Thanks FightingDragon!

Here's a brief summary of the 16.h4 games I found in the IECG database (where I've highlighted the Qc5/Be4 deployment that I mentioned earlier):

1. e4 c5 2. Nf3 d6 3. d4 cxd4 4. Nxd4 Nf6 5. Nc3 g6 6. Be3 Bg7 7. f3 O-O 8. Qd2
Nc6 9. O-O-O d5 10. exd5 Nxd5 11. Nxc6 bxc6 12. Bd4 Bxd4 13. Qxd4 Qb6 14. Na4
Qa5 15. b3 Qc7 16. h4 Bf5 
 (16... Qg3 see Tasic-Daurelle, IECG email 2005/1-0 (57)) 
 (16... h5 see Daurelle-Koch, IECG email 2005/1/2-1/2 (32)) 
17. g4 e5 18. Qc5  
 (Attempting to improve on 18. Qd2 
 see Kunzmann-Kerndl, IECG email 2001/0-1 (67)) 
18... Be6 19. Bd3 
 (19. g5 Rad8 20. Bd3 Nf4 21. Be4 Bd5 22. Bxd5 Nxd5 23. h5 Rfe8 24. hxg6 fxg6 
 25. Nc3 Qb6 26. Ne4 Qxc5 27. Nxc5 Nf4 28. Nd7 Kg7 29. Rd6 Re7 30. Rxh7+ Kxh7 
 31. Nf6+ Kg7 32. Rxd8 Ne6 33. Rd7 Kf7 34. Rxe7+ Kxe7 35. Ne4 Nd4 
 36. Nd2 Ne2+ 1/2-1/2 Bergmann-Kerndl, IECG email 2001) 
19... Nf4 20. Be4 Rac8 21. h5 gxh5 22. Nc3 f6 23. gxh5 Kh8 24. Rd6 Qe7 25. Rhd1 Bf7 26. Bf5 Rc7 27. h6 Bd5 Pijl-Kerndl, IECG 2001/1-0 (72).

FightingDragon, were the Kerndl games the correspondence games you were referring to which continued with 16.h4 Bf5 17.g4 e5 etc.?

Maybe 16...h5 is a viable alternative to 16...Bf5?

Smyslov_Fan wrote on 07/10/06 at 04:53:50:

Generally, the knight stays on a4 and they proceed with the pawn storm.  Black often plays ...e5 at some point, and in some cases it may be useful to have the c5-square unoccupied so the queen can go there where it puts pressure on c6 and continues to keep an eye on c3 (as long as there is no Bc4). 

Quote:
Looks like the bishop never goes to c4  in the 16.h4 line (not in these games anyway).  One useful maneuver which appears in a couple of games is Bd3-e4.  In conjunction with Qc5 this puts some pressure on c6 leading either to an exchange of light-squared bishops (which seems like a good exchange for White) or ties down Black's major pieces to the defense of the c6-pawn.  The deployment Be4/Qc5/Na4 (i.e., N staying put) looks pretty effective in the games where it occurs.  Also having c2-c4 available to hit the Nd5 might be another reason for keeping the bishop off c4.

I've started to look at 16.Nc5 (it certainly makes sense to get the 'dim' knight back into play) and a possible drawback is that it is easy to attack the knight with ...Qe7 either forcing the knight to move again or burdening the white queen with a defensive chore where it has to give up control of a key dark-square (i.e., Qf2 weakens c3 while Qa5 weakens e3).  There are still concrete lines to work out so too soon to draw conclusions.  However, I wonder if the Nc5 outpost works best after queens have been exchanged.

Smyslov_Fan wrote on 07/10/06 at 01:17:31:

Smyslov_Fan,
Great to hear from you in this thread and much obliged for the input - I need all the help I can get!  You make a good case for giving 16.Nc5 serious consideration.

