The Year Computer Chess Overtook The Very Best Human Chess Play.
2 posters
Author
Message
mwyoung
Posts : 880 Join date : 2020-11-25 Location : USA
Subject: The Year Computer Chess Overtook The Very Best Human Chess Play. Wed Feb 08, 2023 10:48 pm
In this analysis we will look at the most famous of the human vs chess computer matches. Played with the best computer chess engines against the best human chess players. We will use centipawn analysis to determine the year in which computer chess programs overtook the best human chess play.
In my analysis GM Carlsen, GM Kasparov, and GM Fischer were able to achieve a average centipawn loss of .12 CP per move average at their peak play over a average of 50 of their best games.
We will not use the match score, but only use the computers accuracy to determine when the best computers overtook .12 average CP loss as when chess computers overtook the best human chess players.
Stockfish 20230203 will be the judge of game accuracy for both the human and computer chess players.
1. e4 {Keene Borik} c5 2. c3 {Alapin. This was not unexpected, since at first glance it looks like a good choice for the computer. Kasparov is one of the world's leading experts in the main line of the Sicilian, so it is natural for the Deep Blue team to select a solid line which is not directly in the mainstream.} d5 {Drawing gasps from the audience and the commentators. This move opens up the center and plays into the kind of position computers normally love. But Garry has prepared something specific for the machine.} 3. exd5 Qxd5 4. d4 Nf6 5. Nf3 Bg4 6. Be2 e6 7. h3 Bh5 8. O-O Nc6 9. Be3 cxd4 10. cxd4 {Stockfish dev-20230203-8d3457a9:} Bb4 {Normally 10...Be7 would be considered, keeping the bishop as a defensive piece for the black king. But Garry has studied this line carefully and is prepared to do open battle against Deep Blue.} 11. a3 Ba5 12. Nc3 Qd6 {OK} 13. Nb5 Qe7 {OK} 14. Ne5 Bxe2 {OK} 15. Qxe2 O-O 16. Rac1 {OK} Rac8 17. Bg5 {OK} Bb6 {Inaccurate} 18. Bxf6 {OK} gxf6 19. Nc4 {OK} Rfd8 {OK Naturally not...} (19... Nxd4 $2 20. Nxd4 Bxd4 21. Qg4+ {and White wins a piece.}) 20. Nxb6 axb6 21. Rfd1 f5 {Best} 22. Qe3 Qf6 $1 {. Strong} 23. d5 {Kasparov was taken completely by surprise. This is the kind of positional sacrifice computers are not supposed to play. Later we found that by sheer brute force Deep Blue had calculated that it could win back the pawn by force.} (23. Qg3+ Kh8 (23... Qg6 $6) 24. Nd6 Rg8 25. Qc3 $1 Rcd8 26. Nxf7+ $1 Qxf7 27. d5+ $16 {Fritz4}) 23... Rxd5 {OK} (23... exd5 24. Qxb6 Qxb2 25. Qxb7 Rb8 26. Qxc6 Rxb5 (26... Qxb5 27. Qf6 $16) 27. Rc3 {threatening Rg3+} f4 28. Rcd3 $16) 24. Rxd5 exd5 25. b3 {The computer is playing excellent moves. It is clear that the pawn on b6 cannot escape its fate. A human player might have played the rote 25.b4, but that allows a knight outpost in some lines after Nc6-e5-c4.} Kh8 {OK Kasparov goes for mate! He is playing too aggressively and adopts the wrong plan against a machine which looks at 100 million positions per second.} (25... d4 $2 26. Nxd4 $16) (25... Qe6 26. Qd2 $1 {followed by Rc1 with ample compensation for the pawn.} (26. Qxb6 $4 Nd8 $1 $19)) (25... Ne7 $5 26. Rxc8+ (26. Qg3+ {John Nunn} Qg7 (26... Qg6 27. Rxc8+ Nxc8 28. Qb8 Qc6 29. Na7 Qc1+ 30. Kh2 $18) 27. Rxc8+ Nxc8 28. Qc7 Qf8 (28... Qa1+ 29. Kh2 Kg7 30. a4 $1) 29. b4 {there are other good moves} Qe8 30. Qxb7 {looks promising for White}) 26... Nxc8 27. Qe8+ Kg7 28. Qxc8 Qa1+ 29. Kh2 Qe5+ 30. g3 Qe2 31. Qxf5 {only move to avoid perpetual check} Qxb5 {and probably drawn.}) (25... Rd8 $1 26. Qxb6 Rd7 {should hold the game quite easily.}) 26. Qxb6 {OK} Rg8 {Best} 27. Qc5 {Naturally not...} (27. Qxb7 Qg5 {threatening mate and forking the rook on c1.}) 27... d4 {[#] Mistake (0.9) Better was probably...} (27... Qg5 28. g3 Qd2 {protecting the weak d-pawn.} 29. Qc3+ (29. Nd6 {Deep Blue would have played this and thought White has a slight edge.}) 29... Qxc3 30. Rxc3 Rd8 31. Rd3 Kg7 {leads to a draw.}) (27... Rd8 $11) 28. Nd6 $1 $16 {Strong} f4 {Best} 29. Nxb7 {OK Time Magazine's correspondent Charles Krauthammer wrote: "Deep Blue's king was under savage assault by a World Champion. Any human Any human under such assault by a world champion would be staring at his own king trying to figure out how to get away. Instead, Deep Blue ignored the threat and quite nonchalantly went hunting for lowly pawns at the other end of the board. In fact, at the point of maximum peril, Deep Blue expended two moves - many have died giving Kasparov even one - to snap one pawn. It was as if, at Gettysburg, General Meade had sent his soldiers out for a spot of apple-picking moments before Beckett's charge, because he had calculated that they could get back to their positions with a half-second to spare. In humans, that is called sangfroid. And if you don't have any sang, you can be very froid. But then again if Meade had known absolutely - by calculating the precise trajectories of all the bullets and all the bayonets and all the cannons in Pickett's division - the time of arrival of the enemy he could, indeed, without fear, have ordered his men to pick apples. Which is exactly what Deep Blue did. It had calculated every possible combination of Kasparov's available moves and determined with absolute certainty that it could return from its pawn-picking expedition and destroy Kasparov exactly one move before Kasparov could destroy it. Which it did. It takes more than nerves of steel to do that. It takes a silicon brain. No human can achieve absolute certainty because no human can be sure to have seen everything. Deep Blue can."