[Event "Chess.com"] [Site "Chess.com"] [Date "2022.06.19"] [Round "3"] [White "Caruana, Fabiano"] [Black "Duda, Jan-Krzysztof"] [Result "1/2-1/2"] [ECO "B90"] [WhiteElo "2783"] [BlackElo "2750"] [Annotator "Rafael Leitao"] [PlyCount "102"] [EventDate "2022.??.??"] [TimeControl "5400+30"] 1. e4 c5 2. Nf3 d6 3. d4 cxd4 4. Nxd4 Nf6 5. Nc3 a6 {I'm always happy when the Najdorf is played in a high level tournament. Is there anyone who prefers to watch the Berlin $2} 6. f3 e5 7. Nb3 Be6 8. Be3 h5 {A modern way to meet the English Attack. I like this move and have played it many times. The idea is to avoid g4. Chess is a developing and dynamic game. One could say that you should never advance on the side where your opponent is attacking, but empty words do not win chess games. The engine says it’s fine, and this is more important. Of course, it’s not so easy to play without a shelter for the king, but if you are afraid of this, you better not play the Sicilian.} 9. Qd2 Nbd7 {The main decision White must make is where to castle. Caruana decides on the more aggressive queenside castling.} 10. O-O-O Be7 11. Kb1 b5 (11... Qc7 { I once played this move against GM Mekhitarian, but I'd rather change the subject for obvious reasons.}) 12. Nd5 Bxd5 13. exd5 Nb6 14. Bxb6 Qxb6 15. Na5 {This is a typical manoeuver in the English Attack, and Black has to be ready to meet Nc6, otherwise he gets a lost position. But here everything is fine.} Rc8 {This is an important theoretical position. I think Caruana was surprised by Duda's opening choice, and he took some time before going for an endgame that is known to be harmless for Black.} 16. Nc6 (16. c4 $5 {is a critical move.} bxc4 17. Nxc4 Qa7 18. Na5 {The knight heads back to his dreamed square.} Rc5 $1 {Precise play, attacking the d5-pawn.} 19. Nc6 Qd7 {There are many correspondence games in this complicated position. Needless to say, most of them ended in a draw.}) (16. Bd3 b4 $1 17. Nc4 Qc5 18. Ne3 a5 {is also ok for Black.} (18... g6 $5 {Even this sacrifice is interesting.} 19. Bxa6 Rc7 { And if Black manages to castle and activate the e7-bishop, White will be in trouble. Remember to always be on the attacking side in positions with opposite-colored bishops.})) 16... Nxd5 $1 {The point of Black's play.} 17. Nxe7 Nxe7 18. Qxd6 Qxd6 19. Rxd6 Nc6 {I found more than one hundred games in this position and nearly all of them ended in a draw.} 20. Rd2 Ke7 {Black is happy that the king is not castled and can be placed immediately in the center. } 21. Bd3 Rhd8 22. Rhd1 g6 23. c3 Na5 24. Re2 {A new move.} (24. Kc2 {was played before.}) 24... Kf6 25. h4 {Avoiding the advance of Black's kingside pawns.} Nb7 26. Red2 Nc5 27. Bf1 Rxd2 28. Rxd2 {Is a bishop better than a knight $2 Is it true that rook and bishop combine better than rook and knight, while queen and knight combine better than queen and bishop $2 There are no hard rules in chess. Always look for the dynamic factors of the position. Here Black's knight is pretty good, and his king is active. He has no problems at all.} Ke6 29. c4 {This is a difficult move to make and this shows that the bishop is easier to play with black. The problem is that after plays b4, although White has a passed pawn, Black has a monster on c5. If he can exchange rooks and start to advance the kingside pawns, it can become dangerous for black. Of course, Caruana knows all that and he will prevent it.