PANFR wrote on 06/27/12 at 12:54:51:


No, 9.g4 is not a forced win for White; he has only a slight advantage! 9...Be6 10.Ne6 fe6 11.0-0-0 Ne5 12.Be2 Qc8 is the main line but I'm not convinced 12...Rc8 is inferior.

Quote:


Well, I would probably agree that Black's outlook after 15...b5 are not great. After all Black has a difficult defensive job ahead and White's attack develops more easily. However the computer estimates the position as almost equal at the beginning and there might be lots of alternatives for Black, which at first seem slightly inferior, but might eventually rehabilitate the line. Like it has happend to many dragon lines before. So we can only use our intuition and speculate.

For curiosity: What are correspondence chess dragoneers do about 13.h4 h5 14.g4 etc. like Anand and Topalov played against Carlsen?

Okay, I had a look at 15...e5 and it definitely appears to be a game-saver. However, I think 16.Bxc4 Rxc4 17.Nb3 is better than 17.Nf5 in view of the analysis given in Roland's replies. Say something like 17.Nb3 Be6 18.Bg5 Qc7 19.Bxf6 Bxf6 20.Nd5 Bxd5 21.Rxd5 and white can try to work on the backwards d-pawn. It's not much, but I think it's safer and more worthwhile than allowing black to exchange-sac on c3 in the 17.Nf5 line.

At any rate, I do agree 15...e5 saves the game for black, and I will keep this in mind for future reference.

Interesting.

But is it you that originally suggested 15 ...e5!? (or Roland) and is from 2011 long time ago 

However 15.Qc1 had been played OTB earlier than the games you gave (by a GM and an IM, that I know of).

Nice game, Qc1 looks very nice for white!

Although not the most popular line in the Dragon, the 12...a6 variation had been 'playable' even on ICCF. However, I found a rare novelty that seemed to upend the entire variation for black. Here's a re-post of my report on the subject:

I recently employed a refutation in one of the Dragon lines using the Yugoslav Attack on an ICCF Master Norm tournament. I researched my opponent's previous games and found he always played the exact same pet line in the Dragon. From there, I found an extremely rare novelty that had only been played twice, but both games were won by white. My game would prove to be the most challenging resistance by black, but the refutation still held and I ended up winning. Here's the game PGN with notes on the critical points:

[Event "WS/MN/083"]
[Site "ICCF"]
[Date "2012.5.2"]
[Round "-"]
[White "Morrow, Wolff"]
[Black "Grego, Luís José Gonzaga"]
[Result "1-0"]
[WhiteElo "2301"]
[BlackElo "2316"]

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.Bc4 Bd7 10.O-O-O Rc8 11.Bb3 Ne5 12.Kb1 a6 { This was the pet line my opponent always played for in his previous games. I knew this, and prepared the 'gem' refutation for him on move 15. } 13.h4 h5 14.Bh6 Nc4 15.Qc1! { This is the new move that kills the entire 12...a6 line. It was played first in Salvador Marques - Alberto da Silva January 2011, and later in Walter - Mesquita Jr. July 2011. In both games, black lost, though not with the best analysis. My opponent would put up maximum resistance, which involved heavy computer analysis. } ( 15.Bxc4 { This is the standard main line, but black has resources to hold at least a draw. } 15...Rxc4 16.Bxg7 Kxg7 17.Rhe1 Qa5 { And the games end up drawing for black in the ICCF archives. } ) 15...b5 16.g4 { Now white gets the pawn march going without black being able to untangle his pieces in time. } 16...e6 { The best try coming from hours of computer analysis. Unfortunately for my opponent, it's still not enough to save the draw. } 17.gxh5 Nxh5 18.Rhg1 b4 19.Nce2 Qf6 20.Rdf1 d5 21.Bxg7 Nxg7 22.Bxc4 dxc4 23.f4 { White's pawn storm is significantly more potent than black's. } 23...b3 24.h5 bxc2+ 25.Qxc2 Nxh5 26.f5 Qg7 27.Rg5 c3 28.bxc3 Qf6 29.Rfg1 e5 30.Rxh5 exd4 31.Nxd4 Qb6+ 32.Qb3 Qxb3+ 33.axb3 Rxc3 34.fxg6 fxg6 35.Rxg6+ Kf7 36.Rxa6 { After the dust has settled, I reach the endgame a clear two pawns up. This is technically lost for black, but it requires some finesse to force the win. } 36...Bc8 37.Rb6 Rd3 38.Nc6 Rd2 39.Ne5+ Ke8 40.Rh1 Rg8 41.Rc1 Bh3 42.Ng6 Rg7 43.Rh1 Rh7 { After this position, I knew I had the win in hand. I can force a Tablebase ending immediately. } 44.Rb8+ Rd8 45.Rxd8+ Kxd8 46.Kb2 Rg7 47.Rxh3 Rxg6 48.e5 { The position is a 6-piece Tablebase ending with white delivering checkmate on move 80. I announced this to my opponent, and he confirmed it and resigned. } 
1-0