} Ne5 {OK} 30. Qd5 {OK} f3 {Mistake (1.2)} (30... Rg6 $14) 31. g3 $1 $18 {Strong} Nd3 {To give you an impression of the volatility of the position let us take a look at some of the lines Kasparov had to consider:} (31... Qf4 {threatening Rxg3+ and Qxc1} 32. Kh2 $2 (32. Rc8 $3 Qg5 33. Rc5 $1 {and White has everything under perfect control. However, if White instead had played} (33. h4 $4 {he would have run into a mating net:} Rxc8 $3 34. hxg5 Rc1+ 35. Kh2 Ng4+ 36. Kh3 Nxf2+ 37. Kh4 Rh1#)) 32... Rxg3 $3 {and Black mates:} 33. Rc8+ Rg8+ 34. Kh1 Rxc8 35. Nc5 Qc1+ 36. Kh2 Qg5 37. Qxe5+ Qxe5+ 38. Kh1 Qg5 39. Ne6 Qg2#) 32. Rc7 {OK} (32. Rc6 $2 Rg5 {is unclear (Hsu)}) 32... Re8 $2 {[#] Mistake (2.6)} (32... Nf4 $5 33. Qxf7 (33. Qxf3 {Deep Blue thought this gave White a large advantage.}) 33... Nxh3+ 34. Kf1 Qxf7 (34... Qa6+ 35. Rc4 {and Qxf3}) 35. Rxf7 Ng5 36. Rd7 Rc8 {with chances of saving the game.}) (32... Qe5) 33. Nd6 $1 $18 {Strong Deep Blue goes for the most ruthless win. Normal players would have opted for} (33. Qxf7 Re1+ 34. Kh2 Qxf7 35. Rxf7 {with a comfortable win for White.}) 33... Re1+ {OK} 34. Kh2 Nxf2 {OK Kasparov is one move from mating the computer. But of course Deep Blue has calculated everything to the end.} 35. Nxf7+ {Best (Easy)} Kg7 (35... Qxf7 36. Qd8+ Kg7 37. Rxf7+ Kxf7 38. Qd5+ Ke7 39. Qxf3 {is an easy win for White.}) 36. Ng5+ Kh6 {Inaccurate} 37. Rxh7+ {Best. Loses game: --- --- Missed win: --- --- Mistake: --- Black=3 Inaccurate: --- Black=2 OK: White=8 Black=9 Best: White=2 Black=3 Strong: White=3 Black=1 Brilliant: --- --- . Weighted Error Value: White=0.07/Black=0.36 Centipawn loss: w=6/b=41 and Black resigned, because of} (37. Rxh7+ Kg6 38. Qg8+ Kf5 39. Nxf3 {after which Black's mate threat has disappeared and his is hopelessly behind on material.}) 1-0
{In game two Kasparov decided to play a conservative opening which required long-term strategy rather than tactical brilliance. He wanted to test the computer, give it a chance to behave like a machine. Garry was prepared to trade short-term advantages for long-term weaknesses.} 1. Nf3 {Keene Borik} d5 2. d4 {A very flexible move order allowing White to keep his options open for as long as possible. But now suddenly the machine started to think! The openings book had beed edited shortly before the game and copied to the wrong machine.} e6 {This caused Kasparov some consternation. It is a very profound reply originally introduced by Ivanchuk. How could Deep Blue find it by pure calculation?} 3. g3 c5 4. Bg2 Nc6 5. O-O Nf6 6. c4 {Here the computer operator, Feng-hsuing Hsu, executed the move 6..cxd4 on the board. Garry's face lit up, he looked like a kid who's parents had given him a big box of candy. After some minutes, however, Hsu suddenly discovered that the computer had displayed 6...dxc4 on the screen. The rules allowed him to call the arbitor and correct the move, with a time adjustment for White. Kasparov was obviously quite upset to have the box of candy snatched away from him.} dxc4 7. Ne5 Bd7 (7... Nxd4 8. e3 Nf5 (8... Nc6 9. Bxc6+ bxc6 10. Qxd8+ Kxd8 11. Nxf7+) 9. Qxd8+ Kxd8 10. Nxf7+) 8. Na3 cxd4 9. Naxc4 {Stockfish dev-20230203-8d3457a9:} Bc5 10. Qb3 O-O { "He won't dare take the pawn against Deep Blue". This was the general consensus in the commentary hall...} 11. Qxb7 Nxe5 {OK} 12. Nxe5 Rb8 {OK} 13. Qf3 {OK} Bd6 {This is all still theory, and it was completely amazing to find out, after the match, that Deep Blue had found every move by pure calculation.} 14. Nc6 {OK This is a new move introduced by Kasparov, who didn't like...} (14. Bf4 {and the simplification after} Nd5) (14. Nxd7 {Kasparov: "After this it was much easier for the computer to find the correct moves."}) 14... Bxc6 {OK} 15. Qxc6 e5 {OK} 16. Rb1 {White has the advantage of the bishop pair and controls the light squares. The latter concept, easily understandable to humans, seems beyond the scope of computers, as we shall see in other games in this match.} (16. b3 $6 {allows Black to use c3 to penetrate White's position , as any Grunfeld expert knows.}) 16... Rb6 17. Qa4 {OK} Qb8 18. Bg5 {OK} Be7 (18... Rxb2 $2 19. Rxb2 Qxb2 20. Bxf6 gxf6 21. Qd7 {with Qf5, Qh5 and Be4 to follow. It is remarkable that Deep Blue recognizes the gravity of this attack and declines the pawn offer on b2.