} (29. Kc2 Na4 $5 {Maybe Caruana wanted to prevent the maneuver Na4-b6, and the knight can go to c4 or d5 at some point. Of course, it's still a draw, but it's easier for Black.}) 29... b4 30. Kc2 a5 31. b3 Nb7 {Preparing the first part of the plan: exchanging rooks. White should start some active play now.} 32. g3 Rd8 33. Rxd8 Nxd8 34. Kd3 Nb7 35. Ke3 Nc5 36. f4 $1 {That's it. White is not advised to sit and wait here.} a4 (36... f6 {A possible drawing line after this move is:} 37. Bg2 a4 38. Bd5+ Ke7 39. bxa4 Nxa4 40. fxe5 fxe5 41. Be4 Kf6 42. Bc2 Nc5 43. Kf3 g5 44. Ke3 g4 {with a mutual fortress on the board. }) (36... f5 {White can wait for e4 and prepare a fortress or try to activate his bishop.} 37. fxe5 $5 Kxe5 38. Bg2 a4 39. bxa4 Nxa4 40. Bd5 Nc3 41. Bg8 Nxa2 42. c5 Nc3 43. c6 Kd6 44. Bf7 Kxc6 45. Bxg6 b3 46. Bxf5 {with a draw. As usual in this endgame, White had to play some careful moves.}) (36... e4 37. Kd4 Kd6 38. f5 $5 (38. Be2 f5 {This looks a bit scary, but Black cannot breakthrough.} 39. Ke3 Ne6 40. Kd2 Kc5 41. Ke3 Nd4 42. Bd1 {and White has a fortress.}) 38... gxf5 39. Be2 Ne6+ 40. Ke3 Ke5 (40... Ng7 41. Kd4 Ne6+ {with a draw.}) 41. Bxh5 f4+ 42. gxf4+ Nxf4 43. Bxf7 Ng2+ 44. Ke2 Nxh4 45. c5 Nf5 46. Bc4 {The position is a draw, although Black is the one having fun.}) 37. bxa4 Nxa4 38. c5 $1 { The easier way to achieve a draw. Caruana activates his bishop.} exf4+ 39. gxf4 Nxc5 40. Kd4 Kd6 (40... b3 {is an immediate draw.} 41. Bc4+ Kf5 42. axb3 Nxb3+ 43. Bxb3 Kxf4 44. Bxf7 Kg4 45. Bxg6 Kxh4 46. Bxh5 Kxh5) 41. Kc4 Ne6 (41... f5 { is a winning try, but it's not enough since the a-pawn is dangerous.} 42. Kxb4 Ne6 43. a4 Nxf4 44. a5 Nd5+ 45. Kb5 {with enough activity for an easy draw.}) 42. f5 $1 gxf5 43. Be2 Nf4 44. Bf3 Ng6 45. Bxh5 Nxh4 46. Bxf7 f4 47. Bh5 f3 48. Bxf3 Nxf3 49. Kxb4 Nd4 50. a3 Nc2+ 51. Kb3 Nxa3 {I love all these final moves when everyone knows the game will finish in a draw. But once a teammate in an Olympiad decided to make an extra centralizing king move in a position of king and bishop against king. His opponent didn’t like it very much. I guess he was lacking a sense of humor that day.} 1/2-1/2 [Event "Chess.com"] [Site "Chess.com"] [Date "2022.06.19"] [Round "3"] [White "Ding, Liren"] [Black "Rapport, Richard"] [Result "1/2-1/2"] [ECO "D87"] [WhiteElo "2806"] [BlackElo "2764"] [Annotator "Rafael Leitao"] [PlyCount "80"] [EventDate "2022.??.??"] [TimeControl "5400+30"] 1. d4 Nf6 2. c4 g6 3. Nc3 d5 {It's strange to see Rapport playing mainstream openings $1} 4. cxd5 Nxd5 5. e4 Nxc3 6. bxc3 Bg7 7. Bc4 c5 8. Ne2 Nc6 9. Be3 O-O 10. O-O b6 (10... Bg4 {To be honest, I never fully understood why the old main line lost its popularity.} 11. f3 Na5) 11. Rc1 Bb7 12. Bb5 {A difficult move. I thought Ding Liren's idea was to take on c6 and play d5, but I was wrong.} (12. Qd2 {The most natural move, but Black is doing fine after} Rc8 13. Rfd1 e6) 12... Rc8 13. Qd2 (13. Bxc6 {This move was played in several correspondence games. As I teach to my students: \"He who says A, must say B\". According to this logic, 13.Bxc6 is indeed the best move. But I'm not sure chess is so simple.} Rxc6 (13... Bxc6 14. d5 Bd7 15. Qd2 {with a complicated game.}) 14. d5 Rd6 15. c4 e6 {and I refer the reader to the many correspondence games he can find in this position. They all finished the way you might expect.