I have played this game as Black against computer( difficulty was set to HARD ). I was trying to test Alatortsev line I found in ECO D, 1987 that claimed Black was equal:

[Title "Taken from ECO D ( published 1987 )"]
[fen ""]

1.d4 d5 2.c4 e6 3.Nc3 Be7 4.cxd5 exd5 5.Bf4 c6 6.Qc2 Bf6 7.e3 Ne7 8.Bd3 g6= 

saying "with the idea of Bf5".

After g4 my position became barely tenable.

Is there something I could have done better in order to reach full equality?

How does modern theory evaluate 8...g6?

Here is the game:

[White "Shredder"]
[Black "AlwaysLearningNewStuff"]
[fen ""]

1.d4 d5 2.c4 e6 3.Nc3 Be7 4.cxd5 exd5 5.Bf4 c6 6.Qc2 Bf6 7.e3 Ne7 8.Bd3 g6= 9.Nf3 Bf5 10.Bxf5 Nxf5 11.g4! $40 Ne7 12.Qb3 Qb6 13.Qa3! Na6 14.g5 Bg7 15.O-O-O Qb4! $7 16.Qxb4 Nxb4 17.a3 Na6 18.e4! $16 $40 O-O-O 19.Be5! Bxe5 20.Nxe5 Rhf8 21.exd5 Nxd5 22.Ne4! Nac7 23.Rhe1 Ne6 24.h4 Kc7 25.Nc5?! Nxc5 26.dxc5 f6!= 27.Nd3 Rde8 28.Kc2 Kd8 29.Rxe8+ Rxe8 30.Nf4 fxg5 31.Nxd5 cxd5 32.Rxd5+ Kc7 33.hxg5 Re2+ 34.Kc3 Rxf2 35.b4 Rf3+ 36.Rd3 Rf5 37.Rh3 Rxg5!= 38.Rxh7+ Kc6 39.Rh8 a5 40.Kb3 axb4 41.axb4 b6 42.cxb6 Kxb6 43.Rb8+ Kc6  44.Kc3 Rb5 45.Rc8+ Kb6 46.Kc4 Rf5 47.Rb8+ Kc6 48.b5+ Kc7 49.Ra8 g5 50.Ra7+ Kb6 51.Ra6+ Kb7 52.Rg6 Re5 53.Kb4 Rf5 54.Ka5 Re5 55.Rg7+ Kb8 56.Ka6 Re6+ 57.b6 Re8 58.Rb7+ Ka8 59.Rg7 Kb8 60.Rb7+ Ka8 61.Rg7 Kb8 62.Rb7+ 1/2-1/2
  • Impressive. I'm sure that you've run some analysis through Shredder. Why did it think that the sides were equal after the 9th move?
    – Tony Ennis
    May 13 '14 at 2:50
  • @TonyEnnis: You have misunderstood. The Chess Informant ECO D ( published 1987 ) gave = after g6 "with the idea of Bf5". After trying it out against Shredder online ( see the link in my post at the top ) I nearly got crushed! But since the edition of ECO is dated, I wonder if evaluation changed nowadays or maybe I have missed something in the game ( like Nh4 instead of Ne7 retreat ) The comments in the game are mine... May 13 '14 at 4:09
  • Probably you should think about 13... h5. White can't stabilize its pawn on g4 and it seems that black knight gets on f5. Black bishop can move diagonal f8-a3 to defen black square which are going to be attacked by white.
    – Kakadu
    May 13 '14 at 9:25
  • @Kakadu: But on ...h5 he will play g5 as in the game. I will be forced to play ...Bg7 as in the game, but now I will have h5 instead of Na6 which seems as a weaker choice... I just do not see how will I defend after o-o-o and then e4. Still, thank you for your contribution. May 13 '14 at 11:29
  • I don't understand the exclamation mark for 18. e4. I think it is a positional mistake, creating an isolated pawn on d4. After 18.e4, I think Black is immediately equal.
    – Wes
    May 13 '14 at 19:54

Deep Rybka's 4 opening book (with pre-calculated scores on positions in it) gives the position resulting by

    1.d4 d5 2.c4 e6 3.Nc3 Be7 4.cxd5 exd5 5.Bf4 c6 6.Qc2 Bf6 7.e3 Ne7 8.Bd3 g6

a +0.52 score, so definitely some advantage for White. It contains the following continuations:

[fen ""]

1. d4 d5 2. c4 e6 3. Nc3 Be7 4. cxd5 exd5 5. Bf4 c6 6. Qc2 Bf6 7. e3 Ne7 8. Bd3 g6 9. Nf3 Bf5 10. h4 (10. O-O Bxd3 11. Qxd3 O-O 12. h4 (12. b4 Nd7))

and several sub variations.

