Sergey Ivanov,
grandmaster

The double-edged Noteboom system becomes very popular nowadays. Just one example: 7...Qe7 8. axb5 Bxc3 9. Bxc3 cxb5 10. d5 Nf6 11. dxe6 ( 11. d6 Qb7 12. b3 ) 11...Bxe6 12. Nd4 Bd7

13. Qf3 Ne4 14. g4 ( 14. Nf5!? ) 14... O-O 15. Bg2 Re8 16. O-O a6 17. Rfd1 Nxc3 18. bxc3 Ra7 , Alander - V.Bagirov, Berlin, 1998.

Another popular line - 11. d5 Nf6 12. bxc4 b4 13. Bxf6 Qxf6 14. Qa4+ Nd7 15. Nd4

15...e5 ( 15...Ke7 16. d6+ Kxd6 17. Rd1 Kc7!? ) 16. Nb3 Ke7 17. Be2 ( 17. Qb5 Ba6 18. Qxa5 Rhb8 19. d6+ Ke8! ) 17...Rhc8 - hasn't still gained a certain estimation.

This variation is of the most frequently use. There are 31 games in Chess Assistant 4.0 on it, played mainly in 1997-98. A detailed survey of this variant is given at the "Chess St.Petersburg", N 2(8)/1998. 15...Qc7 is considered to be less accurate because of 16. e4 e5 17. c5! exd4 18. Rc1! , and the black queen's position on c7 is not good. It lets White to take initiative.

This is a new plan. Not without reasons, White believes that his central pawns, if advance, will be more dangerous than the queensude's black passed pawns. Immediate 16. e4 doesn't seem to be well prepared: 16...e5 17. c5 exd4 18. Rc1 Bxe4 19. c6 Nf8!? , and White has to prove that his activity costs a pawn. Usually White tries 16. c5 . One of the latest examples: 16...Bc6 17. Bc2 Ne4 18. Ba4 Qc7 19. Nd2 Bxa4 20. Rxa4

20...Ndxc5! 21. dxc5 Rad8 22. Bd4 Nxd2 23. Qxd2 e5 , J.Piket - V.Ivanchuk, Monako, 1999.

This is the most reliable reply to the White's idea. Black takes the opportunity to advance his pawns further to bind the opponent's pieces. The possibility 16...e5 is standard too. Possible is 17. Nxe5 (interesting is also 17. dxe5 or 17. Bf5 ) 17...Nxe5 18. dxe5 Nd7 19. e6!? ( 19. c5 Nxe5! 20. Bxe5 Rxe5 21. c6? Bxc6! 22. Rxc6 Rd5 ) 19...fxe6 .

Now erroneous is 20. Qh5?! Nf6 21. Bxf6 Qxd3 22. Qg5 ( 22. Bxg7 Kxg7 23. Rcd1 Qg6 24. Rd7+ Kg8 25. Qxg6+ hxg6 26. Rxb7 Reb8 ) 22...Qg6 23. Qxg6 hxg6 24. Be5 a4 , and White gained a hard ending, S.Ivanov - V.Malakhov, Cappelle la Grande, 1999. But 20. c5 Bc6 ( 20...Rc8 21. Bb5 Bc6 22. Qa4!? ) is stronger: 21. Rc4 e5 ( 21...Nf6 22. Rd4 Qc7 23. Rd6 ) 22. Qc2 Nf6 23. Rh4 e4 24. Bc4 Kh8 25. Rd1 Qe7 26. Rd6

In the case of 18...Bc6 White has a pleasant choice between 19. e5 Nh5 (unfortunately, the square e8 is not free) 20. Nd2 Nf4 21. Be4 Bxe4 22. Nxe4 with the subsequent Nd6 and 19. Bc2 b3 ( 19...e5? 20. d5 Bb7 21. Qd2 Qa5 22. Rb1 ) 20. Bxb3 Bxe4 21. Ne5

19...Rc8? is bad because of 20. dxe5 ( 20. Qa4!? ) 20...Ng4 21. Bb5 Bc6 22. Bxc6 Rxc6 23. Qd4! (but not 23. Nd4 Qh4! or 23. h3 Nxf2! 24. Kxf2 Nxc5 )

20. dxe5 Ng4 is quite good for Black.