}) 13... cxd4 (13... e6 {is the main move.}) 14. cxd4 Qd6 15. Rfd1 Qb4 $6 {The first new move, and the beginning of Black's problems.} (15... Rfd8 $6 {was played in an online blitz. As I wrote before, blitz games have become part of the opening theory.} 16. d5 $1 (16. f3 $6 {was Sjugirov-Mischuk blitz game 2021.}) 16... Ne5 17. Nd4 {and here White achieves what he wants: central dominance and a clear positional advantage. One of the positional drawbacks of the Grunfeld Defense is that White gets an extra pawn in the center.}) (15... a6 {A very natural move, asking White's bishops intentions.} 16. Bxc6 Rxc6 17. Rxc6 Bxc6 18. d5 Bd7 19. f4 {The computer is not worried with White's advancing pawns, but I would be concerned in a game.}) (15... Na5 {is also very natural.}) 16. Qd3 e6 17. h4 $1 {A typical push in this Grunfeld pawn structure. White prepares an attack on the kingside and sometimes the pawn goes all the way to h6. Many readers will associate this advance to the games of AlphaZero, but being a human supporter I would like to suggest the game Portisch-Christiansen, played in London, 1982. In this game, the great Hungarian player made a similar advance, and this is one of the best games I have ever seen.} (17. a3 $6 Qe7 {and White cannot advance the pawn anymore.}) 17... Rfd8 $6 {A natural move, but not the best.} (17... Qa5 {is an interesting exchange sacrifice.} 18. a4 a6 19. Bd2 Nb4 20. Qb1 $1 axb5 21. Bxb4 Qxa4 22. Bxf8 Rxc1 23. Rxc1 Bxf8 {but it's clear White is better here.}) (17... h5 {is the computer suggestion, but I prefer White after} 18. e5 $1 {with idea to play Bg5.}) 18. Bg5 $1 {An unpleasant move to face.} Nxd4 {Rapport is not the one to shy from complications. He decides to sacrifice the exchange.} ( 18... f6 19. Bd2 Qe7 20. h5 {and the kingside is too weak.}) 19. Bxd8 Rxd8 20. h5 $3 {An enigmatic move that shows why Ding Liren is a special player. He is not worried with the hanging b5-bishop nor with the queen on d3. Instead, he just reinforces his position and advances the pawn. Spoiler alert: this pawn could have decided the game.} (20. Ba6 {Is a move a lesser mortal would play.} Bxa6 21. Qxa6 Qb2 22. Kf1 b5 23. Ng3 {Black has some compensation.}) (20. Qe3 $6 Qxb5 21. Nxd4 Qa4 {Remember this variation. It's unclear with the pawn on h4 and winning for White with the pawn on h5.} 22. Nb3 Rxd1+ 23. Rxd1 Qxe4 24. Rd8+ Bf8 25. Qg3 Kg7 $1 26. h5 Be7 27. Rd7 Bh4 $1) 20... Be5 {It's difficult for Rapport to untangle his pieces.} (20... Qb2 21. Qe3 Qxb5 22. Nxd4 Qa4 23. Nb3 Rxd1+ 24. Rxd1 Qxe4 25. Rd8+ Bf8 26. Qg3 {What a difference this little h5-pawn makes $1 Now Black is losing because the king cannot run away.} Kg7 $2 27. h6+ $18) (20... e5 21. Nxd4 Rxd4 22. Qe2 Rxe4 23. Rc4 $1 Rxe2 24. Rxb4 Rxa2 25. Rd8+ Bf8 26. Rc4 {and the rooks crash through.}) (20... Nxe2+ 21. Qxe2 Rxd1+ 22. Rxd1 Qxe4 23. Rd8+ Bf8 24. Qxe4 Bxe4 25. h6 $1 {The little warrior locks the opponent's king.} Bd5 26. a4 {with an easy win.}) 21. a4 (21. Ba6 { This was good now.} Nxe2+ 22. Qxe2 Rxd1+ 23. Rxd1 Qxe4 24. Qxe4 Bxe4 25. Re1 f5 26. Kf1 {White prepares f3 with excellent winning chances.}) 21... Kg7 $2 ( 21... Nxe2+ {was a better chance to survive.} 22. Qxe2 Rxd1+ 23. Rxd1 Kg7) 22. Kf1 $3 {Another majestic move by Ding Liren. Now Nxe2 will not come with a check, and this gives additional options for White.