}) 19. b4 $1 {OK} Bxb4 {Inaccurate This time Deep Blue cannot resist. Maybe this is because White's attack is not quite as direct and the computer cannot "smell" the danger any more. Feng-Hsiung Hsu said this was a "positional mistake caused in part by an evaluation weighting error that was not caught".} (19... Rxb4 20. Rxb4 Qxb4 (20... Bxb4 $4 21. Rb1 a5 22. a3) 21. Qxa7 {and the passed a-pawn, which promotes on a square controlled by the light-squared bishop, gives White very good chances.}) 20. Bxf6 {Best} gxf6 21. Qd7 {OK} Qc8 {OK Now Deep Blue calculates that it cannot afford to allow White to bring his pieces to bear on the black king (Qf5 or Qg4+ and Be4 would be deadly). It is willing to give back the pawn to prevent this.} 22. Qxa7 {OK} Rb8 {Inaccurate} (22... Ra6 23. Qb7 Qxb7 24. Bxb7 Rb6 {In the commentary room it was decided that this gave Black the best chances to hold the position.}) 23. Qa4 {OK} Bc3 {Mistake (1.2)} (23... Bd6 $16) 24. Rxb8 $18 {Best (Easy)} Qxb8 25. Be4 {OK} Qc7 26. Qa6 {OK} Kg7 {. OK} 27. Qd3 $6 {OK} Rb8 $1 {OK Kasparov expected 27...h6 28.Qf3 with options on Qf5 or Qh5 and f4. But the computer finds a disconcerting active defence.} 28. Bxh7 {Mistake (1.4)} (28. Qf3 $18) 28... Rb2 $16 {Best} 29. Be4 {There was a raging debate in the commentary room on whether White could win without the a-pawn. However this is somewhat academic since the pawn will fall anyway.} (29. a4 $6 Rd2 30. Qe4 Qc4) 29... Rxa2 {OK} 30. h4 {Inaccurate} Qc8 {OK} 31. Qf3 {OK} Ra1 $1 {Strong} 32. Rxa1 {OK} Bxa1 33. Qh5 Qh8 {OK} 34. Qg4+ Kf8 {Best White has a dangerous passed pawn on the h-file, Black a weak pawn on f7 and an exposed king. Still the game could probably be held by a strong player, as Kasparov later admitted. However, this requires full understanding of the nature of the position and a strategy required to defend it. The computer lacks this and Kasparov is able to masterfully lure it to its distruction.} 35. Qc8+ Kg7 {Best (Easy)} 36. Qg4+ {OK} Kf8 {Best repeating to gain time to move 40.} 37. Bd5 Ke7 {The computer sees no danger in moving its king away from the kingside defence.} 38. Bc6 Kf8 {OK} 39. Bd5 {OK} Ke7 {a few more minutes saved.} 40. Qf3 Bc3 {Best (Easy)} 41. Bc4 Qc8 $2 {Loses game (2.7)} (41... Kf8 $1 $14) 42. Qd5 Qe6 {OK} 43. Qb5 {Mistake (1.3)} (43. Qb7+ Qd7 44. Qxd7+ Kxd7 45. Bxf7 {is not enough, since Black can now stop the h-pawn with his bishop and has excellent drawing chances. White must capture the f-pawn while still keeping the queen on the board.}) (43. Qb7+ $142) 43... Qd7 $18 {Best} 44. Qc5+ {OK} Qd6 {Best} 45. Qa7+ {OK} Qd7 {Best} 46. Qa8 Qc7 $2 {Best} (46... Qd8 47. Qxd8+ Kxd8 48. Bxf7 {is again better for Black, since it gives him real drawing chances.}) 47. Qa3+ Qd6 {Best} 48. Qa2 {. OK} f5 {Inaccurate A pseudo-active move. It is probably best to keep the pawns on the dark squares and move the king towards the h-file. The commentators and public were also looking ferociously for black counterplay with lines like...} (48... e4 $5 49. Bxf7 e3 50. Qa7+) 49. Bxf7 {Best (Easy) the f-pawn falls with queens on the board.} e4 {[#] OK With the black pawns advancing down the middle the public was getting a little nervous. But both Kasparov, Deep Blue and Fritz in the commentary room knew that Black is losing.} 50. Bh5 $1 {Strong} Qf6 51. Qa3+ {OK} Kd7 52. Qa7+ Kd8 {OK} 53. Qb8+ Kd7 $2 {Mistake (1.6)} (53... Ke7) 54. Be8+ $18 {OK} Ke7 {OK} 55. Bb5 {OK} Bd2 {Mistake (1.6)} (55... Kf7 $142) 56. Qc7+ Kf8 {Best} 57. Bc4 Bc3 58. Kg2 {OK} Be1 59. Kf1 Bc3 60. f4 {OK the f-pawn falls, whether Black captures or not.} exf3 61. exf3 {Cut Off at move 61} Bd2 62. f4 Ke8 63. Qc8+ Ke7 64. Qc5+ Kd8 65. Bd3 Be3 66. Qxf5 Qc6 {now Deep Blue cannot afford to trade queens...} 67. Qf8+ Kc7 68. Qe7+ Kc8 69. Bf5+ Kb8 70. Qd8+ Kb7 71. Qd7+ {but in the end it is forced to do so.} Qxd7 72. Bxd7 {Murray Campbell, who was brought in to operate the machine, phones the control center for permission to resign, but GM Benjamin, who has to make the decision, wants to take one last look at the position. Kasparov stares in disbelief as Deep Blue plays on.} Kc7 73. Bb5 Kd6 {Loses game: --- Black=1 Missed win: --- --- Mistake: White=2 Black=3 Inaccurate: White=1 Black=3 OK: White=23 Black=15 Best: White=3 Black=11 Strong: White=1 Black=1 Brilliant: --- --- . Weighted Error Value: White=0.11/Black=0.28 Centipawn loss: w=11/b=24 and now Campbell is allowed to resign for the machine.} 1-0
