MNb wrote on 08/09/11 at 21:56:14:


Black has abandoned castling queenside, sure, but Black hardly ever castles queenside anyway.

Whereas after 10 0-0 Bf5, aren't we just back in the solid variations (e.g. from Kramnik-Radjabov) that 9...Bf5 10 b3 was trying to avoid?

1.e4 e5 2.Nf3 Nc6 3.Bb5 f5 4.Nc3 fxe4 5.Nxe4 Nf6 6.Nxf6 Qxf6 7.Qe2 Be7 8.Bxc6 dxc6 9.Nxe5 O-O 10.b3 looks inconvenient after 10...Bd6 11.Bb2 Re8.
So 10.O-O and Black has avoided the b2-b3 idea, but abandoned the option to castle Queenside. After 9...O-O 10.O-O there are other moves than 10...Bf5 transposing, but I don't see how they should improve Black's chances. 10...Bd6 11.Nc4 looks good enough for at least some advantage (after 11.d4 c5 12.Be3 Bxe5 Queens will be exchanged, increasing Black's drawing chances).

Jonathan Tait wrote on 08/08/11 at 07:15:21:


An interesting idea. The bishop might for sure be better off somewhere else, since apparently ...Bxc2 is not such a threat. What about 10.b3 anyway?

Vass wrote on 08/07/11 at 20:38:31:

True, 6...Nd5 seems to be a sound alternative, but White still has some options left for inquiry. In the case of 6...Ne4 the simplifications may be more forced. In the 7.Qe2 line, Sokolov's analysis isn't very helpful. Maybe the following gives White some hope. Of course the extra pawn at the end is almost meaningless. But remember Hübner - Ftacnik.

Jänisch-Schliemann thus far:



There is some work left (and alas I now don't have the time for it - end of the schoolyear), but this looks more or less like what we want for White.

I don't mean to be obstructive or dogmatic. If we can prove += with 4.0-0, I would be very happy to go that way, since it's much more economical.

Btw, if we go with 4.c3 f5 we'll have to prove that Khalifman is not right giving += to this whole variation in his book "Openings according to Anand 1.e4, vol 1" (2004).
Hard task!.. 
Edit: So, for now let's stick to 4.c3 Nf6 which gives black better chances for =, imho.

Markovich wrote on 08/06/11 at 23:43:16:

Many of these 36 games were Computer games, I took MegaBase, Corrgames, some of the better collections of engine-games etc. - I checked 14.a4 Be6 at first, but after 15.Nc3 a6 16.Bf1 it seemed to me that the weakness b3 gave Black full compensation without trouble. 25 ply deep, the evaluation is +0.08. Considering the lasting pin and Black's activity, it seems "fair". - My analysis was 14.Bf1 Bd7 15.Nc3 Rac8 16.Bg5 (16.Bf4 Bf5; 16.Bd3 Ba4) 16...f6 17.Bf4 g5 (17...Bf5 18.Qb3+) 18.Bg3 f5 and so on. The computer says: 0.00.   

Nobody is urging you to give up 4.c3 prematurely. There may be hidden improvements in the Ne4 line. One of the advantages of 4.0-0 would be to avoid 4.c3 f5, which isn't entirely harmless. If you are not impressed by 4.c3 f5, but still think that 4.0-0 Nd4 is unpleasant (in my opinion White does have a slight plus), we might prefer 4.c3 Nf6 5.0-0.