} Nxe2 (22... Qe7 23. Qe3 Nxe2 24. h6+ $5 Kg8 25. Kxe2 {With a winning position. The h6 gives white too much attack because it locks the king.}) 23. Qxe2 $2 {After playing a wonderful game, he misses a simple win.} (23. Qxd8 $1 Nxc1 (23... Qxe4 24. Bxe2 Qxg2+ 25. Ke1 Qh1+ 26. Kd2 Bf4+ 27. Kc3 {and the king escapes.}) 24. Qg5 $1 { Followed by h6+. Black can resign.}) 23... Rxd1+ 24. Rxd1 a6 $2 (24... Bd4 { was better.}) 25. Bxa6 Bxa6 26. Qxa6 Qxe4 27. Re1 $2 (27. hxg6 hxg6 28. Qe2 $1 Qf5 29. Kg1 {Black's problem is that he want be able to defend the b6-pawn so White is winning.}) 27... Qd4 28. Qe2 Bf6 29. hxg6 $2 {Another mistake. The h5-pawn was the pride of White's position. Ding Liren should advance it to h6 somehow, instead of exchanging it. He could have achieved this with a brilliant maneuver suggested by the machine.} (29. Qe4 $1 Qc5 (29... Qd2 30. Qe3 $1 Qb4 31. h6+ {achieving the desired goal to advance the pawn.}) 30. Qf3 $1 gxh5 (30... Qxh5 31. Qxh5 gxh5 32. Ke2 {with an easy win.}) 31. Qg3+ Kf8 32. Qf4 Kg7 33. Re4 {Avoiding Bd4 and preparing Rc4. The computer claims white is totally winning here. The heavy pieces will coordinate a decisive attack.}) 29... hxg6 30. Qe4 Qd2 {and now Rapport has decent compensation.} 31. Re2 Qd1+ 32. Re1 Qd2 33. g3 Bd4 34. Qe2 Qc3 35. Rd1 Qc6 36. Qg4 e5 37. Rxd4 exd4 38. Qxd4+ Kg8 39. Qd8+ Kg7 40. Qd4+ Kg8 {A big miss for Ding Liren.} 1/2-1/2 [Event "Chess.com"] [Site "Chess.com"] [Date "2022.06.19"] [Round "3"] [White "Firouzja, Alireza"] [Black "Nakamura, Hikaru"] [Result "1/2-1/2"] [ECO "E36"] [WhiteElo "2793"] [BlackElo "2760"] [Annotator "samsh"] [PlyCount "107"] [EventDate "2022.??.??"] 1. d4 {The third round saw four draws, but some interesting games. Ding Liren showed more good preparation with White and clearly looks like he came ready with a lot of ideas, but his actual gameplay has been lackluster. The most interesting game of the day saw Firouzja play his first game with the white pieces. He managed to put Hikaru under a lot of pressure, but ultimately was unable to win against extremely good defense.} Nf6 2. c4 e6 3. Nc3 Bb4 4. Qc2 O-O 5. a3 Bxc3+ 6. Qxc3 d5 7. Bg5 {All of this is very topical. Now, in the past, Hikaru has preferred h6, which he has clearly analyzed very well.} dxc4 ( 7... h6 8. Bxf6 Qxf6 9. Nf3 dxc4 10. Qxc4 Nc6 11. Qc3 Re8 12. Rd1 e5 13. d5 Nb8 14. e3 Bg4 15. Be2 Nd7 16. O-O Bxf3 17. Bxf3 e4 18. Qxc7 exf3 19. Qxd7 fxg2 20. Rfe1 Qxb2 $1 {Hikaru once blitzed out all of these moves with Black against me at the 2021 Grand Chess Tour. He held the game very easily, and in the time since then, I was not able to find much in the way of improvements for White. I'm not totally sure why he decided to deviate.}) 8. Qxc4 b6 9. Rd1 Ba6 10. Qa4 h6 11. Bh4 Qe7 12. Nf3 Rd8 {We are still following a heavy mainline of theory. Hikaru has had this position some 5 times in the past, though generally from a different move order.} 13. Qc2 {But now this is rare. In the upcoming moves, Hikaru really spent a lot of time and I was a bit worried. The position is certainly not bad yet, but it is very sharp and Firouzja had clearly come well-prepared.} c5 14. e4 Bxf1 15. Rxf1 g5 16. Nxg5 hxg5 17. Bxg5 Nc6 18. Qc1 { Firouzja was still blitzing to here. According to the machine, everything holds for Black. But this is very deceptive and in human terms, it looks quite challenging to play.