1. e4 {Keene Borik} c5 2. c3 d5 3. exd5 Qxd5 4. d4 Nf6 5. Nf3 Bg4 6. Be2 e6 7. O-O Nc6 8. Be3 cxd4 {Stockfish dev-20230203-8d3457a9:} 9. cxd4 Bb4 10. a3 Ba5 11. Nc3 Qd6 {OK With the exception of the moves 7.h3 Bh5 we have the same position as in game one. Kasparov is willing to take up the battle under the same conditions.} 12. Ne5 {Obviously the Deep Blue team have been studying the first game and reached the conclusion that Black had achieved a satisfactory position. GM Joel Benjamin, responsible for Deep Blue's openings preparation, has a new move for Kasparov.} Bxe2 (12... Nxe5 13. dxe5 Qxd1 14. Bxd1 Bxd1 15. exf6 {and White has a clear advantage.}) 13. Qxe2 Bxc3 {OK} (13... Nxe5 14. Bf4 Nf3+ 15. Qxf3 Qc6 16. Qxc6+ bxc6 17. b4 {is better for White.}) 14. bxc3 Nxe5 {OK} 15. Bf4 {OK} Nf3+ {OK} 16. Qxf3 {Best (Easy)} Qd5 {OK} 17. Qd3 {OK} Rc8 {OK} 18. Rfc1 {OK In this line Joel Benjamin had intended...} (18. Be5 {but after} Nd7 19. Bxg7 Rg8 20. Qxh7 Nf6 21. Qh6 Nh5 {Black is winning a piece! Benjamin: "Thank heavens we didn't put 18.Be5 into the book."}) 18... Qc4 {OK} 19. Qxc4 {OK} Rxc4 { In the commentary hall the GMs were predicting an easy win for Kasparov. Black has a clear target - the pawn on c3 - which he can attack it with the moves ...Nd5, ...Kd7 and ...Rc8. But Deep Blue comes up with truly remarkable counterplay, which stuns the World Champion.} 20. Rcb1 $1 {OK} b6 {OK} 21. Bb8 $1 {OK "The computer played the only moves that don't lose," Garry told us later. "I'd like to know how many humans would find this plan." But in the commentary hall Fritz is predicting all of Deep Blue's moves almost instantaneously.} Ra4 {OK} 22. Rb4 {Best "Another inspired move," writes Raymond Keene in his book on the match. Again, Fritz took 0:00 seconds to come up with White's move.} Ra5 {OK} 23. Rc4 {OK "Any human would play 23.c4, almost instinctively, without thinking," said Kasparov.} O-O 24. Bd6 {OK} Ra8 {OK} 25. Rc6 {Note that White's rook has moved from the passive c1 to the dominant square c6.} b5 {OK} 26. Kf1 Ra4 {OK} 27. Rb1 a6 {OK} 28. Ke2 {OK} h5 29. Kd3 {OK} Rd8 {OK} 30. Be7 Rd7 {OK} 31. Bxf6 gxf6 32. Rb3 {OK} Kg7 33. Ke3 e5 34. g3 {OK} exd4+ 35. cxd4 Re7+ {OK} (35... Rdxd4 36. Rxa6 $1) 36. Kf3 Rd7 {OK} 37. Rd3 Raxd4 {OK} 38. Rxd4 Rxd4 39. Rxa6 {Best (Easy). Loses game: --- --- Missed win: --- --- Mistake: --- --- Inaccurate: --- --- OK: White=12 Black=19 Best: White=3 --- Strong: --- --- Brilliant: --- --- . Weighted Error Value: White=0.02/Black=0.02 Centipawn loss: w=2/b=2 and draw agreed by the Deep Blue team, although the computer showed a positive score.} 1/2-1/2
1. Nf3 {Keene Borik} d5 2. d4 c6 {This time the book is on and Deep Blue deviates from the second game.} 3. c4 e6 4. Nbd2 Nf6 5. e3 Nbd7 6. Bd3 Bd6 7. e4 {opening up the position again. Kasparov is going to slug it out against the computer, and the heart of the computer expert on his team sinks.} dxe4 8. Nxe4 Nxe4 9. Bxe4 O-O 10. O-O h6 11. Bc2 e5 12. Re1 {Stockfish dev-20230203-8d3457a9:} exd4 13. Qxd4 Bc5 14. Qc3 a5 {OK} 15. a3 {OK} Nf6 {} 16. Be3 {Inaccurate A calm move which gives White a slight positional advantage.} (16. Bxh6 {Against a human, Kasparov confided, he would have seriously considered this sacrifice, but not against a machine running at hundred million nodes per second. "The slightest miscalculation and you are dead!"} gxh6 17. Rad1 Nd7 (17... Qb6 18. Qxf6 Bxf2+ 19. Kh1 {and White is winning, because after} Bxe1 {White has} 20. Ng5 {and mate.}) 18. Re4 f5 19. Re6 {with a powerful attack.}) 16... Bxe3 {OK} 17. Rxe3 Bg4 18. Ne5 {OK} Re8 19. Rae1 Be6 {OK} 20. f4 {Inaccurate} Qc8 {Inaccurate} 21. h3 {OK} b5 {. OK In this position Kasparov thought for quite some time, summoning up the courage for a very daring move.} 22. f5 {Inaccurate Kasparov had to do a tremendous amount of calculation, considering e.g.} (22. f5 Bxf5 23. Nxf7 Rxe3 (23... Kxf7 24. Rxe8 Nxe8 25. Bxf5 {Black cannot capture the bishop because of Rf1. White is winning.}) (23... Bxc2 24. Nxh6+ $1 gxh6 25. Qxc2 Kg7 {the threat was Qg6+} 26. Rxe8 Nxe8 27. Re7+ Kf6 (27... Kf8 28. Qh7 {is hopeless for Black}) 28. Qh7 {and White will soon win.}) 24. Nxh6+ gxh6 25. Rxe3 Nh5 (25... Bxc2 26. Qxf6 Qf8 27. Qxc6 {with good chances for White}) 26. Qe5 Bxc2 27. Qxh5 {threatening mate} Qf8 28. cxb5 cxb5 29. Qxb5) 22... Bxc4 {OK Before it could make this move the computer crashed. For about 20 minutes the team worked at the terminal in the playing hall, while a distraught Kasparov, torn out of his calculations, complained bitterly about the disturbance. To compound his displeasure Deep Blue, in the end, plays a very good move. This looked a bit suspicious, but later we discovered that Fritz finds the same move in about 30 minutes. Deep Blue, being about 500 times faster, must find it in about four seconds.} 23. Nxc4 bxc4 24. Rxe8+ {OK} Nxe8 25. Re4 {OK} (25. Qxc4 {was better according to Kasparov}) 25... Nf6 26. Rxc4 {OK The position looks very promising for White because of Black's pawn islands and the permanently weak pawn on c6. But Deep Blue plays very forcefully to consolidate its position and even go for the full point.} Nd5 27. Qe5 {OK} Qd7 {OK} 28. Rg4 {OK} f6 {OK} 29. Qd4 {OK} Kh7 30. Re4 {OK} Rd8 {OK} 31. Kh1 $2 {Later Kasparov criticised this move off the diagonal because it gives the opponent unnecessary mate threats.