[Event "Aimchess US Rapid KO 2021"] [Site "chess24.com INT"] [Date "2021.09.04"] [Round "3.11"] [White "Artemiev, Vladislav"] [Black "Carlsen, Magnus"] [Result "0-1"] [ECO "A15"] [WhiteElo "2704"] [BlackElo "2847"] [Annotator "Monokroussos,Dennis"] [PlyCount "70"] [EventDate "2021.08.31"] [EventType "k.o."] {[%evp 0,70,27,-5,-5,-5,14,16,22,22,26,12,12,7,18,-9,8,28,20,20,33,10,20,18, -12,-12,-25,-3,11,-11,-11,-9,-9,5,38,36,29,27,33,18,32,-12,-8,7,14,7,11,-6,68, 37,47,37,67,59,82,83,86,67,78,71,67,69,101,-62,-84,-150,-161,-161,-161,-562, -1046,-29989,-29990]} 1. c4 Nf6 2. Nf3 c5 3. Nc3 e6 4. e3 d5 5. d4 a6 6. cxd5 exd5 7. g3 (7. Be2 Nc6 8. O-O Bg4 9. b3 Rc8 10. dxc5 Bxc5 11. Bb2 O-O 12. h3 Bh5 13. Nh4 Bg6 14. Nxg6 hxg6 15. Bf3 d4 16. Na4 Ba7 17. Bxc6 Rxc6 18. Bxd4 Bxd4 19. exd4 Rd6 20. Qf3 b5 21. Nc5 Rxd4 22. Rad1 Rxd1 23. Rxd1 Qb6 24. Ne4 Nxe4 25. Qxe4 Rd8 26. Rc1 b4 27. g3 a5 28. Kg2 Qd6 29. Rc6 Qd5 30. Qxd5 Rxd5 31. Ra6 g5 32. Kf3 f6 33. Ke3 Kf7 34. Ra7+ Kg6 35. g4 Rc5 36. f3 f5 37. Ra6+ Kh7 38. gxf5 Rxf5 39. Kf2 g6 40. Ra7+ Kh6 41. Kg3 Rc5 42. Ra8 Rf5 43. Ra7 Rc5 44. Ra8 Rf5 45. Ra7 {1/2-1/2 (45) Giri,A (2764)-Carlsen,M (2863) chess24.com INT 2020}) 7... c4 8. Bg2 Bb4 9. Bd2 O-O 10. Ne5 Nc6 11. O-O Re8 12. Nxc6 bxc6 13. b3 {Carlsen had previously found himself on the white side of this position - twice - with only half a point to show for his efforts. The engine isn't overly enamored by White's position either, so it's a funny choice for Artemiev.} a5 (13... Bxc3 14. Bxc3 Ne4 15. Rc1 a5 16. bxc4 Ba6 17. f3 Nxc3 18. Rxc3 Bxc4 19. Rf2 Qg5 20. Qc1 h5 21. Bf1 Bxf1 22. Kxf1 Rab8 23. Rxc6 h4 24. Kg2 Rxe3 25. Rc8+ Rxc8 26. Qxc8+ Kh7 27. Qd7 f6 28. Qb5 Qf5 29. g4 h3+ 30. Kg3 Qe4 31. Qxa5 f5 32. gxf5 Qxf5 33. Qb5 Re6 34. Re2 Rg6+ 35. Kf2 Qf4 36. Qb1 Kh6 37. Qd3 Qxh2+ 38. Ke3 Re6+ {0-1 (38) Carlsen,M (2862)-Dubov,D (2702) chess24.com INT 2020}) 14. bxc4 $146 (14. Qc2 cxb3 15. axb3 g6 16. Rfc1 Bf5 17. Qd1 Qd6 18. Ra2 h5 19. Na4 h4 20. Nc5 Kg7 21. Bxb4 axb4 22. Rxa8 Rxa8 23. Ra1 Rh8 24. Qe1 hxg3 25. hxg3 Ne4 26. Nxe4 Bxe4 27. Bxe4 dxe4 28. Ra5 Rh5 29. Rxh5 gxh5 30. Kg2 c5 31. dxc5 Qxc5 32. Qd1 Qc3 33. Qd5 Kf6 34. Qd6+ Kg7 35. Qd5 Kf6 36. Qd6+ Kg7 37. Qd5 Kf6 {1/2-1/2 (37) Carlsen,M (2845)-Mamedyarov,S (2790) Shamkir 2019}) 14... Ba6 15. Qc2 Bxc4 16. Rfb1 Bxc3 17. Bxc3 Ne4 18. Be1 h5 19. Qd1 h4 20. g4 a4 21. a3 Bb3 22. Qe2 Re6 (22... c5 $142 $17) 23. Rc1 Qg5 24. f4 $1 Qe7 25. Bb4 $11 Qc7 26. f5 (26. g5 $142) 26... Ree8 27. Qe1 Qd8 28. Rxc6 h3 $5 {A nice idea in Artemiev's time trouble. White's position is rather hole-y, and won't be easy to defend.