} Rxd4 (18... e5 19. d5 Nd4 20. Rd3 {For example, the computer swears this position is dead equal. With Black, I would seriously worry about losing in 3 more moves.}) 19. Rxd4 Nxd4 20. Qf4 Nc2+ $1 {I like this decision a lot. It is not the best move and White has a little pull in the endgame, but I think it should be defensible and simplifying down a bit will help the defense become easier.} (20... Qd7 {The computer, with nerves of steel, shows this is a draw.} 21. Bxf6 Nc2+ $1 22. Ke2 Qb5+ 23. Kd2 Qxf1 $1 { By some miracle, White is not giving mate here.} 24. Qg5+ {But the game goes on all the same $1} Kf8 25. Qg7+ Ke8 26. Qg8+ Kd7 27. Qxf7+ Kc6 28. Qxe6+ Kb5 29. Kxc2 {And apparently this is equal. No human can play chess like this.}) 21. Kd1 Qd7+ 22. Kxc2 Qa4+ 23. Kb1 Qxe4+ 24. Qxe4 Nxe4 25. Bf4 {Black's position is a bit unpleasant thanks to White's potential for an outside passed pawn mass on the kingside, but it should be defensible. From here on out, Hikaru started to play really well.} Nf6 $1 {Regrouping the knight to d5 is a good place to start.} 26. h4 Nd5 27. Bd2 f5 28. g3 Kf7 29. f3 Rh8 30. Bg5 c4 ( 30... e5 $5 {This may have been easier.}) 31. g4 b5 {White surely has better chances, but it is very hard for him to coordinate the kingside pawns. The bishop is stuck on g5, but h4-h5 is not really a thing if Black can respond with Rh8-g8.} 32. Rd1 $1 {I like this decision a lot. White's best plan will be to threaten gxf5, forcing Black to exchange the f-pawns. Then, the king on f7 will become less secure.} fxg4 33. fxg4 a5 34. h5 $1 {Now this move becomes possible. It looks like there is an issue on the g-file, but because Black was forced to exchange the f-pawns, the check on f1 messes up his coordination.} Rg8 35. Rf1+ $1 {Black's position is really difficult, but Hikaru played amazing to keep it together. One of his must underappreciated traits is how good a defender he is. I've only ever gotten one bad position against him in classical chess, and I got crushed. I've gotten a lot of good positions... but whenever his position becomes close to critical, it is like playing against a tablebase. I've drawn every other time.} Ke8 $1 {Now, White's pawn is forced forward to h6 to avoid losing material. But, this allows Black to set up a light square blockade.} (35... Kg7 {The machine prefers this one, and I despise the move. After} 36. Bc1 $1 {Black is unable to exchange the rooks, g5 is coming, and the position looks critical to ,e/}) 36. h6 Rf8 $1 {The reason my computer disliked Ke8 was that it thought White wins after trading rooks here. Only when I let it run in this position did it realize that will be a draw.} 37. Rxf8+ (37. Rc1 {In hindsight, Firouzja could have tried this, hoping to open another front with a3-a4 someday. But I doubt he would win the game.} Kf7 38. a4 Kg6 39. Bd2 bxa4 40. Rxc4 Rf1+ {Black has some work to do, but he should hold with good defense.}) 37... Kxf8 38. Bd8 a4 $3 {This takes serious nerves. It feels like allowing White's pawns to g6 and h7 should lose routinely as the bishop can come to the long diagonal to force the h-pawn through, but somehow Black is holding it together.} (38... Kf7 {It feels much more natural to pitch the a-pawn and try to blockade on the light squares instead, but now White can win on the queenside.