} Qc7 32. Qf2 {OK} Qb8 {OK} 33. Ba4 $2 {OK A plausible move, but later Garry called the whole plan "rubbish". 33.b3 was clearly better.} c5 {OK} 34. Bc6 {OK Originally White had planned...} (34. Qe1 Qxb2 35. Be8 {threatening Bg6+ and mate, but now he realised that Black has} Nf4 $1 36. Rxf4 Rxe8 37. Qxe8 Qc1+ 38. Kh2 Qxf4+) 34... c4 {OK} 35. Rxc4 {OK After a safety move like...} (35. Re1 {Black has} Nf4 {and ...Nd3, with prospects of playing for a win.}) 35... Nb4 {OK threatening ...Re1+ and instant death.} 36. Bf3 {OK} Nd3 {OK} 37. Qh4 {Inaccurate Kasparov is in time trouble. Against the most awesome tactical force on the planet he has to find a series of only moves. His hands trembles as he makes them.} Qxb2 38. Qg3 {OK} Qxa3 39. Rc7 {OK} Qf8 {OK} 40. Ra7 {OK} Ne5 {OK} 41. Rxa5 {OK} Qf7 {. OK Here Kasparov offered a draw. After consultation with GM Benjamin the Deep Blue team turned down the offer.} 42. Rxe5 {Inaccurate played after about ten minutes of deliberation. Kasparov finds the safest way to a draw.} fxe5 43. Qxe5 Re8 44. Qf4 {OK Black has no real winning chances.} Qf6 (44... Rf8 45. Bg4 Kh8 46. Qg3 {and now what?}) 45. Bh5 Rf8 {OK} 46. Bg6+ {OK} Kh8 {OK} 47. Qc7 Qd4 48. Kh2 {OK} Ra8 49. Bh5 Qf6 {OK} 50. Bg6 {OK} Rg8 {Loses game: --- --- Missed win: --- --- Mistake: --- --- Inaccurate: White=5 Black=1 OK: White=23 Black=19 Best: --- --- Strong: --- --- Brilliant: --- --- . Weighted Error Value: White=0.06/Black=0.03 Centipawn loss: w=6/b=3 and draw agreed. Kasparov was feeling the pressure in what he described as a human defence against the onslaught of the machine. "I'm tired from these games, and if I was playing against a human, he would be exhausted too. But I'm playing against something that knows no such feelings and is always playing with the same strength." Deep Blue had thus far achieved a rating performance of 2775!} 1/2-1/2
1. e4 {Keene Borik} e5 2. Nf3 Nf6 {Instead of the dangerous and exhausting Sicilians Garry prepared a save Petroff for this game, an opening he had only played twice before in his life.} 3. Nc3 Nc6 {turning it into a four knights.} 4. d4 exd4 5. Nxd4 Bb4 6. Nxc6 bxc6 7. Bd3 d5 8. exd5 cxd5 9. O-O O-O 10. Bg5 c6 11. Qf3 {Stockfish dev-20230203-8d3457a9:} Be7 12. Rae1 Re8 13. Ne2 h6 {OK} 14. Bf4 {OK} Bd6 15. Nd4 Bg4 {OK} 16. Qg3 {OK} Bxf4 {OK} 17. Qxf4 Qb6 {OK} 18. c4 {A surprise for Kasparov, who had to now calculate 18...Qxb2...} Bd7 (18... Qxb2 19. Nxc6 Qxa2 20. Ne7+ Kf8 21. Nxd5 Nxd5 22. Qxg4 {which is quite comfortable for White.}) 19. cxd5 cxd5 20. Rxe8+ Rxe8 21. Qd2 Ne4 {OK} 22. Bxe4 {OK} dxe4 23. b3 Rd8 {. OK The position is slightly better for Black, but Kasparov offered a draw. Deep Blue also realized this and showed -0.10 to -0.15 for itself, but advisor Joel Benjamin decided to decline the offer. "The draw offer came much too early," said project manager C.J. Tan. "We are scientists and wanted to continue the experiment." But quite apart from that, with queens on the board and a bit of tactics left Benjamin wanted to see if Kasparov would go astray.} 24. Qc3 {OK} f5 {OK Garry: "I was tired and wanted to conserve energy for the last game. So I offered the draw in a slightly better position. Of course I had seen a promising, aggressive continuation. Now the opponent forced me to play it. My thanks to them for this assistance."} 25. Rd1 {This move will cause White some headaches on the d-file. A more plausible continuation would have been...} (25. g3 {and after e.g.} Rc8 26. Qe3 {"the game would end in four or five moves in a draw" (Kasparov).}) 25... Be6 {Inaccurate} 26. Qe3 {Inaccurate Deep Blue still doesn't smell the danger that the pin on the d-file poses.} Bf7 {OK} 27. Qc3 {Mistake (1.0) just moving around aimlessly.} (27. Rd2 $15) 27... f4 {Mistake (0.9) Deep Blue now knew it was in deep trouble.} (27... Qf6 $19) 28. Rd2 $17 {Best} Qf6 {OK replacing one pin by a new one. After the game Garry told us he should have immediately played 28...Rd5 because instead of} 29. g3 $2 {Loses game (2.3) the variation} (29. Ne2 Rxd2 30. Qxd2 Qa1+ 31. Nc1 {offered some drawing chances.}) (29. Ne2 $17) 29... Rd5 {Mistake (1.3)} (29... Rd6 $142) 30. a3 $2 {Mistake (2.6)} (30. gxf4 Qxf4 31. Qe3 Rg5+ 32. Kh1 Qg4 33. Qg3 e3 34. Qxg4 Rxg4 35. Re2 exf2 36. Rxf2 Rxd4 $19 {is a simple line which Deep Blue can see in seconds.}) (30. h3) 30... Kh7 {Inaccurate} 31. Kg2 $2 {Mistake (1.} (31. h3) 31... Qe5 $19 {. OK} 32. f3 {Deep Blue was showing it's operator how it could finish itself off in this game. From all the losing variations for Black it chose the one that puts up the most resistance. In the commentary hall we were looking at other possible moves with Fritz. Everything was losing:} (32. gxf4 Qxf4 33. Qe3 Rg5+ $19 34. Kf1 (34. Kh1 Qg4 35. f3 (35. Qg3 e3 36. fxe3 Qe4+ 37. Qf3 Bd5 38. Qxe4+ Bxe4+ 39. Rg2 Rxg2)) 34... Qxh2 35. Qxe4+ Bg6 36. Qe3 Rg4 37. Rd1 Qg2+ 38. Ke1 Re4) (32. b4 e3 33. fxe3 fxe3 34. Rd3 Qe4+ 35. Kg1 e2 $19) 32... e3 33. Rd3 {OK} e2 34. gxf4 {OK} e1=Q 35. fxe5 {OK} Qxc3 {OK} 36. Rxc3 Rxd4 {Best (Easy) and the extra pieces is enough for Black to win comfortably.} 37. b4 Bc4 {OK} 38. Kf2 {OK} g5 {OK} 39. Re3 Rd2+ 40. Ke1 Rd3 {OK} 41. Kf2 {OK} Kg6 {OK} 42. Rxd3 {Mistake (194.9)} (42. f4 $142) 42... Bxd3 $19 43. Ke3 {OK} Bc2 44. Kd4 Kf5 45. Kd5 h5 {Loses game: White=1 --- Missed win: --- --- Mistake: White=4 Black=2 Inaccurate: White=1 Black=2 OK: White=10 Black=15 Best: White=1 Black=1 Strong: --- --- Brilliant: --- --- . Weighted Error Value: White=0.62/Black=0.15 Centipawn loss: w=51/b=14} 0-1
