I am trying to find a line in the Dunst Opening where:

  1. White's guaranteed to develop his bishop to an active square such as c4 or b5.
  2. Black doesn't get to suffocate White by playing d3.

So far I have been unsuccessful. Here are two lines I've found for Black that essentially refute White's whole setup.

This is the line with 5.Nc3:

[FEN ""]

1. Nc3 d5 2. e4 d4 3. Nce2 e5 4. Ng3 Be6 5. Nf3 f6 6. Bb5+ c6 7. Ba4 Na6 8. Bb3 (8. O-O? Nc5 9. Bb3 Nxb3 10. axb3 d3!) (8. d3? Qa5+) Bxb3 9. axb3 d3! 10. O-O Nb4

5.c3 has another refutation:

[FEN ""]

1. Nc3 d5 2. e4 d4 3. Nce2 e5 4. Ng3 Be6 5. c3 a6! 6. Nf3 Nc6
  • f6 looks risky. As White I would be tempted to reply with Nxe5 followed by Qh5+. Even if that is premature, it shows that Black is taking on some risks by defending e5 without the knight. Commented Jul 19, 2021 at 12:38
  • As this is feasible for black, it should be better for white. You might have to strike with f4 before Ng3. It could be that the whole opening is unsound and the great masters of yesteryear didn't play this for that reason.
    – Mike Jones
    Commented Jul 19, 2021 at 19:20
  • it's not good. I use to play it against the scandinavian, but poor results and engine evaluations convinced me that it's not a good way for white to play for an advantage. Commented Jul 20, 2021 at 23:17

3 Answers 3


OK, so after a bit more research I have concluded that this variation is in fact playable for White, even with perfect play from Black! Here's my analysis:

The critical line is

[FEN ""]
1.Nc3 d5 2.e4 d4 3.Nce2 e5 4.Ng3 Be6 5.Nf3 f6 6.c3

In this position, we're threatening to win a pawn with 7.cxd4 exd4 8.Qa4+ Nc6 9.Bb5. Black has a couple of ways to parry this threat: 6...c5, 6...Nc6, 6...a6 and 6...Qd7.

  • If Black plays 6...c5, he immediately gives our bishop a comfy square with 7.Bb5+. After we play 8.d3 on the next move, all of our major opening headaches are pretty much gone.

[FEN ""]
1.Nc3 d5 2.e4 d4 3.Nce2 e5 4.Ng3 Be6 5.Nf3 f6 6.c3 c5 7.Bb5+ Nd7 8.d3
  • The most testing move is 6...d3!, trying to completely suffocate our position. Against this we can sacrifice our knight for 3 pawns:

[FEN ""]
1.Nc3 d5 2.e4 d4 3.Nce2 e5 4.Ng3 Be6 5.Nf3 f6 6.c3 d3! 7.Nxe5!? fxe5 8.Qh5+ Bf7 9.Qxe5+ Ne7 10.Qb5+ Nd7 11.Qxd3 

Objectively, White is much worse here according to the engine, but this seems fine to me for faster time controls.

  • Against 6...Nc6, we do get to play 7.Bb5. Now if Black doesn't want to let us consolidate with 8.d3, he has to play 7...d3 himself. Here we just 8.O-O. Now we're already threatening to win the pawn with 9.Ne1, so Black has to immediately question our bishop with 8...a6. After 9.Bxc6 bxc6, we play 10.Qa4, developing with tempo on the c6-pawn. After Black defends it, we can play 11.b3 with the idea of developing our entombed bishop to a3.

[FEN ""]
1.Nc3 d5 2.e4 d4 3.Nce2 e5 4.Ng3 Be6 5.Nf3 f6 6.c3 Nc6 7.Bb5 d3 8.O-O a6 (8...Qd7? 9.Ne1) 9.Bxc6 bxc6 10.Qa4 Qd7 11.b3

(Black did get to play d3 in this variation, but at least we aren't getting suffocated here.)

  • Against 6...a6, we can play 7.cxd4 exd4 8.b3, which has the idea of trading off the bishop via 9.Bc4. Black doesn't have a good way to stop this: 8...b5? is met strongly by 9.a4!, and if 8...d3? then Black loses a pawn after 9.Bb2 followed by 10.Qb1.

[FEN ""]
1.Nc3 d5 2.e4 d4 3.Nce2 e5 4.Ng3 Be6 5.Nf3 f6 6.c3 a6 7.cxd4 exd4 8.b3 Nc6 (8...b5? 9.a4!) (8...d3? 9.Bb2 Nc6 10.Qb1 Nb4 11.Nd4 Qd7 12.a3) 9.Bc4
  • Against 6...Qd7, we can play 7.cxd4 exd4 8.Qc2, which prevents 8...d3 and has the same idea of trading off the bishops via 9.Bc4. Again, if 8...b5? then 9.a4!. However, note that if 8...Nc6 is played then we want to play 9.Bb5 instead of 9.Bc4, to avoid a nasty line where our c1 bishop can get buried. (9.Bc4 should be played against any other move.)