} 29. Bxh3 Qg5 (29... Bc4) 30. Qe2 Rac8 (30... Nf6) 31. Rac1 $4 {Natural, but simply a blunder.} (31. Rxc8) (31. Rcc1) 31... Bc4 $1 32. Rxc8 ({The exchange sac doesn't help, because White can't keep taking on c4 due to the loose e3-pawn.} 32. R6xc4 Rxc4 33. Rxc4 dxc4 {is hopeless for White.} 34. Qxc4 Qxe3+ 35. Kg2 Nf2 36. Qc3 Nd3 $19 {will soon net White's queen (or king).} ) 32... Bxe2 33. Rxe8+ Kh7 {Materially, White is fine, but the threats of ... Bxg4 and ...Qxe3+ are crushers.} 34. Bd2 Bxg4 35. Bg2 Bf3 0-1 [Event "Aimchess US Rapid KO 2021"] [Site "chess24.com INT"] [Date "2021.09.04"] [Round "3.12"] [White "Carlsen, Magnus"] [Black "Artemiev, Vladislav"] [Result "1-0"] [ECO "D32"] [WhiteElo "2847"] [BlackElo "2704"] [Annotator "Monokroussos,Dennis"] [PlyCount "57"] [EventDate "2021.08.31"] [EventType "k.o."] {[%evp 0,57,27,27,29,-5,-5,-5,25,20,29,11,35,6,13,3,18,-51,-51,-7,18,-14,0,6, 33,40,47,81,94,-122,-60,-50,-39,-77,-77,-102,-109,-106,-110,-147,-118,-117, -144,-125,-87,-95,-95,-95,-124,-131,-41,-31,0,0,14,30,24,28,330,338]} 1. d4 d5 2. c4 e6 3. Nc3 c5 4. cxd5 exd5 5. dxc5 $5 d4 6. Na4 (6. Ne4) 6... b5 $1 7. cxb6 axb6 8. b3 b5 $1 9. Nb2 Bb4+ $146 10. Bd2 Bxd2+ (10... Bc3 $1) 11. Qxd2 Nf6 12. e4 $1 Nxe4 (12... O-O {is a reasonable alternative, which will be met by} 13. f3 ({or} 13. Bd3 {.})) 13. Bxb5+ Bd7 14. Qb4 $5 (14. Bxd7+ Qxd7 15. Qc2 O-O 16. Ne2 (16. Qxe4 $4 Re8 $19) 16... Re8 $44) 14... Nc3 $1 15. Bxd7+ Qxd7 16. Kf1 Nc6 17. Re1+ Kd8 18. Qb6+ Qc7 19. Qc5 $5 {Carlsen keeps taking risks, both because of Artemiev's customary lag on the clock - and in this game it seemed he may have had connection problems as well - and because Artemiev typically seems more comfortable in queenless positions.} Rxa2 $15 20. Nd3 Re8 21. Nf3 Ra5 22. Qc4 Rxe1+ 23. Ndxe1 Qd7 24. h4 Qd5 25. Rh3 Qxc4+ 26. bxc4 $11 { Black would be better, were it not for the coming Nxd4. Oddly, Black has no sensible answer to this small tactic, which assures White of equality.} Ra4 $6 (26... Ra1 27. Nxd4 Nxd4 28. Rxc3 Kc7 29. f3 Kc6 30. Kf2 Ra2+ 31. Ke3 (31. Kf1 Nf5 $11) 31... Nf5+ $11) 27. Nxd4 Rxc4 28. Nf5 {Threatening Nxg7, which can't be stopped. Nevertheless, the position remains drawish against correct play by Black.