1. Nf3 {Keene Borik} d5 2. d4 c6 3. c4 e6 4. Nbd2 Nf6 {Stockfish dev-20230203-8d3457a9:} 5. e3 c5 {a tempo-losing move to deviate from game four. The Deep Blue team has prepared a line that should open up the position.} 6. b3 Nc6 7. Bb2 cxd4 {OK} 8. exd4 Be7 {OK} 9. Rc1 O-O {OK} 10. Bd3 {OK} Bd7 {Inaccurate} 11. O-O {OK} Nh5 {Inaccurate A very strange move which absolutely nobody at the site liked.} 12. Re1 Nf4 {[#]} 13. Bb1 {OK} Bd6 14. g3 Ng6 15. Ne5 {Mistake (0.} (15. a3 $16) 15... Rc8 {Mistake (0.9)} (15... Ncxe5 $1 $14) 16. Nxd7 $16 {OK} Qxd7 17. Nf3 Bb4 18. Re3 Rfd8 {Inaccurate} 19. h4 {Inaccurate} Nge7 $2 {Loses game (2.1)} (19... Be7 $16) 20. a3 $18 {Best} Ba5 {OK} 21. b4 {OK} Bc7 22. c5 {Inaccurate Four consecutive pawn advances which drive back all the black pieces, which are stumbling over each other on the queenside.} Re8 23. Qd3 {OK} g6 {OK} 24. Re2 Nf5 {Inaccurate} 25. Bc3 h5 {OK} 26. b5 {Keene calls this "Kasparov's strategy of strangulation" and points out that 26.b5 doesn't just attack the knight but establishes "a giant, crawling mass of white pawns, rather resembling a colossal army of soldier ants on the move."} Nce7 27. Bd2 {OK} Kg7 28. a4 Ra8 {Inaccurate} 29. a5 {OK} a6 30. b6 {OK} Bb8 {. OK Kasparov has shut the black bishop and rook out of play to the end of the game. Black's position is lost.} 31. Bc2 Nc6 {OK} 32. Ba4 Re7 33. Bc3 Ne5 $2 {Mistake (2.1) actually accelerating the end.} (33... f6) 34. dxe5 $18 {Best} Qxa4 {OK} 35. Nd4 {OK} Nxd4 36. Qxd4 Qd7 {IM Otto Borik that other variations also lose:} (36... Qxd4 37. Bxd4 {and Rb2, c6 wins, e.g.} Re8 38. Rb2 Rc8 39. c6 Rxc6 40. Rxc6 bxc6 41. b7 Ra7 42. Bxa7 Bxa7 43. b8=Q Bxb8 44. Rxb8) (36... Qc6 37. Bd2 Rd7 38. Bg5 Kf8 39. Bf6 {and now f3, Rg2 and g4 wins.}) 37. Bd2 {OK} Re8 38. Bg5 {OK} Rc8 39. Bf6+ {OK} Kh7 {OK} 40. c6 $1 {OK} bxc6 {OK} (40... Rxc6 41. Rec2 Rxc2 42. Rxc2 Qe8 43. Qc5 {wins}) 41. Qc5 {OK} Kh6 42. Rb2 Qb7 43. Rb4 $1 {OK. Loses game: --- Black=1 Missed win: --- --- Mistake: White=1 Black=2 Inaccurate: White=2 Black=5 OK: White=16 Black=11 Best: White=2 --- Strong: --- --- Brilliant: --- --- . Weighted Error Value: White=0.15/Black=0.32 Centipawn loss: w=16/b=30 And the Deep Blue team reseigned for the machine. Why did Black resign? IM Malcolm Pein explained this on the Internet: "Black has four pieces left plus his king. The rook on a8 and the bishop on b8 cannot move. If the queen on b7 moves it allows b7, winning a rook. If the rook on c8 moves White can play Qxc6, forcing an exchange of queens. After that there are many ways to win, the most prosaic being double on the c file and play Rc8. So we are left with Kh7! The simplest way then is Qe7 Qxe7 Bxe7 threatening b7 and if Rc8-e8 then b7 Ra7 Bc5 etc. Note that had Kasparov left his rook on b2 Black would have ...Bxe5 gaining a tempo."} (43. Rb4 Qd7 (43... Re8 44. Qxc6 Qxc6 45. Rxc6 Kh7 46. b7 Ra7 47. Rbb6 Rg8 48. Rc8 d4 49. Rbc6 g5 (49... d3 50. Rxg8 Kxg8 51. Rc8+ Kh7 52. Rh8#) 50. Rxg8 Kxg8 51. Rc8+ Kh7 52. Rh8+ Kg6 53. Rg8+ Kf5 54. f3 {Threat: Rxg5#} Bxe5 55. Bxe5 {Threat: ditto} Kxe5 56. b8=Q+ {and mate.}) (43... Kh7 44. Qe7 Qxe7 45. Bxe7 Re8 46. b7 Ra7 47. Bc5 Rd8 48. Bxa7 Bxa7 49. Rxc6 Bb8 50. Rc8 {winning}) 44. b7) 1-0
{Deep Junior did not manage to take revenge. It had to relay on its tenacity to save a half point. My analysis is based on Andre Schulz' remarks on www. chessbase.de from yesterday evening. Deep Junior gelingt es nicht Revanche zu nehmen. Er musste im Gegenteil hart kämpfen, bevor seine große Zähigkeit ihm einen halben Punkt gesichert hat. Meine Anmerkungen basieren auf der Analyse von Andre Schulz von gestern:} 1. e4 c5 2. Nf3 e6 ({Normally Kasparov prefers the Najdorf variation: Kasparov bevorzugt sonst die Najdorf-Variante:} 2... d6 3. d4 cxd4 4. Nxd4 Nf6 5. Nc3 a6) 3. d4 cxd4 4. Nxd4 a6 5. Bd3 Bc5 ({Another set up is} 5... Nf6 6. O-O d6 7. c4 Be7 8. Nc3) 6. Nb3 {Stockfish dev-20230203-8d3457a9:} Ba7 7. c4 Nc6 8. Nc3 d6 9. O-O Nge7 10. Re1 O-O 11. Be3 {OK} e5 {OK} 12. Nd5 {Inaccurate} a5 {Kasparov has managed agian to reach a position, which is not easy to play for the computer. Kasparow ist es erneut gelungen, eine Stellung herbeizufughren, die dem Computer nicht liegt.} 13. Rc1 a4 {Kasparov fights for the square d4. Kasparov will das Feld d4 freiräumen.