The pawn stands slightly more actively, taking c4 from the white knight, and opening a path to b6 for the bishop.

It was useful to make luft.

Losing at once. He could resist further with 33.h3! Qa3 34.Bc3.

White resigned.

The knight jump into d3 is natural and strong – Black gets the two bishops and active piece play. But in view of the fact that White will have to part with his light-squared bishop, Black first provokes a weakness:

If 17.bxc5 Bxa1 18.cxd6 Qxd6 19.Nxb7 Qe7 20.Nc5 Rb8 21.Nb3 Qf6 22.Kf2 Bc3 23.Qc2 0-0, White does not get full compensation for the exchange.

Better was 22.bxc5, strengthening the position in the centre. However, because of the weakness of the kingside light squares, the advantage would still lie with Black.

With such a bishop, White has to surrender a rook, and the rest is a matter of technique.

White resigned.

The assessment of such positions depends on how well White controls d5. If the white bishop were on b3, instead of e2, the only way for Black to get decent play would be to sacrifice the exchange on c3, and then take on e4 – once again, the whole battle revolves around the central squares. However, with the bishop on e2, Black can eliminate the blockade of d5 with a pawn break: