1.d4
Nf6
2.c4
e6
3.Nf3
d5
Nisipeanu refrains from his main weapon against 3.Nf3, the Queen's Indian, probably finding himself under the impression of Topalov's brilliant achievements in this opening during the past years.
4.Nc3
Be7
But only this was meant to be the real surprise for Topalov. Prior to this game, Nisipeanu had played the Orthodox Queen's Gambit only once in his life, at the youth Balkaniad back in 1993. As a coincidence, his opponent in that game was also Bulgarian: Vladimir Georgiev. Topalov might have expected the Vienna variation (4...dxc4) or the Semi-Slav (4...c6), which occured occasionally in the European Champion's games along the past years. For instance, he defeated Topalov's second, Silvio Danailov, in the Meran in the B-group of the Wijk aan Zee festival in 2000.
5.Bf4
[From strict strategic point of view, the classical 5.Bg5
might seem sounder, because it increases the pressure over the d5-square. By choosing 5.Bf4, White usually gives up any ambitions regarding the influence over the centre and aims for animated minor piece play instead. We can draw a similar parallel in the open games, between the Ruy Lopez and the Italian Game. The modern history of 5.Bf4 started in 1978, when Kortschnoj repeatedly employed it during his World Championship match against Karpov. In spite of the fact that Karpov generally managed to hold his own, it could be felt that there was a lot of potential behind White's setup.]
5...c5
[The counter-attack in the centre based on the advance of the c-pawn is a logical consequence of the lack of pressure against the d5-pawn. However, Black usually inserts the moves 5...0-0
6.e3
and only now plays 6...c5
. Topalov had some recent experience against it: 7.dxc5
Bxc5
8.a3
Nc6
9.cxd5!?
(He also used the classical move 9.Qc2
, Kortschnoj's favourite, but after 9...Qa5
he chose the modern 10.Nd2
instead of the more established 10.Rd1. The game soon took a very interesting course: 10...Bb4
11.cxd5
exd5
12.Bd3
d4
13.0-0
Bxc3
14.Nc4
Qh5
15.bxc3
Nd5
16.Bg3
dxe3
17.Rae1
Be6
18.fxe3
Nde7
19.Nd6
b6
20.Rb1
and White's numerous pawn weaknesses were compensated by his hyper-activity in Topalov-Kasimdzhanov, Linares 2005. Black's queen soon got into big trouble, causing him decisive material losses.) 9...Nxd5
10.Nxd5
exd5
11.Bd3
Bb6
12.0-0
d4
13.e4
Bc7
14.Bxc7
Qxc7
15.h3
Qb6
16.b4
Be6
17.Rc1
h6
18.Qd2
with a more pleasant position for White in Topalov-Morozevich, San Luis 2005. The early attack of the centre by means of 5...c5 chosen by Nisipeanu has never made part of the main stream of theory, but suddenly became popular in 2005. The European Champion might have relied on the fact that his mighty opponent had awarded this variation less time during his home preparation than to the main lines.]
6.dxc5
Na6
Black intends to capture the c-pawn without the loss of tempo implied by Be7xc5.
7.cxd5
Nxd5
8.Nxd5
exd5
9.e4!?
[It appears that it is not so easy to surprise the World Champion in the opening! Topalov played this new move relatively quickly, leaving little doubt about the fact that he had prepared it at home. Previously, 9.e3
had been played, when after 9...Nxc5
10.Bb5+
Bd7
11.Bxd7+
Qxd7
12.0-0
0-0
White retains the more pleasant but not necessarily better position.]
9...0-0
However, Nisipeanu had done his homework properly, too, since he played this move rather quickly. We could hardly dream for a more promising start of the game.
10.exd5
Bxc5
This looks like a deviation from Black's initial intentions, but, given the new circumstances, the capture with the knight might have put less pressure on White, allowing him to keep his strong central pawn without having to make any major concessions. After the text move, the white king is vulnerable along the e-file and the e1-a5 diagonal. White has to act carefully.
11.Be2
[Played solidly. The attempt to accelerate the rhythm of his development with 11.Bxa6?!
would be a typical mistake, allowing 11...Qa5+!
12.Nd2
Qxa6
when White would have serious problems getting castled, while his d5-pawn would lack satisfactory defence. In case of 13.Qe2
, Black could maintain the tension with 13...b6!?