} g6 $4 (28... Nd5 29. Nxg7 Ne5 $11 {/+/=}) 29. Nd6 {The problem isn't the f-pawn; it's the black knight on c3. Well, really it's all of Black's pieces, which are, as a unit, on collectively the worst squares possible. The king on d8 runs into Nb7+ in case of ...Rc8, and Black's knights are both hanging once the rook leaves the c-file, while they also restrict Black's rook so that it can't safely remain on the c-file to keep either knight under protection.} (29. Nd6 Rd4 (29... Rc5 30. Nb7+ $18) 30. Nxf7+ Ke7 31. Rxc3 $18) 1-0 [Event "Aimchess US Rapid KO 2021"] [Site "chess24.com INT"] [Date "2021.09.04"] [Round "3.13"] [White "Artemiev, Vladislav"] [Black "Carlsen, Magnus"] [Result "1-0"] [ECO "A04"] [WhiteElo "2704"] [BlackElo "2847"] [Annotator "Monokroussos,Dennis"] [PlyCount "76"] [EventDate "2021.08.31"] [EventType "k.o."] {[%evp 0,76,27,27,58,16,22,-9,18,0,24,14,14,24,50,4,-4,4,6,11,-2,7,-5,-19,26, -18,-20,-27,-25,-24,-16,-16,9,-6,6,6,10,6,12,1,13,20,31,42,20,36,28,35,33,26, 28,22,113,103,130,107,107,107,121,122,108,113,112,118,174,111,223,254,217,204, 193,193,189,189,192,195,198,209,220]} 1. Nf3 c5 2. c4 Nf6 3. g3 Nc6 4. Bg2 d5 5. cxd5 Nxd5 6. Nc3 g6 7. h4 h6 8. O-O Bg7 9. d3 Be6 10. Bd2 b6 11. a3 Rc8 12. Rc1 O-O 13. Nxd5 Bxd5 14. b4 Nd4 15. Nxd4 Bxg2 16. Ne6 Qd7 17. Kxg2 Qxe6 18. bxc5 Rxc5 19. Rxc5 bxc5 20. Be3 c4 21. dxc4 Qxc4 22. Qd2 a6 23. Rc1 Qe4+ 24. Kg1 Rb8 25. Qa5 Qe5 $2 {Not a mouse slip, exactly, but not the intended move, either. What happened was that Carlsen initially intended to play 25...Qe6, but after moving the piece with his mouse changed his mind, and put it back on e5, thinking that was where it came from. Only after some time had elapsed did he realize that the queen came from e4, not e5 - I imagine he noticed that White's queen was hanging, was confused for an instant, and then realized what had happened. Pro tip: I've made that sort of error in online blitz before as well, so what I do now is move the queen to a square it can't possibly go to (e.g. h8) and then let go of it. (The software will put the queen back on the correct square.) After this error, White collected the pawn and won without difficulty - and Carlsen, probably irritated by his error, didn't put up much resistance, either.} 26. Qxa6 Kh7 27. Bf4 Qb5 28. Qxb5 Rxb5 29. Rc7 g5 30. hxg5 hxg5 31. Be3 Kg6 (31... Bf6) 32. a4 (32. Rxe7 {seems fine.}) 32... Rb2 $2 { This lets White's a-pawn go straight to a7, with a quick and easy win.