  • +1 Normally, I don't think it's a good idea to rely on computer evaluations, but here, from a human perspective, it can be said that Black's knight on e7 is misplaced, the knight on b8 is not yet developed, whereas all of White's minor pieces are already on good squares. So White does have a small advantage.
    – Wes
    May 14 '14 at 3:37
  • 3
    @Wes I am aware of the risks of relying on a computer, but to answer the OP's question "How does modern theory evaluate 8...g6?" I would put a strong computer's evaluation over any human's. With that being said, the computer doesn't give explanations for its score while humans do. May 14 '14 at 4:37
  • Human+computer > computer. Computers still lack the foresight of humans. In fact it is dangerous to rely on computer evaluations (even the best ones), because other human players have access to those very same computers and they can look deeper and "understand" (that's the key word) the key factors of the position while at the same time using the computer to analyze specific variations. So unless you can explain why it is advantageous for a side, it's generally a bad idea to say it's advantageous because computer evaluates it to be so.
    – Wes
    May 15 '14 at 0:11

I'll play.

So, first I checked on the 11.g4. Stockfish didn't care for it, instead preferring

[FEN "rn1qk2r/pp3p1p/2p2bp1/3p1n2/3P1B2/2N1PN2/PPQ2PPP/R3K2R w KQkq - 0 11"]

1.O-O-O O-O 2.Kb1 Nd7 3.g3 Nb6 4.h4 Ng7 5.Bh6 Re8 6.h5 Nc4 7.g4 Qd7 8.g5 Be7 9.Bxg7 Kxg7 10.Rdg1 Kg8 11.hxg6 fxg6 12.Ne5 Nxe5 13.dxe5[+.84]

So White has a bit of an advantage in that line.

After 11.g4?!, however, black retorts with (flip the board colors)

[FEN "rn1qk2r/pp3p1p/2p2bp1/3p1n2/3P1BP1/2N1PN2/PPQ2P1P/R3K2R b KQkq - 0 11"]

1... Nh4 2.Nxh4 Bxh4 3.Qb3 b6 4.O-O O-O 5.Rac1 Nd7 6.Ne2 Rc8 7.Rc3 Nf6 8.f3 h5 9.h3 Qe7 10.Qa4 Qd7 11.Qa6 c5 12.Be5 hxg4 13.hxg4 Rfe8 14.b3 Bg5[+.32]

g4 is a weak move because it begs Black to exploit the loose knight on f3. If the Knight does not capture, White is punished for not castling by Ng2+ followed by Nxf4 which nukes White's Bishop, isolates the d pawn, and removes White's opportunity to castle. And there's a fully open center file with Queens on the board.

But while Stockfish doesn't like 11.g4, it likes Black's response to it even less. After 11...Ne7 we see

[FEN "rn1qk2r/pp2np1p/2p2bp1/3p4/3P1BP1/2N1PN2/PPQ2P1P/R3K2R w KQkq - 1 12"]

1.g5 Bg7 2.h4 Qd7 3.h5 gxh5 4.Ne5 Bxe5 5.Bxe5 Rg8 6.Bf4 Ng6 7.Rxh5 Nxf4 8.exf4 Na6 9.O-O-O O-O-O 10.Rxh7 Rdf8 11.a3 Nc7 12.Rg1 Ne6 13.Ne2 Kb8[+1.45]

Black's sensible looking move cedes a real advantage. The problem is that Black is giving White too many tempos on the Kingside. At move 15, Black has not one piece past the second rank. His position is very passive and he has not castled. Bleh.

To somewhat answer the OP's original question, I think Black has decent chances after 11. g4 Nh4. However, I don't know if this counts as 'full equality' - that's up to the OP to decide.


I let the machine run all night at 10 minutes per move. Here's the game with the 11. g4 Nh4 variation. White is unable to convert his (its?) slight advantage. The game finishes with a draw by repetition.

[FEN " 1/2-1/2"]
1.d4 d5 2.c4 e6 3.Nc3 Be7 4.cxd5 exd5 5.Bf4 c6 6.Qc2 Bf6 7.e3 Ne7 8.Bd3 g6 9.Nf3 Bf5 10.Bxf5 Nxf5 11.g4 Nh4 12.Nxh4 Bxh4 13.Qb3 Qb6 14.Qxb6 axb6 15.Ke2 Nd7 16.a4 Be7 17.g5 f6 18.gxf6 Nxf6 19.f3 Kd7 20.Be5 Rhf8 21.Rhg1 Bd6 22.Bxd6 Kxd6 23.b4 Nh5 24.b5 Ng7 25.Rgb1 Nf5 26.bxc6 bxc6 27.Rxb6 Rfe8 28.Nb5+ Kd7 29.Rb7+ Kd8 30.Na7 Nxd4+ 31.Kd3 Re7 32.Rxe7 Kxe7 33.Kxd4 Rxa7 34.Kc5 Ra6 35.a5 Ke6 36.h3 g5 37.Ra2 Kd7 38.Ra3 h6 39.e4 dxe4 40.fxe4 Ke6 41.Kd4 h5 42.Ra2 Kd6 43.Ke3 c5 44.Ra1 Ke5 45.Rg1 Kf6 46.Rf1+ Kg6 47.Ra1 Kf6 48.Rf1+ Kg6 49.Ra1 Kf6 50.Rf1+ Kg6 51.Ra1 Kf6 52.Rf1+ Kg6 *

I prefer 6... Nf6 to Bf6. That would have put your N on its best square, and refrained from moving your B twice so early in the opening. Also, the N on f6 protects the h pawn, which your B does not.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.