; The consolidation of the d5-pawn by means of 11.Bc4
would have allowed Black deprive his opponent from the right to castle in more economic way than in the game with 11...Re8+
. After the apparently modest development of the bishop to e2, it is obvious that Black has compensation for the pawn in view of his slight lead in development. However, finding the right path through the jungle of variations is far from easy for both sides.]
11...Bb4+
[Instead of trying to win his pawn back, Nisipeanu forces White to give up the idea of castling, hoping that this would ensure him long term compensation. Black had several ways of striving for material equality, but apparently only one of them would have been satisfactory. In case of 11...Nc7?!
White could safely give up his active bishop with 12.Bxc7
(12.d6
would not save the pawn after 12...Ne6
13.Bg3
Qa5+
14.Nd2
Qb4
eventually followed by ...Rd8 if for any reasons ...Bxd6 would not work.) 12...Bb4+
13.Nd2
Qxc7
14.0-0
and although Black can hope to hold on because of his firm blockade on d6, this looks like a rather one-sided affair.; Going for the b2-pawn with 11...Qb6?!
12.0-0
Qxb2
looks pretty risky in view of 13.Be5
when the black king side starts looking vulnerable, while the d5-pawn considerably restricts Black's activity.; The attempt to improve upon this last line with 11...Qa5+?!
12.Bd2
Qb6
13.0-0
Qxb2
would be refuted by 14.Rc1
followed by Bc3, with similar consequences.; However, the simple 11...Nb4
12.d6
Be6
suggested by IM Mircea Pavlov looks a satisfactory way of solving the opening problems. Black threatens ...Nd5 followed by the capture on d6. It is not very clear whether White could retain even some symbolic advantage in the symmetrical ensuing position. IM Francisc Nemeth suggested 13.Ng5
as a possible refutation of Black's plan, but after 13...Qf6
White risks losing stability. In a brief statement for the Radio made after the game, Nisipeanu mentioned that at some moment he deviated from his initial analysis because he thought that there would be a better possibility for him. Judging from the fact that he invested some time in taking the decision of giving check, we could assume that this was the moment he referred to.]
12.Kf1
Bd6
This pawn has to be safely blocked. Besides, the bishop was obviously misplaced on b4. White needs three whole tempi in order to get castled in artificial way, for instance h4, Kg1-h2. This strongly suggests that Black has at least adequate compensation for the sacrificed pawn. However, things are not that simple. First of all, White can do perfectly well without connecting his rooks yet, since he has the plan based on the advance of the h-pawn at his disposal. This could lead to the activation of the rook either directly (after h4-h5 and Rh4) or indirectly (in case of Black's ...h6, there would always be a possibility of a knight jump to g5 sustained by the h4-pawn, when the opening of the h-file would be dangerous for Black. Besides, Black's development is not that impressive yet, while his king's position is quite vulnerable. All these aspects make the position quite difficult to evaluate and, more important, to play over-the-board.
13.Bg5!?
[White cannot afford to allow Black to develop naturally. With his last move, Topalov underlines the fact that the enemy queen has problems finding adequate squares. The neutral retreat 13.Be3
would have offered Black a good position after 13...Bf5
followed by ...Rac8.]
13...Qa5
[Only ulterior analysis will reveal the objective merits of this move, but my personal feeling in this moment is that it is one of the main reasons for Black's ulterior problems. The queen will be very exposed on the queen side, allowing White the radical activation of his forces without loss of time. We can see a similar picture after 13...Qb6
14.Nd2
Qxb2
15.Nc4
Qb4
16.Rc1
eventually followed by Bd3 and Bd2-c3, when the black king would start feeling unsafe, while some of the other pieces are hanging.; 13...f6?!
is hardly a solution either. The weakness of the e6-square would leave White with a permanent advantage in case of returning the pawn by means of Nd4-e6, followed by g3 and Kg2.; However, the awkwardly looking 13...Qd7!?
deserves serious attention. Black plans to concentrate his forces against the d5-pawn with ...b6, ...Bb7, ...Nc7 and, if allowed, ...Rad8. This looks like a rather slow course of events, but we should not forget that White is also several tempi away from completing his development.]
14.a3
Nc7
[With the queen on a5, this move has obvious drawbacks, by cutting Her Majesty's retreat. In the press centre the normal developing move 14...Bd7
suggested by IM Andrei Cioara was considered to be a satisfactory continuation, and if 15.b4
then 15...Qa4
followed by the occupation of the c-file. Abstractly speaking, this looks very logical. Indeed, Black intends to take advantage by the passivity of the h1-rook by initiating active operation on the opposite wing. However, after 16.Nd2
followed by Nc4 or Ne4 Black would face concrete problems because of his lack of stability on d6.]