} (32... Rb1+ 33. Kg2 Ra1) 33. a5 Ra2 34. Rc6+ Kf5 35. a6 Ke4 36. a7 Bd4 37. Bxd4 Kxd4 38. Rc7 g4 {After White picks up the second pawn, the win will be trivial.} 1-0 [Event "Aimchess US Rapid KO 2021"] [Site "chess24.com INT"] [Date "2021.09.04"] [Round "3.14"] [White "Carlsen, Magnus"] [Black "Artemiev, Vladislav"] [Result "1/2-1/2"] [ECO "B51"] [WhiteElo "2847"] [BlackElo "2704"] [Annotator "Monokroussos,Dennis"] [PlyCount "102"] [EventDate "2021.08.31"] [EventType "k.o."] {[%evp 0,102,27,22,38,45,81,51,66,63,67,67,90,61,54,48,52,11,12,9,18,22,26,27, 27,-1,36,18,17,17,23,24,17,27,47,31,33,15,14,-13,4,-2,-5,0,-5,3,7,-7,17,32,32, -5,-5,-3,22,18,13,3,12,-11,1,-19,-45,-59,-60,-48,-41,-41,-50,-53,-53,-51,-54, -51,-45,-50,-53,-51,-40,-40,-8,-3,-3,-2,-2,-2,-2,-2,-1,-1,-1,-6,-6,-10,-6,0,0, 0,0,0,0,0,0,0,0]} 1. e4 c5 2. Nf3 d6 3. Bb5+ Nd7 4. a4 Ngf6 5. Nc3 g6 6. a5 Bg7 7. O-O O-O 8. a6 bxa6 9. Bxa6 Bxa6 10. Rxa6 Nb8 11. Ra3 Nc6 12. d3 Qb6 13. b3 Nd7 14. Bb2 Qb7 15. Qa1 a5 16. Nd5 Bxb2 17. Qxb2 a4 18. Qc1 e6 19. Ne3 axb3 20. Rxb3 Qc7 21. Qb2 Rfb8 22. Rxb8+ Rxb8 23. Qc3 Nde5 24. Nxe5 Nxe5 25. Ra1 Nc6 26. Qf6 Qd8 27. Qxd8+ Rxd8 28. Ra6 Nb4 29. Rb6 d5 30. exd5 exd5 31. d4 cxd4 32. Rxb4 dxe3 33. fxe3 Rc8 34. Rb5 Rxc2 35. Rxd5 Kg7 36. Kf1 Kf6 37. Rd7 h5 38. h4 Ke6 39. Ra7 f6 40. Rg7 Kf5 41. e4+ Kxe4 42. Rxg6 Kf5 43. Rh6 Kg4 44. Rxf6 Kxh4 45. Rf4+ Kg5 46. Rf3 Rc4 47. Kf2 Rf4 48. Rxf4 Kxf4 49. g3+ Kg4 50. Kg2 h4 51. gxh4 Kxh4 {A very well-played game by both sides; unfortunately for Artemiev, it meant that Carlsen won the first day's mini-match.} 1/2-1/2 [Event "Aimchess US Rapid KO 2021"] [Site "chess24.com INT"] [Date "2021.09.05"] [Round "3.21"] [White "Carlsen, Magnus"] [Black "Artemiev, Vladislav"] [Result "1-0"] [ECO "D37"] [WhiteElo "2847"] [BlackElo "2704"] [Annotator "Monokroussos,Dennis"] [PlyCount "99"] [EventDate "2021.08.31"] [EventType "k.o."] {[%evp 0,99,27,20,29,-5,-5,-5,-4,14,46,28,42,5,7,-13,13,38,68,59,57,35,36,16, 12,3,3,3,-1,2,-2,-6,-6,1,27,24,24,1,-7,18,5,-22,21,-6,29,7,15,-11,15,-3,16,-2, 6,18,-7,-15,0,2,0,0,0,0,39,40,50,44,59,57,58,52,49,93,93,98,129,119,104,104, 107,107,147,153,389,307,472,378,402,444,909,909,956,486,486,486,530,528,590, 597,1299,556,562,565]} 1. d4 d5 2. c4 e6 3. Nf3 Nf6 4. Nc3 Nbd7 5. cxd5 exd5 6. Qc2 c6 7. Bg5 Be7 8. e3 Nh5 9. Bxe7 Qxe7 10. O-O-O Nb6 11. Bd3 Bg4 12. Kb1 g6 13. h3 $146 (13. Rc1 {has been almost automatic here.}) 13... Bxf3 14. gxf3 O-O-O 15. Qb3 Kb8 ({The engine suggests the inhuman} 15... Kc7 {.