[FEN ""]
1.Nc3 d5 2.e4 d4 3.Nce2 e5 4.Ng3 Be6 5.Nf3 f6 6.c3 Qd7 7.cxd4 exd4 8.Qc2 Nc6 (8...b5? 9.a4!) (8...Kf7 9.Bc4) 9.Bb5 (9.Bc4? Nb4 10.Qb3 Bxc4 11.Qxc4 b5 12.Qb3 d3)
  • 3
    Why would anyone downvote this answer without giving a reason? Please show any error that you may have found.
    – Philip Roe
    Commented Jul 20, 2021 at 17:31
  • I also don't understand the reason for a downvote. I didn't check all of the analysis, but it seems reasonable enough. Commented Jul 20, 2021 at 17:57
  • 1
    I would like to add that after 1. Nc3 d5 2. e4 d4, white scores scores 54% wins vs 42% black wins according to lichess.org database. This is much better than average, which is almost equal.
    – Akavall
    Commented Jul 20, 2021 at 23:38

The opening 1.Nc3 is much better than people think, and I have had very good results with it, especially if I pick the opponent. The perfect choice is someone strong enough to try and refute it but not strong enough to do so. The line you mention with 4..Be6 is the only line that worries me, but it only appears if Black is either very well informed or else very perceptive and confident. The line with Nxe5 (John Coleman) is best prepared with c3 to provoke ..d3. That Pawn then gets picked up by the marauding Queen. The material is then three Pawns for a Knight and the better player will win, but maybe you don't want to take that risk. If you have to admit that, then your objective of an active Bishop development without allowing ..d3 is not achieveable. The best plan in my opinion is to give the check in your first line to provoke ..c6 (the Pawn is really wanted on c5). Then Ba4 is possibly a playable gambit after..Qa5+, c3 dxc3, 0-0!, or Be2 is not bad either. I hope you have read "Knight on the Left" by Harald Kielhack. This is one of the best general chess books I know, and much superior to the common "funny opening" booklets. The point is that 1.Nc3 has so many transpositional possibilities that the author has to deal with all of them and then discuss the differences made by the move order.

  • I think I actually found a way to get active bishop development while (mostly) preventing d3, see my answer below. Obviously though, the gambit line you mention sounds way more appealing / less convoluted, so I'll probably end up going with that. Also, thanks for pointing me to that book!
    – James Ko
    Commented Jul 20, 2021 at 4:43
  • 2
    I think you show pretty convincingly that the only move to give White any problem is 6..d3. I would be very happy to have White in any other line.Then the sacrifice is critical and as you say, the engines like Black. There have been a few correspondence games that seem to confim that, and White will be on the defensive for quite a while. But OTB, if you have played it before and Black hasn't, then White has an awful lot of pawns
    – Philip Roe
    Commented Jul 20, 2021 at 21:09

One thing to keep in mind is that practically this line is both extremely rare and fine for white against normal play. I'm around 1800, I've played 1.Nc3 d5 2.e4 d4 3.Nce2 e5 4.Ng3 559 times, of those I've seen 4. ... Be6 5.Nf3 f6 exactly 9 times (i.e. 1.6% of those games) and my record in those 9 games is 4 wins, 4 losses, and 1 draw.

You're right that against an engine this is not a very sound opening and you'll end up around -1 in these kind of lines (often with black playing h5 at some point), but even in those positions you're not lost and in practice against human opponents it's totally fine.

  • I'm curious how often you face 2..c6, transposing to a Caro. Something tells me that someone starting with 1. Nc3 probably did not want to play a Caro Kann :P Commented Feb 11, 2022 at 1:14
  • If you're a Caro-Kann player I'd recommend just playing 1.Nc3 d5, 2. e4 dxe4 and you're in a Caro-Kann like position without needing to commit to c6. At any rate I've seen 2. ... c6 in 76 games out of 988 (7.7%). It's not my favorite thing to see, but I don't mind it. Nowadays I play the two knights Caro-Kann, which is perfectly playable and usually as much outside of my opponents wheelhouse as mine. I've scored quite well with the two knights (61% winning percentage) even though I don't know it well. Commented Feb 11, 2022 at 1:22
  • By far the response that I score worst against is 1. ... c5. I'll play a Grand Prix against that, but I lose more games than I win. I used to hate 1. ... Nf6, but do fine since I've switched to the Vienna Gambit (it's nice not need to know the Nc6 variations in the Vienna). I'm also a little bit below even against the French, but that's mostly before I found a good line, and it's basically gone fine since I learned some better lines in the two knights French. Commented Feb 11, 2022 at 1:37
  • 1
    3 Nce2, presumably.
    – Rosie F
    Commented Feb 11, 2022 at 6:29

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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