15.Bc4
b5
16.b4
Qb6
17.Bb3
Generally speaking, Black should be content with the fact that White has made absolutely no progress with the development of his king side yet. However, the quality of development of all his other pieces is better than Black's. For instance, the g5-bishop denies the rooks access to d8, while his colleague exterts some potential pressure along the a2-g8 diagonal, which could count in certain tactical lines. Comparing with the similar position that could have arisen after 13...Qd7, Black's position looks less harmonious. His queen is vulnerable and actually not so active on the left wing.
17...a5
A logical move, activating the a8-rook.
18.Be3
[Now that there is a point of tension on b4, White decides to transfer his bishop to the queen side. In the press centre, the main worries were connected with 18.Rc1
when after 18...Bb7
19.Be3
Qa6
20.Ng5
the threat Qd3 with a strong attack would be very unpleasant. In case of 20...h6
, besides jumping with the knight to e4, White could renew his initial threat with 21.h4
when Black's position looks critical. In certain cases, the intrusion of the rook to c6 would reveal the hidden force of the b3-bishop.]
18...Qb7
19.Bc5
[Topalov chooses a safe continuation, increasing his domination in the centre. The alternative would have been 19.Ng5
with the threat Qd3, when after 19...Bf5
20.g4
Bg6
21.h4
Black could survive with 21...Rfe8
22.h5
Be4
although after 23.Nxe4
Rxe4
White should be better. The only trap he has to avoid is 24.Qf3?
Nxd5
25.Qxe4
Nxe3+
winning tones of material for Black.]
19...Bxc5!!
[Nisipeanu played this completely unexpected move after long thinking. The objective merits of the exchange on c5 are hard to judge without thorough analysis, but the double exclaim mark attached to the move refers to Black's courage of changing the generally unfavourable course of events by radical and apparently highly risky means. Consolidating the d6-square would leave White free hands on the opposite wing, for instance 19...Rd8
20.Ng5
Bf5
21.Nxf7!
Kxf7
22.Qf3
g6
23.g4
and Black is in dire straits. The careless 23...Bxc5?
only makes things worse because of the discovered check 24.d6+
, winning the queen.; Or, similarly, 19...Ra6
20.Qd3
h6
21.h4
axb4
22.Ng5
f5
23.Bxb4
Bxb4
24.d6+
Kh8
25.d7
Bxd7
26.Qxd7
and now, in view of the threat Nf7+ followed by Qxf5+, Black has to play 26...Qc8
defending the f5-square and hoping for survival, although his position still looks precarious because of the weakness of the light squares around his king.]
20.bxc5
a4
21.Ba2
b4
22.axb4
Qxb4
This is the position Nisipeanu was aiming for. White's pawns look impressive, but because of his delay in development they could become vulnerable, too. In the following phase of the game, the World Champion proved his ability of adapting himself to the new circumstances by returning one of the pawns at the right moment in order to complete his development and maintain an edge.
23.Qd4
Na6
24.c6
Qxd4
25.Nxd4
Nb4
26.Ke2!!
[Even though we cannot draw any definitive conclusion about the concrete position on the board, I believe that it is safe to consider this to be the key move of the game. The d5-pawn will be lost, but for the first time in the game it will be White who will enjoy the better development, because of his more active king. Finding himself in slight time trouble already, Nisipeanu will fail to find satisfactory ways in what could be a defensible position. One of the merits of White's last move is that it refrains from "winning" a piece with 26.d6
Rd8
27.d7
. Indeed, after 27...Bxd7
28.cxd7
Rxd7
29.Nf3
Nxa2
30.Rxa2
Rd1+
31.Ne1
Re8
32.Re2
Ra8
White's developing problems would persist and Black would have little trouble forcing a draw, for instance 33.f3
Ra1
34.Kf2
a3
35.Rf1
a2
36.Nc2
Rxf1+
37.Kxf1
a1Q+
with complete equality.]
26...Nxa2
27.Rxa2
Ba6+
28.Kd2
Bc4
29.Ra3
Bxd5
30.c7!
White needs to keep the d8-square under control, preventing Rfd8 in order to ensure the knight's stability in the centre. Besides, this pawn will be a permanent threat for Black. If White will manage to install his knight on d6, Black would be completely paralysed.