Scholar,
I guess you are right that there are a number of candidate moves to consider for White.  I've no experience at all in correspondence chess (and still need a bit of work in learning how to analyze properly).  I think I've developed some bad habits for choosing candidate moves too narrowly and superficially.

Now comes the hard work: analyzing each and making a choice (I'm happy enough with 16.h4 so far, but need to dig a bit deeper and look at the other possibilities) ...

BTW, I found 6 games in the IECG archive (links below) where 15...Qc7 was played (compared to only a single OTB game found searching the Chessbase 'Big Database 2006').  16.Bc4 was played in one game [Ratings W-1036, B-1577, result: draw] and 16.h4 was played in the other five games [Average Ratings W-2265, B-2297, results: (1-0, draw, 0-1) = (2, 2, 1)].  With such a small number of games, of course there are no meaningful statistical conclusions to draw (if there ever are).  Just the same, I thought it was interesting that in all the higher level IECG games 16.h4 was preferred over Van den Doel's 16.Bc4.

IECG home page: http://www.iecg.org/
IECG games archive: http://pasparis.club.fr/current.html

OstapBender wrote on 07/08/06 at 22:13:41:


Well, for example: h4, Bc4, c4, g4, Nc5 -- I'm not saying that all of these are good, but a few of them seem pretty solid, and certainly all of them need to be considered.

1.e4 c5 2.Nf3 d6 3.d4 cxd4 4.Nxd4 Nf6 5.Nc3 g6 6.f3 Bg7 7.Be3 0-0 8.Qd2 Nc6 9.0-0-0 d5 10.exd5 Nxd5 11.Nxc6 bxc6 12.Bd4 Bxd4 13.Qxd4 Qb6 14.Na4 Qa5 15. b3 Qc7

http://www.france-echecs.com/diagramme/imgboard.phpfen=r1b2rk1/p1q1pp1p/2p3p1/3n...

The moves under consideration are 16.Bc4 and 16.h4 (but I'm open to other suggestions).  Does everyone agree that 16.Qc5 (compared to 13...Qc7 14.Qc5) is a pointless gesture here - so there are some clear difference in White's choices resulting from the inclusion of b3 and Na4.

Scholar, I wonder what your thoughts are on the key differences between 16.Bc4 and the other ...Qc7 line of 13...Qc7 14.Bc4.  If you want to wait until I choose my next move, this is understandable (I guess you shouldn't have to play both sides of the board ).

By the way, thanks to FightingDragon for the suggestions on White's responses to 15...Qc7.

FightingDragon wrote on 07/04/06 at 09:12:14:


Below is the full game van den Doel-Rogers, Dutch Cht 2004:

1. e4 c5 2. Nf3 d6 3. d4 cxd4 4. Nxd4 Nf6 5. Nc3 g6 6. Be3 Bg7 7. f3 Nc6 8. Qd2 O-O 9. O-O-O d5 10. exd5 Nxd5 11. Nxc6 bxc6 12. Bd4 Bxd4 13. Qxd4 Qb6 14. Na4 Qa5 15. b3 Qc7 16. Bc4 Rd8 17. g4?! Be6 18. Nc5 Nf4 19. Nxe6 Nxe6 20. Qe3 Nf4 21. h4 Ng2 22. Qe4 Qf4+ 23. Qxf4 Nxf4 24. Rde1 e6 25. Re4 Nd5 26. Kb2 Nf6 27. Rf4 Kg7 28. Bd3 Rd6 29. Re1 Rad8 30. Rc4 Nd5 31. Be4 h6 32. Rc5 Nf6 33. b4 Nd5 34. Kb3 Ne7 35. a4 f5 36. Bd3 Kf6 37. Bc4 Nd5 38. b5 cxb5 39. axb5 R8d7 40. Bf1 Nf4 41. Rc8 Rd2 42. Bc4 Rf2 43. Re3 Rh2 44. h5 fxg4 45. fxg4 gxh5 46. Rf8+ Rf7 47. Rxf7+ Kxf7 48. Re4 Rh3+ 49. Kb2 Nd5 50. Bxd5 exd5 51. Ra4 Ke6 52. Rxa7 Kd6 53. Ra6+ Kc5 54. Rxh6 hxg4 55. Rg6 Kxb5 56. Rxg4 1/2-1/2

Based on what FightingDragon (and Golubev?) say, it looks like White needs to improve upon 17. g4 (highlighted above).