}) 16. a4 $14 Qe6 17. a5 Nc4 18. a6 (18. Qb4) 18... b6 (18... b5 $5) 19. Na4 Rc8 20. Ka1 { Preparing Rc1, so there's no ...Nd2+ fork.} Ka8 21. Rc1 Rc7 22. Nc3 Rhc8 23. h4 Nf6 24. Rc2 Nd6 (24... b5) 25. Rhc1 (25. e4) 25... c5 $1 26. Nb5 Nxb5 27. Bxb5 Qd6 (27... c4) 28. Qa3 $1 $14 Kb8 29. Kb1 Qe7 30. Rc3 Ng8 31. dxc5 Rxc5 32. Rxc5 Rxc5 33. Rxc5 Qxc5 34. Qa4 {Because of the pawn on a6, Black's king is more vulnerable than its counterpart. Right now White intends Qf4+, followed by further harassment of the Black king.} Qd6 35. Qd4 f6 36. h5 $1 Ne7 37. hxg6 (37. h6 {was also possible, AlphaZeroing on both flanks.}) 37... hxg6 38. Qh4 Kc7 39. Qh8 Qb4 $6 (39... d4 $1 40. Qa8 Nc8 {keeps holding on.}) 40. Qe8 Kd6 $2 ({Black needed to retract the previous move, when the situation isn't yet fatal.} 40... Qd6) 41. Qd7+ $18 Ke5 (41... Kc5 42. Ba4 $1 Kc4 43. Qc7+ $1 Qc5 ( 43... Kd3 44. Qc2#) 44. Qxa7 $18) 42. Bd3 $1 {Threatening 43.f4+ Qxf4 44.Qxe7#. } f5 43. Qxa7 {There's no perpetual check, so White is clearly winning.} Qe1+ 44. Kc2 Qb4 (44... Qxf2+ 45. Kb3) 45. Qc7+ Kf6 46. a7 Qa4+ 47. Kd2 Nc6 48. Qxb6 Kg5 49. Bc2 Qxa7 50. Qxc6 1-0 [Event "Aimchess US Rapid KO 2021"] [Site "chess24.com INT"] [Date "2021.09.05"] [Round "3.22"] [White "Artemiev, Vladislav"] [Black "Carlsen, Magnus"] [Result "1/2-1/2"] [ECO "B20"] [WhiteElo "2704"] [BlackElo "2847"] [Annotator "Monokroussos,Dennis"] [PlyCount "98"] [EventDate "2021.08.31"] [EventType "k.o."] {[%evp 0,98,20,22,38,-8,-6,-21,0,9,13,31,71,71,55,49,51,63,57,49,47,45,35,33, 40,31,22,26,-24,-23,-24,-25,0,-25,0,0,19,7,21,0,7,0,20,26,69,53,61,68,58,61,59, 62,62,60,55,58,70,65,44,34,34,34,34,34,45,45,43,58,58,50,50,50,36,26,26,-9,7, -24,0,0,-13,-15,-6,-6,-34,-52,-17,-22,-18,-44,-33,-36,-3,-23,-3,-6,-3,-12,-3, -3,-2]} 1. e4 {Not exactly Artemiev's favorite move, but desperate times require desperate measures.} c5 2. g3 $5 {Not considered particularly dangerous, but the point is not to win with some super-prep but to get a game.} d5 3. exd5 Qxd5 4. Nf3 Bg4 5. Bg2 Qe6+ 6. Kf1 {All normal - if one can call playing 6.Kf1 "normal" - but here Carlsen sank into thought. He would of course already know about this line, but spent the time figuring out what plan he would go for.} Nc6 7. h3 Bxf3 (7... Bh5 {is more than 30 times more common (and I suspect Carlsen knew this as well), but may have felt that it was better - safer - to avoid the main line of the variation, where Artemiev may have discovered something particularly new and interesting.