Bad was 21. Bc2 Nxc5 22. Bxa4 Nxa4 23. Nxe5 ( 23. Qb3 Nc3 ) 23...Nb2 24. Qd4 Rxe5! 25. Qxe5 Nd3 After 20...Ba4 the white queen has only two free squares, and it is quite difficult to decide which of them is better. Now I understand that my choice was wrong. It turned out to be that I was under impression of my home analysis, where I paid the main attention to the move 21. Qd2 . Nevertheless, 21. Qe2 is stronger. Now after

  A) 21...Qa5 22. c6 Nc5 23. Bxe5 (worse is 23. Nxe5 Nb3 24. Rb1 Nxa1 25. Nc4 Qd8! 26. Rxa1 Nd5 ) the white queen doesn't find itsself under the fork and after 23...Nxd3 ( 23...b3 24. Bc3 ) 24. Qxd3 b3 25. Bxf6 gxf6 26. Nd4 White's chances are obviously better. But, as a matter of the fact, after 21. Qe2 the pawn b4 is not under attack, so Black has the option of

  B) 21...Ra5 22. c6 Nc5 , and in the case of 23. Bxe5 ( 23. Rb1!? 23...b3 24. Bc3 ) 23...Bxc6! 24. Bxf6 gxf6 25. Rxc5 Rxc5 26. dxc6 b3 the position is unclear.

Of course, bad is 21...b3? 22. c6 b2 ( 22...Nf8 23. Bxe5 b2 24. Rc3 ) 23. cxd7 Qxd7 24. Bxb2 axb2 25. Qxb2

White can't prevent black knight to occupy the square c5 : 22. Qe3 Ng4! 23. Qg5 Nxc5 24. Rxc5 Qxc5 25. Qxg4 f6 with a very dangerous threat of b4-b3-b2 .

Deserving attention was 23. Bxe5!? Nb3 24. Qg5 h6 25. Qg3 Nh5 26. Qg4 Nxc1 27. Rxc1 For example, 27...g6 (or 27...Rxe5 28. Nxe5 Nf6 29. Qf5 29...b3 30. c7 ) 28. c7 Rec8 29. d6 .

I had been thinking for 45 minutes here, but, nevertheless, didn't bring myself to 24. Qg5 . May be, computer analyses will allow to prove that White would have lost in this case too. But at the practical game this aggressive move was apparently the best solution. The game could develop like this: 24...h6 25. Qg3 ( 25. Qf5!? ) 25...Nxc1 (if 25...Nxa1 , then 26. Rxa1 Nh5 27. Nc4 or 26. Nc4 is possible) 26. Nc4 (it seems that strong is 26. Nd7 Nh5 27. Nf6+ Nxf6 28. Bxf6 g6 29. Rxc1 with attack, but a counterattack exists: 26...b3! 27. Nxf6+ Kh8 28. Rxc1 b2 29. Qe3 bxa1=Q 30. Rxa1 Qc3 ) 26...Nh5! ( 26...Qxd5 27. Bxf6 )

  A) 27. Nxa5 Nxg3 28. Rxc1 Nxe4

  B) 27. Qe3 Nxd3 ( 27...Qxd5!? ) 28. Nxa5 Nxe1 29. Qxe1 Rxa5 30. Qxb4 Rb5

  C) 27. Qd6!? Qd8!? ( 27...Nxd3 28. Nxa5 Nxe1 29. Qxb4 ) 28. Rxc1 ( 28. Qxd8 Rexd8 29. Rxc1 Bxc6 ) 28...Qg5 29. Ne3 Nf4 30. h4

30...Nh3!+ ( 30...Qxh4 31. g3 Nh3+ 32. Kg2 Qh5 33. Rh1 ) 31. Kh2 Qf4+ 32. Qxf4 Nxf4

There was another way for White: 24. Qe3 .