FightingDragon, do you have any games or analysis featuring 16.h4 (l like the look of this move)?  I will see if I can find any in the IECG correspondence database (since the 'Big Database 2006' I got with Chessbase doesn't seem to include correspondence games).

Scholar wrote on 07/08/06 at 03:54:08:

Scholar,
I'm not so much saying that I think you're wrong about 15...Qc7 as I am proclaiming that I'm too obtuse (perhaps acuity will come with experience) to entirely understand your reasoning.

As for critical lines, who knows, but here is what I've seen recommended and/or played:

after 13...Qc7 there are 
14.Qc5 seems to be Golubev's recommendation, but maybe just because 14.Bc4 is more complex (less clear?). Now
14...Nxc3 [more common than 14...Qf4+ 15.Kb1 Bf5, which also might critical] 15.Qxc3

14.Bc4 (recommended by Dearing) 14...Nb6 [or 14...e5 15.Qd2 (15.Qc5 Nxc3 16.bxc3) 15...Nb6 (15...Be6)] 15.Be2 [I've also seen 15.Ne4 but haven't looked at it to form an opinion.]

15...Qc7 is much less explored.  In the only game I've seen mentioned (Van den Doel-Rogers, Dutch Cht 2004) White doesn't get much after 16.Bc4 Rd8 but this seemed to have more to do with subsequent play than any defect with 16.Bc4 (maybe it's less effective here than in the line 13...Qc7 14.Bc4 since the white knight has moved to a4?).  I also like the idea of playing 16.h4 leading into typical kingside pawn play.  No critical lines to reveal here.

My comment that I didn't see 15...Qc7 as an improved version on 13...Qc7 was based on my impression that forcing White to play b3 (often played later anyway) and Na4 (usefully eyeing c5) doesn't seem to improve Black's prospects.  I understand your point about Na4 removing some pressure from d5, but I'm not entirely appreciating the significance of this.

Of the three moves now under consideration for Black:

I think that 15...Bf5 would be the most difficult to face, but still think there should be prospects for White in either the 16.g4 or 16.Qc5 lines (whether I can uncover these is an entirely different issue).  Chris Ward's suggestion that Black looks OK these lines notwithstanding - I like Chris's section a lot, but Ithink he may be just a tad biased towards Black in his assessment of Dragon positions.  Right now I'm leaning toward 16.Qc5 (which, until proven one way or the other, seems more critical than 16.g4).

I think that 15...Be6 might be the least interesting of the three moves - although it could be quite important if the endgame after 15...Bf5 16.Qc5 is shown to be dangerous.

I think that 15...Qc7 is in many respects the most interesting choice, particularly considering the differences of opinion in this thread regarding its merits relative to the more straightforward 13...Qc7.  I would probably continue with 16.h4 (although I certainly need to dig a bit more deeply to be sure) and my gut feeling is that Black's defense against the typical kingside attack is not so easy.

The Delchev-Cebalo (2005) annotated by Ward on Chesspublishing.com features the 15...Bf5 16.g4 Be6 17.Qe5 variation.  Ward indicates that Black looks OK in the game as well as after 16.Qc5.  White appears to be the one in need of an improvement to be able to fight for an edge.  Maybe this is one of Black's better options versus 9.0-0-0 d5 10.exd5.  If this holds up for Black, perhaps someone will want to try the White side of 9.0-0-0 d5 10.Qe1 or 10.Kb1.