}) 8. Bxf3 Qd7 9. d3 Nf6 10. Nd2 $146 {A new move, but there were hardly any prior games, and none at the super-GM level.} e6 11. Nc4 Be7 12. a4 O-O 13. c3 {Preparing Kg2, but weakening the d3-pawn.} (13. Kg2 {is what White would like to do, but it allows } Nd4 {, swapping off White's powerful bishop.}) (13. a5) 13... Rad8 ({After the game, Carlsen expressed his dismay at using the a8-rook to go here. He had a concrete idea in mind, but he had completely missed Artemiev's idea. Against that, it would have been better to have chosen this rook.} 13... Rfd8 14. Kg2 Nd5 (14... Qxd3 $6 15. Qxd3 Rxd3 16. a5 {isn't too different from the game; White is better.}) 15. a5 b5 {is a line where it makes sense to have the second rook on a8 rather than f8.}) 14. a5 $1 Qxd3+ $6 (14... Nd5) 15. Qxd3 Rxd3 16. Ke2 Rd7 17. a6 Nd5 18. axb7 Rxb7 {White has a very nice position. In return for the sacrificed pawn, he has the bishop pair and targets on a7 and c5.} 19. Ra6 Nb8 20. Ra2 Nd7 21. Bd2 Rc7 22. Rha1 N7b6 23. Ne3 Nxe3 24. Bxe3 Nc8 25. Ra6 g5 26. Rc6 Rxc6 27. Bxc6 Kg7 28. Ra5 h6 29. Bxc5 Bxc5 30. Rxc5 Nd6 31. Bf3 Kf6 32. b4 Rd8 33. g4 Rd7 34. c4 Ke7 35. Rc6 f6 36. Bg2 Nf7 37. Ra6 Ne5 {I don't think White was ever winning, but he has been better, to varying degrees, ever since the pawn sac. Carlsen confessed after the match that he expected to lose this game, but his resilience and Artemiev's unsuccessful time management let him escape. This position is already drawn, as 38.c5 Nd3 gets rid of one of White's powerful queenside pawns. So Artemiev tried to push Black's rook off the d-file with} 38. Bc6 {, but after} Rc7 {he is nevertheless losing one of his queenside pawns all the same. The next moves are forced, and the result is a dead drawn rook ending.} 39. Bb5 Nxc4 40. Bxc4 Rxc4 41. Rxa7+ Kd6 42. Rh7 Rxb4 43. Rxh6 Ke5 44. Kf3 Rb3+ 45. Kg2 Rb2 46. Rg6 Ra2 47. Rh6 Rb2 48. Rg6 Ra2 49. Rh6 Rb2 1/2-1/2 [Event "Aimchess US Rapid KO 2021"] [Site "chess24.com INT"] [Date "2021.09.05"] [Round "3.23"] [White "Carlsen, Magnus"] [Black "Artemiev, Vladislav"] [Result "1-0"] [ECO "B92"] [WhiteElo "2847"] [BlackElo "2704"] [Annotator "Monokroussos,Dennis"] [PlyCount "97"] [EventDate "2021.08.31"] [EventType "k.o."] {[%evp 0,97,22,22,38,38,81,60,55,24,24,24,44,26,52,54,48,41,41,44,65,60,48,45, 44,50,103,98,112,106,106,125,147,160,175,158,171,145,145,160,145,116,182,135, 