  A) 24...Nxc1 25. Rxc1 b3 26. Nc4 Qb4 ( 26...Nxd5? 27. Qg5! ) 27. Bc3 But Black has a simple

  B) 24...Nxa1 25. Nc4 Qd8 26. Rxa1 Nxd5

Black has to prevent the white bishop to consolidate on the square d4 . Dubious was 25...Qc7?! 26. d6! Qxd6 27. Nxf7 ( 27. Qxf7+ Kh8 28. Qb7 Bb3 ) 27...Qf4 , and White has at least a perpetual check.

26. d6 is refuted by 26...b3! 27. Qxa3 Bxc6 .

This exchange was necessary - otherwise White couldn't defence the pawn e4 .

Black was threatened with 29. Nxa3 and the unpleasant transfer of the knight c4 to the f5 through the e3 .

I refused 29. Ne3 because of 29...Qxc1 30. Qxc1 Bxd3 . On the face of it, black pawns can't be stopped. But it turns out that after 31. Ng4! the white queen and the knight cooperate perfectly and produce dangerous threats to the black king. For example:

  A) 31...b3? 32. Nf6+ Kh8 ( 32...Kf8 33. d6 Kg7 34. Qg5+ Kh8 35. Nh5 ) 33. Qh6 Bxe4 34. Nh5 Rg8 35. Qf6+

  B) 31...a2 32. Nf6+ Kh8 33. Qa1 Re7 34. d6 That's why black would have faced with the dilemma: to admit the line B) or to choose between 30...a2 and 30...b3 with unclear play.

After the brown study my opponent decides to continue the struggle, and he didn't make the mistake as I did. He had only a draw after 29...Bxc4 30. Qg5+ Kf8 31. Qh6+ Ke7 32. Qg5+ , and Black can't avoid a perpetual check.

White would have had more practical chances after 30. Qh6 30...b3 31. d6 or 31. f5 In this case Black had to continue with 30...Qf8 , protecting the king.

In the case of 31. Nb6 black has several ways to the win: 31...Bxd3 (possible is 31...b3 32. Nxa8 b2 33. Rd1 a2 ) 32. Nxa8 .

  A) 32...Ra7 33. c7 ( 33. d6 Rxa8 34. d7 Bb5 ) 33...Rxa8 34. c8=Q+ Rxc8 35. Rxc8+ Kg7 36. Ra8 Bxe4

  B) 32...b3 33. d6 b2 34. Rd1 Kf8! 35. d7 ( 35. dxe7+ Kxe7 36. c7 Ba6 ) 35...Rxd7 36. cxd7 Ke7

The rook must attack passed pawns. If 31...Rea7 , then White could gain good chances to a draw by sudden 32. Nxa3! ( 32. c7 a2 33. Ra1 b3 ; 32. Ne5 b3! 33. c7 a2 ) 32...Bxc6 ( 32...Bxd3 33. c7 bxa3 34. d7 Rxc7 35. Rxc7 Ba6 36. Ra7! Rb8 37. f3 ) 33. Nb1 .

After 32. Nxa3 Bxc6 the pawn d6 is undefendable. 32. Nb6 Bxc6 33. Nxa8 Rxd6 also didn't help.

Of course, impossible was 32...Bxc4?? 33. d7!

The decisive mistake. Losing was also 33. e5 a1=Q 34. c8=Q+ Rxc8 35. Rxa1 Bxc4 36. Bxc4 Rxe5 . White had to play 33. Nb6!? Rxd6 ( 33...a1=Q 34. c8=Q+ Rxc8 35. Rxa1 ) 34. Nxa8 Rc6!? ( 34...Bd7 35. Bb5 Bc8 36. Rf1 36...b3 37. Bc4 ) 35. c8=Q+ Rxc8 36. Rxc8+ Kg7 37. Bxb5 a1=Q+ 38. Bf1 , and even after taking the rook for the pawn b the Black's victory is not clear.

White couldn't longer escape: 34. e5 Bxc4 35. d7 Ree8 36. dxe8=Q+ Rxe8 37. Bxc4 b2 .

The easiest way.