} 14. Bxa7 {OK} Rxa7 15. Nd2 {OK the knight is not well placed here. Der Springer steht hier nicht glücklich.} (15. Na1 $5 Nxd5 16. cxd5 Nd4 17. Nc2 {should be OK for White. sollte für Weiß OK sein.}) 15... Nd4 16. Qh5 {Inaccurate more a gesture than a threat. Mehr eine Geste als eine wirkliche Drohung.} Ne6 {aims for the very strong square c5. strebt auf das sehr gute Feld c5} 17. Rc3 (17. b4 axb3 18. axb3 Nd4 19. Ra1 Rxa1 20. Rxa1 Nxd5 $15) 17... Nc5 18. Bc2 Nxd5 {OK easier than trying to ignore and play around the Nd5 with einfacher als der Versuch, den d5 mit} (18... Nc6 {zu ignorieren und zu umspielen.}) 19. exd5 {The computer play's for an attack, which just isn't there. The knight c5 is now a giant. sehr verpflichtend, weil der c5 nun zum Riesen wird. Insgesamt spielt Weiß auf einen Angriff, der nicht stellungsgemäß ist (den es so nicht gibt).} ({I like} 19. cxd5 {more. gefällt mir besser.}) 19... g6 {the thematic ...f7-f5 does not run away. ...g6 allows the introduction of the Ra7 into the defence of the kingside with b7-b6, if neccessery. Das thematische ...f7-f5 läuft nicht weg. ...g6 ermöglicht es, den a7 ggf. mit b7-b6 in die Verteidigung am Königsflügel einzubeziehen.} 20. Qh6 {OK} f5 {OK Black's cental pawn mass is very impressive. Die schwarzen Zenrumsbauern schränken das weiße Spiel stark ein.} 21. Ra3 Qf6 {A devilish trap into which Deep Junior falls. It just likes the material too much. eine teuflische Falle, in die Deep Junior promt hineinfällt. Der Computer achtet zu sehr auf das Material und nicht auf die Koordination seiner Figuren.} (21... Ra6 {and und}) (21... Qb6 {came into consideration as well. kamen auch in Frage.}) 22. b4 {Mistake (0.0)} (22. Nf3 $11) 22... axb3 $2 {Mistake (0.0)} (22... Nd7 $16) 23. Rxa7 $17 {Best} bxc2 {OK White has to lose time now to neutralize the pawn c2. The queen h6 and the rook a7 are far away. Weiß muss nun Zeit verlieren, um den c2 zu neutralisieren. Der a7 ist ebenso wie die h6 weitab vom Schuss.} 24. Rc1 {Mistake (1.0)} (24. Ra3 $17) 24... e4 25. Rxc2 {[#] OK} (25. Qe3 $2 f4 26. Qa3 $2 Qd4 $19) 25... Qa1+ $2 {Mistake (1.5) Kasparov played this move too fast. von Kasparov zu schnell gespielt.} (25... f4 $1 {threatens e3 and cuts the queen off from the defence of White's first rank um e3 zu drohen und die Dame von der Verteidigung der Grundreihe abzuschneiden, war viel stärker:} 26. Nf1 (26. Ra8 e3 27. Nf1 Qf5 28. Rxc8 Rxc8 29. Qh3 Qxh3 30. gxh3 exf2+ 31. Rxf2 g5 $17) (26. h4 Qa1+ 27. Nf1 e3 28. h5 Bf5 $19) 26... e3 27. Nxe3 (27. fxe3 fxe3 28. Qxf8+ Kxf8 29. Ra8 Ke7 30. Rxc8 Qf5 $19) 27... Bf5 28. Rc1 Qb2 29. Rd1 fxe3 30. Qxe3 Bd3 $17) (25... f4 $1 $19) 26. Nf1 $11 {Best (Easy)} f4 {OK} 27. Ra8 $1 {Best (Easy) The only defence with the idea Qxf8+ followed by Rxc8+. Probably Kasparov had missed this ressource. Die einzige Verteidigung, mit der Idee xf8+ nebst xc8+. Diese Ressource war Kasparov vermutlich entgangen.} e3 {OK} (27... Nd3 $5 {was interesting, e.g. war interessant, z.B.} 28. Re2 Qd4 29. Qh4 Qxc4 {with initiative. mit Initiative.}) 28. fxe3 {OK} fxe3 29. Qxf8+ {OK} Kxf8 30. Rxc8+ {OK} Kf7 {Loses game: --- --- Missed win: --- --- Mistake: White=2 Black=2 Inaccurate: White=2 --- OK: White=8 Black=6 Best: White=3 --- Strong: --- --- Brilliant: --- --- . Weighted Error Value: White=0.13/Black=0.16 Centipawn loss: w=12/b=12} (30... Kf7 31. Rc7+ Kf8 (31... Ke8 $2 32. Re2 Kd8 33. Rxh7 Qc1 34. Rxe3 {is dangerous for Black. ist gefährlich für Schwarz.}) 32. Rc8+ Kf7 {with perpetual check. mit Dauerschach.}) 1/2-1/2
1. d4 d5 2. c4 c6 3. Nc3 Nf6 4. e3 e6 5. Nf3 Nbd7 6. Qc2 {Stockfish dev-20230203-8d3457a9:} b6 {Schwarz wählt ein selteneres System. Der Hauptzug 6...d6 stand in der ersten Matchpartie auf dem Plan.} 7. cxd5 {"Eigentlich" sollte dieser Zug nicht viel versprechen. 7.e2 und 7.d3 sind die Alternativen.} exd5 8. Bd3 Be7 {OK} 9. Bd2 {Durch das frühe ...b7-b6 hat Schwarz den c6 geschwächt, aber andererseits ist (im Vergleich etwa zu Stellungen aus dem klassichen Damengambit) der schwarzfeldrige Läufer auf d2 passiver platziert als auf f4 oder g5.} O-O {Schwarz sollte nicht so früh seine Königsstellung festlegen.} (9... Bb7 {erscheint flexibler:} 10. h3 (10. g4 Nxg4) 10... c5 $132) 10. g4 {Inaccurate Kasparov bleibt der aggressiven Linie treu, die ihm bereits in der ersten Matchpartie den Erfolg gebracht hatte.} Nxg4 {Mistake (1.2) JUNIOR muss die Folgen dieses Zuges genau berechnet haben. Nur haarscharf verliert Schwarz nun nicht sofort. Andererseits war Schwarz auch schon zu diesem Zug gezwungen, da Weiß 11.g5 droht und sowohl nach 10..h6 oder 10...g6 bekommt Weiß eine willkommene Angriffsmarke.} (10... g6 $1 $11) 11. Rg1 {Mistake (1.3)} (11. Bxh7+ $16) 11... Ndf6 {Mistake (1.4)} (11... Ngf6 12. O-O-O $44) (11... Ngf6 $1 $11) 12. h3 $16 {Best (Easy)} Nh6 13. e4 {OK} (13. O-O-O Kh8 $13) 13... dxe4 {Best} 14. Bxh6 {Mistake (0.9)} (14. Bxe4 $16) 14... exd3 $14 {OK} 15. Rxg7+ {Inaccurate} (15. Bxg7 Ng4 (15... dxc2 $2 16. Bxf6+ Bg4 17. Rxg4#) 16. Qxd3 Kxg7 17. hxg4 $14 {sollte Weiß einen leichten Vorteil geben. Nachdem er lang rochiert hat, kann Weiß weiterhin sein Spiel gegen den schwarzen König richten.