Every Tempo Matters

In position with opposite castlings the main question is "Who is faster?" Every tempo counts and the speed of the attack is richly awarded at the end. This is what the Polish GM Dariusz Swiercz managed to prove in round five of the Qatar Chess Masters. In a sharp Gruenfeld the former junior world champion managed to grind down the fearsome Anton Korobov from Ukraine:

[Event "Qatar Masters Open"]
[Site "?"]
[Date "2015.12.24"]
[Round "5.8"]
[White "Korobov, Anton"]
[Black "Swiercz, Dariusz"]
[Result "0-1"]
[WhiteElo "2713"]
[BlackElo "2646"]
[Annotator "Bojkov, Dejan"]
[PlyCount "70"]
[EventDate "2015.??.??"]
[WhiteClock "0:07:01"]
[BlackClock "0:02:29"]

1. d4 Nf6 2. c4 g6 3. f3 {Diagram [#] A popular move order.} d5 4. cxd5 Nxd5 5.
e4 Nb6 6. Nc3 Bg7 7. Be3 O-O 8. Qd2 Nc6 9. O-O-O Qd6 10. Nb5 {Mamedyarov
introduced this move in the modern practice back in the distant 2004...} ({
White can postpone the jump and go first} 10. Kb1 Rd8 {and only then} 11. Nb5
Qd7 12. d5 a6 13. Nc3 Na5 14. Bd4 {with unclear play in the blitz game Aronian,
L (2784)-Volokitin,A (2639) Berlin 2015}) 10... Qd7 ({Noone had ever tested
the computer move} 10... Qd8) 11. f4 Qe6 {[%csl Ra2][%cal Ge6a2] Diagram [#]
The point behind Black's previous move. He needs to counter attack as quick as
possible.} 12. Nc3 ({With his kingside undeveloped White cannot afford the
McDuck's approach} 12. Nxc7 $2 Qxa2 13. Nxa8 Na5 $1 {[%csl Rb2,Yc1][%cal Ra5b3,
Ra2a1,Rf8c8,Rc8c1,Rb6c4,Rg7b2] Diagram [#] leads to lethal attack for Black.})
12... Nc4 13. Qe2 N6a5 (13... Nxe3 14. Qxe3 {is aslo possible and was tested
in the game Wang Hao-Svidler, Stavanger 2013.}) 14. Bf2 $146 {[%csl Gf2]
Diagram [#] Korobov decided to keep this bishop. Since Black did not take it
on the previous move some players believed that he may never take it.} ({
Afresh example is} 14. Nf3 c5 15. e5 cxd4 16. Nxd4 Qb6 17. Nf5 Nxe3 18. Nxe3
Nc6 19. Ned5 Qd8 20. g3 Be6 21. Bg2 {with advantage for White, Karjakin,S
(2780)-Giri,A (2720) Wijk aan Zee 2013}) 14... c5 {A sharp position occurred.
Both sides castled in different flanks and in these situations the main
question is who is faster. Swiercz tries to open files as early as possible
and he has every right to do so. Even the rough, mechanical calculation
reveals that he is already ahead in the development. The other logical
possibilities are} (14... b5 $5) ({And} 14... Rd8 $5) 15. Nf3 ({Bad is} 15.
dxc5 $2 Bxc3 16. bxc3 Qf6 {[%csl Rc3,Rf4][%cal Rf6c3,Rf6f4] Diagram [#] with
double attack against c3 and f4.}) (15. d5 {fails to impress neither after} Qb6
) ({However, the immediate} 15. e5 {is interesting with the idea to meet} b5 {
with} 16. d5 Qf5 17. g3 {The threat is Bf1-h3 when White wants to make use of
his knight on g1. But Black may pretend that he did not see it and go for} b4 {
when} 18. Bh3 $6 {Bumps into} bxc3 $1 19. Bxf5 $2 cxb2+ 20. Kc2 Bxf5+ 21. Kc3
Rab8 {[%csl Rc3] Diagram [#] and Black wins. Such sharp lines are typical for
positions with open kings.}) 15... b5 16. e5 {Building some barricades seems
more useful at the moment} (16. Kb1 $5 Qa6 17. Qc2) 16... Qa6 ({White would be
happy to close the position after} 16... b4 17. Ne4 b3 18. a3) 17. Ne4 $2 {
[%csl Ya2,Rc1] Diagram [#] It is not that obvious, but White badly needed this
knight in the defense of his king. Correct was} (17. dxc5 Bf5 (17... Bh6 18. g3
Bg4 {is another possibility.}) 18. Nd4 {with unclear play.}) 17... Bf5 $1 {
A cold shower. Instead of defending, Swiercz brings another piece in the
attack.} 18. g4 ({Probably here Korobov realized that on the planned} 18. Nxc5
{Black has the killer} Nb3+ $3 {[%csl Rc1] Diagram [#]} 19. Nxb3 (19. axb3 Qa1#
) 19... Qxa2 20. Rd3 Rac8 {and White has to resign.}) ({White is clearly worse
after both} 18. Kb1 Nc6 $17) ({Or} 18. g3 Nc6 $17) 18... Bxg4 19. Nxc5 {
Now that the bishop is away from the b1-h7 diagonal it seems as White can do
this. But...} Nb3+ $1 {[%csl Ra1,Ra2,Rc1][%cal Ra6a1] Diagram [#] Comes anyway.
} 20. Kb1 {There is no choice} (20. axb3 Qa1+ 21. Kc2 Qxb2+ 22. Kd3 Bf5+ 23.
Ne4 Qxb3#) ({Or} 20. Nxb3 Qxa2 {we do not count.}) 20... Nxc5 21. dxc5 {
The smoke had cleared. Black kept all his active pieces and White has no
defenders of the queenside, nor development. With natural play Swiercz
converts his indisputable advantage.} f6 {[%cal Rg7a1] Diagram [#]} (21... Rad8
{was quite convincing too.}) 22. Rg1 Bh5 23. Qe4 fxe5 24. Bxc4+ bxc4 25. Rd7 ({
Computer claims White's best chance is the endgame after} 25. Nxe5 Bxd1 26.
Rxd1 Rad8 27. Rxd8 Rxd8 28. Qxc4+ Qxc4 29. Nxc4 Kf7 {but Black is obviously
winning here as well.}) 25... Bxf3 26. Qxf3 Qe6 {[%csl Rb2][%cal Ga8b8,Gf8c8,
Rg7b2] Diagram [#]} 27. Rgd1 ({Or} 27. Qg4 {when Black has a pleasant choice
of winning continuations} Rf5 $19 ({And} 27... Qxg4 28. Rxg4 exf4 29. Rxe7 Rab8
$19)) 27... e4 $1 28. Qe2 Rfb8 29. R1d4 c3 $1 {Diagram [#]} ({Even better than
} 29... Bxd4 30. Rxd4 c3 31. b3 Rd8 {which also wins.}) 30. b4 ({In case of}
30. b3 {Swiercz can enjoy another sac on the b3 square} Rxb3+ $1 {[%csl Rb1]
Diagram [#]} 31. axb3 Qxb3+ 32. Kc1 (32. Ka1 c2 $1 $19) 32... Qa3+ 33. Kc2 Qb2+
34. Kd1 Qb1#) 30... Bxd4 31. Rxd4 a5 32. b5 c2+ $1 {[%csl Yb1][%cal Rb8b5,
Re6a2] Diagram [#] It is over.} 33. Qxc2 (33. Kxc2 Qxa2+ 34. Kd1 Qb1+ 35. Kd2
Rxb5 $19) 33... Rxb5+ 34. Kc1 e3 35. Re4 exf2 0-1

Report.