} 39. Bxa5 Kg6 40. Bd2 { White's king will eventually walk to f3 and e4 and d4 to win the game on the queenside. If Black had his a-pawn here on a4, then at some moment b5-b4 would come for counterplay.}) 39. g5 Kf7 40. h7 Kg7 41. g6 {This looks so hopeless. Black's knight is forever stuck guarding f6, the king can't move, all White needs to do is bring in the king to win the game. But this proves impossible.} c3 $1 {A final important move, but a natural one. Black needs to trade off a pawn before he can sit and wait.} (41... Kh8 42. Ba5 $1 {White wins with a check on c3.}) 42. bxc3 Nxc3+ 43. Kc2 Nd5 {Now, funnily enough, it turns out White has no way to bring his king in. As soon it touches d3 or e2, a check on f4 will ruin his day. Black can put his pawn on e5, stopping any hope of the bishop reaching the long diagonal before Nf4 can come. I have to imagine Firouzja was incredibly frustrated that the position is not winning.} 44. Bg5 { White is ready for Bc1-b2.} (44. Ba5 e5 $1 {White does not get to c3.} 45. Bc3 $2 Nxc3 $1 {Now Black even wins.} 46. Kxc3 e4 $1 47. Kd4 {And here we see why trading the b-pawn for the c-pawn was so important.} b4 $1 {With b2 and c4 still onboard, Kxc4 would win routinely. As is, White even loses $1}) 44... e5 $1 {Just in time. Now, the diagonal will remain blocked for one critical move, and Black is in time to meet Bc1 with Ne7.} 45. Kd2 Kh8 46. Ke1 {Allowing the king to e4 would certainly lose, but this is a slow plan for White, and now Black can mobilize the queenside.} b4 $1 47. axb4 a3 {White has to bring the bishop back.} 48. Bc1 Nxb4 $1 {No more check on f6.} 49. Kd2 Nd5 $1 {The final only move. Black's knight makes it to f4 just in time.} 50. Bxa3 Nf4 $1 51. Bb2 Nxg6 52. Bxe5+ Nxe5 53. Kc3 Kxh7 54. Kd4 {A great defense from Hikaru.} 1/2-1/2 [Event "Chess.com"] [Site "Chess.com"] [Date "2022.06.19"] [Round "3"] [White "Radjabov, Teimour"] [Black "Nepomniachtchi, Ian"] [Result "1/2-1/2"] [ECO "E04"] [WhiteElo "2753"] [BlackElo "2766"] [Annotator "Rafael Leitao"] [PlyCount "59"] [EventDate "2022.??.??"] [TimeControl "5400+30"] {Nothing happened in this game. I suspect Radjabov was frustrated with his yesterday game, and Nepomniachtchi felt ok with an easy and quick draw with Black.} 1. d4 Nf6 2. c4 e6 3. Nf3 d5 4. g3 dxc4 5. Bg2 c5 6. O-O Nc6 7. dxc5 Qxd1 8. Rxd1 Bxc5 9. Nbd2 c3 $1 {An important move to disturb White's pawn structure, otherwise he gets a free Catalan style advantage. Of course, this is well known.} 10. bxc3 O-O 11. Nb3 Be7 12. Nfd4 Bd7 13. Nxc6 Bxc6 14. Bxc6 bxc6 15. Be3 (15. c4 {This is the main move.}) 15... Nd5 16. Bd4 c5 17. Nxc5 { A new move.} (17. Be5 c4 $1 (17... Bf6 $6 18. Bxf6 Nxf6 19. Nxc5 (19. c4 Rfc8 { with equality in Mikhalevski-Michalczak, Zurich 2009.}) 19... Rfc8 20. Na4 Nd5 21. Rab1 Nxc3 22. Nxc3 Rxc3 23. Rd7 {and Black will suffer a bit in this rook endgame.}) (17... f6 18. c4 $1 (18. e4 Nb6 19. Bd6 Kf7 20. Bxc5 Bxc5 21. Nxc5 Rfc8 22. Nb3 Rxc3 {with a draw.}) 18... Nb6 19. Bd6 {It should be a draw anyway, but Black will have to play a few good moves.}) 18. Na5 (18. e4 Nxc3 19. Bxc3 cxb3 20. axb3 {followed by a handshake.}) 18... Rfc8 {with a good position for Black.}) 17... Bxc5 18. Bxc5 Rfc8 19. Bd4 Nxc3 20. Bxc3 Rxc3 { Now it's over.} 21. Rac1 Ra3 22. Rc2 h5 23. Rd7 a6 24. Kg2 g6 25. e3 Kg7 26. Rcc7 Rf8 27. Rd6 Re8 28. Rdd7 Rf8 29. Rd6 Re8 30. Rdd7 1/2-1/2