Oh, sure. That is why I wrote "for the time being".
I would like to expand on the subject. History buffs should know, that Rubinstein did not invent 4...Nd4. There are at least 9 games (3 of them seeing Marshall handling the black pieces) before 1912. Of course Rubinstein delivered some major contributions. These led to the conclusion, that 4...Nd4 is better than 4...Bc5 5.Nxe5 Nd4/5.0-0 0-0 6.Nxe5 Nd4, another Marshall idea.

We all know, that there are two "problems" with 4...Nd4.
1) White may chose the most boring line of the entire opening theory: 5.Nxd4 exd4 6.e5 (this move is almost an insult) dxc3 7.exf6 Qxf6 8.dxc3 Qe5+ 9.Qe2. I think even TN will trust statistics here: more than 80% draws. You don't want to play this against a weaker opponent.
2) White can play a slow manoeuvring game with 5.0-0 Nxb5 6.Nxb5 c6 7.Nc3.

So 4...Bc5 may be worth a look. The idea is to transpose with 5.Nxe5 Nd4 6.Bc4 Qe7 /6.Ba4 0-0. White has two independent ideas:
A) 5.0-0 0-0 6.Nxe5 Nd4 7.Bc4 d6 8.Nf3 Bg4 9.Be2 is Schlechter's recommendation, who denied Black all compensation. I am not so sure after Nxe2+ 10.Qxe2 Re8 11.d3 d5. Instead 10...Nh5 11.d3 f5 has been played a few times, but 12.d4! looks good.
B) 5.Nxe4 Nd4 6.Be2 (6.0-0 0-0 see A) d5 7.Nd3 Nxe4 (Fine's idea) 8.Nxe4 dxe4 9.Nxc5 Qg5 10.Nxe4 Qxg2 11.Ng3 (Black's piece sac is incorrect, says Keres) h5! and White faces some problems. William Hartston 30 years ago wrote a book called "Black countergambits". This contains some analysis, which I alas have lost. Three sample lines:
B1) 12.c3 h4 13.Bf1 Qc6 14.cxd4 hxg3 15.f3 Be6 followed by 0-0-0 and/or Rxh2.
B2) 12.h4 Bg4 13.Bxg4 hxg4 14.d3 Nf3+ 15.Ke2 Nd4+ with (more than?) enough compensation.
B3) 12.Bxh5 Rh6 13.d3 Re6+ 14.Be3 Rxe3+ and Black is better.

So is Marshall's version a risky, but interesting option to head for the Rubinstein Gambit?

MNb wrote on 02/07/07 at 01:34:15:


I am no real expert in this line nor do I place much faith in stats from a small sample (sometimes even large samples), nevertheless the idea behind 8...c6 seems to be to avoid 8...Ne8 9.Nd5 favored by Motylev, although its not clear just how dangerous this is for Black.      

Following up on your 10...d5!? idea, which I think might be a viable alternative depending on the evaluation of 11.Nf4 Qf6 12.Nfxd5 cd5 13.Bxe8 Be6 14.Ba4 Nb3 15.Qf3 Qxf3 16.gf3 Nxa1. This may all be mute however as 10...Qf6 seems convincing enough.   

Ok it seems that this Knight sacrifice is in fact too poisonous to accept, and should 8....Ne8 9.Nd5 prove troublesome for Black then this is a useful line to know.

Topper    

Willempie wrote on 02/06/07 at 15:58:32:


I think after your 13...Bf5! White is busted.

The longer you analyse White's position the more you realise just how bad it really is.

Topper  

I don't have a board, but after 11.h3 d5, is 12.Nxd5 possible?

I suspect that if White has to give back the piece with 11.Nf4 then it is well and truly over for him, as this leaves him with no compensation at all for his rotten position.

Maybe he has try some combination of h3 and g4 at some point or perhaps h3 without g4. I haven't analysed these possibilities fully yet but I remain doubtful as to White's survival chances.

Topster

To get the ball rolling, and after having given this position a fair bit of analysis since posting the original query in the Corus A thread, I have come to the conclusion that acceptance of this Knight sacrifice is not just risky but bordering on suicidal.

Here is a sample variation of what I believe to be the critical line to illustrate the point: 

1. e4 e5 2. Nf3 Nc6 3. Nc3 Nf6 4. Bb5 Nd4 5. Ba4 Bc5 6. Nxe5 O-O 7. Nd3 Bb6 8. e5 c6 9. exf6
Re8+ 10. Kf1 Qxf6 11. g3 d5 12. Kg2 Re2!! 13. Nxe2? [However 13. Nf4
Bg4 14. Ncxe2 Bf3+ 15. Kf1 Re8 -/+ isn't much better for White] Qf3+ 14. Kg1 Bh3 15. Ndf4 Nxe2+ 16. Qxe2 Bxf2+ 17. Qxf2 Qd1+ 18. Qf1 Qxf1 Mate

Since natural moves seem to fail, it isn't clear to me how White will manage to successfully untangle, co-ordinate and ultimately consolidate his extra piece.

What say you guys? 

Tops