1-0
Conclusion: 6.a3 a6 is better for White, and the positions with the a3-b4-c5-d4-e3-f2 pawn chain offer few prospects for Black.
Quote:Also, in your 6...b6 variation I'm not sure black's best is 8...dxc4, since 8...c5 would be more thematic, and how black is usually supposed to handle the position. Even in the line you gave, however, I'm not sure white is better after 9...Nbd7 with either Nh5 or c5 to follow depending on white's response...
#1) 5. Bf4 0-0 6. a3 b6 7. e3 Bb7 8. Bd3 c5 and I can't find anything for white that doesn't fizzle out quickly after either 9. cxd5 or 9. dxc5.
1. d4 d5 2. c4 e6 3. Nc3 Nf6 4. Nf3 Be7 5. Bf4 O-O 6. a3 b6 {If Black doesn't
want to transpose to the main lines with 6...c5 then this is the best
alternative, which was recommended by Kaufman. However, White should still
keep a small edge.} 7. e3 Bb7 (7... c5 {is interesting when} 8. dxc5 bxc5 9.
Bd3 Bb7 (9... Bd6 10. Bg5 $1 {is good for White, since} Nbd7 11. cxd5 exd5 12.
Nxd5 h6 13. Nxf6+ Nxf6 14. Bh4 {doesn't give Black enough for the pawn.}) 10.
O-O Nbd7 (10... Nc6 11. cxd5 Nxd5 12. Nxd5 exd5 13. Qc2 g6 14. Be2 $14 {and
Rybka doesn't think White has more than a symbolic edge but I disagree - White
will play Rfd1 and Rac1 and can exploit the weakened dark squares with Bh6 and
Qc3. White can also prepare a timely e4 to break up the hanging pawns, and b2
can be easily defended.}) 11. cxd5 exd5 12. Qc2 c4 13. Be2 {and with both a
d4-outpost for the knight and the d5-pawn as a potential target, White is
better. Rybka likes Black's bishop pair after} Nh5 14. Bg3 Nxg3 15. hxg3 Qa5 {
, but the weakness of d5 and White's superior minor pieces give him the edge
after} 16. Rfd1 Nf6 17. Nd4 {, with Bf3/Nf5 and b3 being two worthwhile ideas.}
) 8. Bd3 (8. cxd5 $5 Nxd5 9. Nxd5 Bxd5 {is suggested by Rybka but is innocuous
due to} 10. Qc2 c5 11. dxc5 bxc5 $1 12. e4 Bb7 {and with ...Nc6-d4 coming,
White has no edge. If White avoids e4, then ...f5 gives Black sufficient
counterplay.}) 8... dxc4 (8... c5 {is better here than after 7...c5, since
after} 9. dxc5 {Black does not have to transpose to 7...c5 with 9...bc5 10.0-0
but can instead play 9...Nbd7 or 9...dc4.} Nbd7 $5 {I rather like this move,
sacrificing a pawn for a lead in development and some initiative, although
White still keeps a plus with best play.} (9... dxc4 10. Bxc4 Qc8 $1 11. O-O
Qxc5 12. Qe2 Nbd7 13. Rac1 $14 {Kaufman. Rybka claims it's equal after} Bxf3
14. gxf3 Qh5 {, but it is obvious to the human eye that White's bishops are a
strong asset and Black also has some weak squares on the queenside.}) 10. cxd5
(10. c6 Bxc6 11. O-O dxc4 (11... Nc5 12. Ne5 Bb7 13. Be2 Nfe4 $11) 12. Bxc4 Qc8
13. Qe2 {looks nice for White, but he has no edge after} Nh5 $1 14. Rfd1 Nxf4
15. exf4 Bxf3 16. Qxf3 Qxc4 17. Rxd7 Bf6 $11) 10... Nxd5 11. Nxd5 Bxd5 12. c6
$1 Bxc6 (12... Nc5 13. c7 $1 Qd7 14. Bc2 $16) 13. Bxh7+ Kxh7 14. Qc2+ Kg8 15.
Qxc6 Nc5 (15... Rc8 16. Qb5 Rc5 {(this is necessary as Rd1 was threatened)} 17.