30...Be6
[The regrouping initiated by this move fails to solve all Black's problems. Capturing the g2 pawn with 30...Bxg2
would have made some sense. Black would probably not be able to sacrifice the exchange for the c-pawn, but in certain cases giving up his bishop and starting a king side counter-attack might be just sufficient for a draw. The presence of the a-pawn would also cause White some loss of time during the process of defending his other wing. However, from a practical point of view it is rather difficult to take such a decision.]
31.Nb5
Bd7
[This move does not really cross White's plans. More stubborn seems to have been 31...Ra5
32.Rb1
Bd7
33.Nd6
Ra6
when the white king would be too far from the centre yet to ensure the stability of his knight.]
32.Nd6
g6
[Black decides to activate his king and avoid any back rank problems at the same time. The counterplay initiated by 32...Ra6
33.Rd3
a3
would fail to the simple 34.Nc4
Be6
35.Nxa3
]
33.Rc1
Kg7
34.Rc4!?
[White intends to keep both wings under control. Just as after the opening, Black has to permanently watch the possibility of possible rapid king side attacks. However, the immediate activation of the white king was also possible, for instance 34.Ke3
(Avoiding the d-file in order to prevent ...Ra6, which could be answered now by the simple c8=Q) 34...Kf6
35.Kd4
Ke7
36.Kc5
Ra6
and now the simple 37.Re1+
reduces Black's previous efforts of centralisation to nothing.]
34...Ra6
35.Rd4!?
[Another interesting moment. Topalov probably rejected 35.c8N
because of 35...Rd8
when Black could hope to give up one of his rooks for the enemy knights at the right moment and then defend his a4-pawn by placing his bishop on b3, which would lead to a draw.]
35...Ra7
36.Ne8+!
A simple tactical operation, leading to a very promising double-rook ending.
36...Rxe8
37.Rxd7
With the simple threat of Rd8
37...Raa8
[The most stubborn answer. In case of 37...Rc8
White would launch an irresistible attack with 38.Rf3
a3
39.Rdxf7+
Kh6
(39...Kg8
allows 40.Rf8+
when the c-pawn would promote.) 40.Rh3+
Kg5
and now 41.f4+
Kg4
42.Rg3+
followed by Rxh7 leads to mate.]
38.Kc3
The decisive centralisation.
38...Kf6
39.Kb4
Ke6
40.Rd4
[From practical point of view, this looks like the safest method. White threatens Re3+, followed by the exchange on e8 and Rd8, ensuring the promotion of the pawn. However, the apparently double edged 40.Rad3
would have been sufficient for a win, too after 40...a3
41.Rd8
a2
42.Rxe8+
Rxe8
43.Re3+
Kf6
44.Rxe8
a1Q
45.c8Q
and the white king will hide from checks on the eighth or seventh rank of the queen side. ]
40...Ke5?!
[This last move in time trouble makes things easier for White. At some moment, it was considered to be the decisive mistake in a difficult position, but apparently other moves did not save the game either. 40...Rec8
would have given White the possibility of displaying impressive rook activity with 41.Re3+
Kf6
42.Rf4+
Kg7
43.Re7
a3
(or 43...Rf8
44.Ka3
followed by Rxa4, with a winning rook ending. Black's king would be just one step too far.) 44.Rfxf7+
Kg8
(44...Kh6
leads to a curious mating attack after 45.Rxh7+
Kg5
46.h4+
Kg4
47.Re4+
Kf5
48.Rhe7!!
threatening g4 followed by R4e6 mate.) 45.Rg7+
Kf8
46.Rxh7
Kg8
47.Reg7+
Kf8
48.Rd7
Kg8
49.Rhe7
a2
50.Re1
when the least White will obtain would be a winning rook ending with two extra pawns.; Or if 40...f5
, clearing the f7-square for the king, then 41.Re3+
Kf7
42.Rd7+
Kf8
43.Rxe8+
Kxe8
44.Rxh7
again with an obvious win.]
41.Rc4
Rec8
[IM Mircea Pavlov suggested 41...Ke6
42.Kc5
Kd7
43.Kb6
Kc8
in order to block the pawn with the king, as generally recommended in the endgame. However, this method would not have saved the game here because of Black's complete passivity. After 44.Re3!
suggested by IM Nemeth, Black cannot avoid mate on the eighth rank (either on e8 or d8). Remarkably, Black would have maintained chances to save the game in case of the absence of his a-pawn, which would have allowed starting efficient counterplay based on ...Ra1.]
42.Kb5
Kd5
43.Rc2
being reduced to complete passivity, Black resigned. 1-0