128,128,186,137,186,185,195,214,247,252,254,197,201,207,207,186,306,207,230, 129,134,180,180,212,258,258,264,332,352,365,343,372,377,370,402,402,402,427, 409,383,415,639,1186,478,520,546,515,546,524,929,992,580,992,634,1785,2150]} 1. e4 c5 2. Nf3 d6 3. d4 cxd4 4. Nxd4 Nf6 5. Nc3 a6 {Artemiev isn't a Najdorf player, but as he had to win this game (and then the next one) to stay alive, sharper is better.} 6. Be2 {Quiet chess?} g6 (6... e5) (6... e6) 7. g4 { Apparently not!} h6 (7... b5) (7... Bg7) 8. Be3 Bg7 9. f3 b5 10. a3 (10. Qd2 Bb7 11. O-O-O b4 12. Nd5 {is playable - 10.a3 isn't bad, but it isn't necessary either.}) 10... Bb7 11. Qd2 Nc6 {Not a very "Najdorf-y" move.} (11... Nbd7 {is more (stereo-) typical.}) (11... h5 $1 {may be best, resolving the kingside and making it at least possible for Black to castle short in a line like} 12. g5 Nfd7 13. O-O-O O-O {, with good chances to mix things up.}) 12. O-O-O b4 $5 13. Nxc6 Bxc6 (13... bxc3 14. Qxc3 Qc7 15. Na5 Qxc3 16. bxc3 $16 { is much better for White--don't be fooled by the seeming weakness of the doubled, isolated c-pawns. They are not falling any time soon, and Black's a6-pawn is even weaker.}) 14. axb4 Qb8 15. Nd5 Bxd5 (15... Nxd5 $142 16. exd5 Bb5 17. c4 Ba4 18. Rde1 O-O 19. Bd3 Rc8 {may be Black's best try. White is objectively fine, but the position is far from "accident-proof" for him.} 20. Bxh6 Bxh6 21. Qxh6 Qxb4 22. Qd2 $14) 16. exd5 a5 17. b5 a4 18. Bd4 $1 { Artemiev said he had missed this idea. Black's attack is already over.} O-O ( 18... a3 19. b3 a2 20. Kb2 $18) 19. Kb1 a3 20. b3 Qc8 21. c4 a2+ 22. Ka1 Re8 23. Rhe1 (23. Qb2) 23... e5 24. dxe6 Rxe6 25. Bd3 d5 26. Qb2 (26. cxd5) 26... dxc4 27. Bxc4 Rxe1 28. Rxe1 Qd8 29. h4 g5 30. hxg5 hxg5 31. f4 gxf4 32. g5 Qxd4 $1 {A very nice try.} (32... Ne8 33. Bxg7 Nxg7 34. g6 {is even worse - now it's Black who's going to get mated.}) 33. Qxd4 Ng4 34. Bxf7+ $1 {Otherwise Black would reach an equal ending.} Kf8 (34... Kxf7 $4 35. Re7+) 35. Re8+ $1 ( 35. Qxg7+ Kxg7 36. Re7 {is stronger for a computer, but at a short time control game - especially one where a draw is as good as a win - such an adventurous approach is foolhardy.}) 35... Rxe8 36. Qxg7+ Kxg7 37. Bxe8 f3 38. b6 $1 {Only this move wins, though other moves draw - and that, again, is sufficient to win the match.} Ne3 39. b7 Nc2+ 40. Kxa2 Nb4+ 41. Ka3 Na6 42. Bb5 Nb8 43. Bd3 Nd7 44. Kb4 f2 45. Kb5 Kf7 46. b4 Ke6 47. Ka6 Kd6 48. Ka7 Kc7 49. g6 1-0