Qe2 Qc8 (17... e5 $5 18. b4 exf4 $1 {(the best practical chance)} (18... Rc8
19. Bg3 e4 20. Nd4 {and I agree with Rybka that Black has no compensation for
the pawn.}) 19. bxc5 fxe3 20. Rd1 Qc7 21. fxe3 Bxc5 22. O-O {and White has
excellent winning chances with Nd4 coming next.}) 18. O-O Rc2 19. Qb5 {and
Black has compensation owing to his more active pieces but I would much rather
be in White's shoes. Once he gets his rooks to the c- and d-files, he will
have good winning chances.}) 16. O-O $1 (16. Rd1 Nd3+ 17. Kf1 Rc8 18. Qb5 (18.
Qe4 $6 Rc1 19. Rxc1 Nxc1 20. Nd4 Nb3 $1 $36 {is fine for Black - White's king
is very vulnerable and already he has to think about how to secure the draw.})
18... Rc1 19. Rxc1 Nxc1 20. Nd4 Bf6 21. Be5 Bxe5 22. Qxe5 Qa8 $11) 16... Rc8
17. Qb5 a6 18. Qb4 (18. Qc4 b5 19. Qe2 Qd3 {gives Black ample compensation as
his chances of eventually regaining the pawn are high.}) 18... Nd3 19. Qb3 Nxf4
20. exf4 Qc7 (20... Bf6 $5) 21. Qe3 $14 {and I slightly prefer White.}) 9. Bxc4
Nbd7 {This logical move was suggested by Bryan Paulsen, which is equivalent to
Kaufman's given 9...Bd6 and 9...c5: Black is solid but White keeps a small and
stable edge.} 10. O-O c5 (10... Nh5 11. Be5 $1 (11. Bg3 $6 {justifies Black's
play and indeed he should equalise with} a6 12. Rc1 b5 13. Bd3 Nxg3 14. hxg3 c5
$132 {with a full share of the play and perhaps more.}) 11... c5 (11... Nxe5 {
may well be best:} 12. dxe5 $1 (12. Nxe5 Nf6 13. f4 {with the intention of f5
also looks promising for White but I prefer 12.de5 since it takes the sting
out of a ...c5 advance.}) 12... Qxd1 13. Rfxd1 Rfd8 14. Nb5 $14) 12. d5 $1 {(a
thematic answer to ...c5)} exd5 13. Bxd5 Bxd5 14. Qxd5 Nhf6 15. Qb3 $1 (15. Qb7
Qc8 16. Qxc8 Raxc8 17. Rfd1 Nxe5 18. Nxe5 {also gives White an edge, albeit
not enough to really trouble Black.}) 15... Nxe5 16. Nxe5 Bd6 17. Nc4 {and
White's kingside majority is more mobile than Black's queenside majority,
which gives him the advantage. Play could continue} Qe7 18. Rad1 Rad8 19. Nxd6
Rxd6 20. Rxd6 Qxd6 21. Rd1 Qe6 22. Qa4 a5 23. e4 $14 {and White has nagging
pressure along the d-file and on the queenside.}) 11. dxc5 (11. d5 {isn't as
convincing here:} exd5 12. Nxd5 Bxd5 13. Bxd5 Nxd5 14. Qxd5 Nf6 15. Qe5 Re8 16.
Rad1 Qc8 {and White has no edge.}) 11... Rc8 (11... Qc8 {is slightly better
for White - see 8...c5 9.dc5 dc4 10.Bc4.}) (11... Nxc5 12. Qc2 Rc8 (12... Qc8
13. Rfd1 $14) 13. Rad1 Ncd7 14. Qe2 $14) 12. Qe2 (12. Qc2 $6 {is inferior due
to} Bxf3 13. gxf3 Rxc5 14. Bd3 Nd5 $1 {when I prefer Black's position.}) (12.
Rc1 $5 Nxc5 13. Qc2 Nce4 14. Rfd1 Qe8 15. Nxe4 Bxe4 16. Qb3 {and Rybka
underestimates White's chances - he has a stable advantage because Black's
f8-rook is out of the game, and his pieces have no targets while Black's
queenside pawns are vulnerable to attack. White can pile up the pressure with
Ne5, when c6 is quite weak.}) 12... Nxc5 (12... Bxc5 $6 13. Ba6 $1 Bxa6 14.
Qxa6 $16) 13. Rfd1 Nd5 (13... Qe8 14. Ne5 $14) 14. Rac1 Nxc3 (14... Bf6 $6 15.
Nb5 $1) 15. Rxc3 Bd5 16. Bxd5 exd5 {and White can keep a small edge in a few
ways, the best of which is probably} 17. Be5 $1 $14 *
Conclusion: Black has a solid position in the 6...b6 line, but with accurate play White keeps a small edge. Bryan's ideas are interesting but no better or worse than Kaufman's suggestions.