}) 15... Kh8 16. Qxd3 {OK} Rg8 {Best (Easy)} 17. Rxg8+ Nxg8 18. Bf4 {OK} f6 {OK Damit hat JUNIOR seine Königstellung gesichert und Kasparov hätte eigentlich klar werden müssen, dass angesichts des offenen Charkters der Stellung JUNIOR nicht mehr verlieren sollte.} 19. O-O-O {OK} Bd6 20. Qe3 {OK} Bxf4 {OK} 21. Qxf4 Bxh3 {OK Unbeeindruckt nimmt JUNIOR einen Bauern weg und lässt zu, dass Weiß nun über die h-Linie den schwarzen König angreifen kann. Weiß hat ausreichend Kompensation für den Bauern - nicht mehr, aber auch nicht weniger.} 22. Rg1 Qb8 {OK} 23. Qe3 {Inaccurate} Qd6 {Inaccurate} 24. Nh4 {OK} Be6 25. Rh1 Rd8 26. Ng6+ {Best} Kg7 27. Nf4 {OK} Bf5 28. Nce2 {Best Weiß bringt eine weitere Figur zum Königsflügel.} Ne7 29. Ng3 {Best} Kh8 {OK} 30. Nxf5 {OK} Nxf5 31. Qe4 {OK} Qd7 {OK} 32. Rh5 {Mistake (1.3) Kasparov unterläuft ein Rechenfehler, der ihm den d4 und damit die Partie kostet.} (32. Ng6+ Kg7 33. Nf4 {hält das Gleichgewicht:} Re8 $5 (33... Kh8 $11) 34. Qg2+ Kf8 35. Ng6+ $5 hxg6 36. Qxg6 Qg7 37. Qxf5 Qg5+ $11 {mit Übergang in ein ausgeglichenes Turmendspiel.}) (32. Qg2 $11) 32... Nxd4 $17 {Best (Easy)} 33. Ng6+ $2 {Loses game (2.7)} (33. Rh1 $17) 33... Kg8 $19 {Best (Easy)} (33... Kg7 $2 34. Rxh7+ Kxh7 35. Nf8+ $18) 34. Ne7+ {OK} Kf8 {Best (Easy)} 35. Nd5 {OK Das traurige Eingeständnis des Fehlers.} (35. Rxh7 {war Kasparovs Absicht, aber dann setzt Schwarz matt:} Nb3+ 36. Kc2 (36. axb3 Qd1#) 36... Na1+ 37. Kc3 Qd2+ 38. Kc4 b5+ 39. Kc5 Qd6#) 35... Qg7 {Best} 36. Qxd4 {OK} Rxd5 {Best (Easy). Loses game: White=1 --- Missed win: --- --- Mistake: White=3 Black=2 Inaccurate: White=3 Black=1 OK: White=12 Black=8 Best: White=4 Black=7 Strong: --- --- Brilliant: --- --- . Weighted Error Value: White=0.30/Black=0.18 Centipawn loss: w=25/b=12} ({Nach} 36... Rxd5 37. Rxd5 cxd5 {kann Weiß den d5 wegen} 38. Qxd5 Qg5+ {nicht schlagen und verbleibt daher in einem hoffnungslosen Damenendspiel mit zwei Minusbauern.}) 0-1
[/pgn]
mwyoung
Posts : 880 Join date : 2020-11-25 Location : USA
Subject: Re: The Year Computer Chess Overtook The Very Best Human Chess Play. Thu Feb 09, 2023 12:24 am
1. e4 c5 2. Nf3 Nc6 3. d4 cxd4 4. Nxd4 e6 5. Nb5 d6 {Stockfish dev-20230203-8d3457a9:} 6. c4 {White has installed the Maroczy-bind. He has more space, but Kasparov just sets up a hedgehog. Weiß errichtet den Maroczyaufbau. Er hat mehr Raum, doch Kasparov baut sich einfach igelartig auf.} Nf6 7. N1c3 a6 8. Na3 {this knight is not well placed here and has already moved quite often. Black should have equality. Dieser Springer steht nicht glücklich und hat schon oft gezogen. Schwarz sollte Ausgleich haben.} Nd7 {Inaccurate to throw Deep Junior out of its opening book. Mit diesem Zug wirft Kasparov Deep Junior aus seinem Eröffnungsbuch.} ({Üblich ist} 8... Be7 {is the usual move.}) 9. Nc2 Be7 {OK} 10. Be2 {OK} b6 11. O-O Bb7 12. h3 {one of those weakening pawn moves the white players make so easily in the hedgehog. It does not fit into the following white play. Einer von den schwächenden Bauernzügen, die Weiß im Igel so gerne macht. Er passt überhaupt nicht zum folgenden Spiel von Weiß.} O-O 13. Be3 {Inaccurate} Rc8 {OK} 14. Qd2 Nce5 {OK the Nc6 was a small problem in Black's position. Normally it is redeployed to d7 via e5 and thanks to h3 Kasparov does just this now. Der c6 war ein kleines Problem in der schwarzen Aufstellung, weil er im Weg stand. Normalerweise wäre er via e5 nach d7 überführt worden, was Kasparov nun dank h3 problemlos nachholt.} 15. b3 Nf6 {Best} 16. f3 {OK does not fit to h3 (the dark squares on the kingside are now weak), but signales that Junior wants to play on the queenside, which is right here, I think. Passt zwar nicht zu h3 (die schwarzen Felder am Königsflügel sind nun schwach), ist aber dennoch nicht so schlecht, weil er signalisiert, dass im Folgenden am Damenflügel gespielt werden soll, was ich hier für besser halte als am Königsflügel anzurennen.} Qc7 {OK} 17. Rac1 Rfe8 {Inaccurate} (17... b5 $6 {plays with the fire due to spielt mit dem Feuer:} 18. cxb5 Qxc3 19. Qxc3 Rxc3 20. Bd4 Rc7 21. bxa6) 18. a3 {Inaccurate} Ned7 {Inaccurate Black has reached a typical hedgehog position Schwarz hat eine typische Igelformation eingenommen.} 19. Rfd1 {Inaccurate} Qb8 {Inaccurate leaves the vis a vis of the Rc1 and increases the harmony in Black's camp geht aus dem vis a vis mit dem c1 und erhöht die Harmonie im schwarzen Lager.} (19... Nc5 $5 20. Rb1 d5 {came strongly into consideration. kam stark in Betracht.}) 20. Bf2 {Inaccurate} (20. Nd4 $6 {allows lässt} d5 {zu.}) 20... Rcd8 {invites Junior to play on the queenside, which is not good as the computer follows a strong plan then. dieser Zug lädt Junior dazu ein, am Damenflügel zu kommen, was ich nicht für gut halte, weil der Rechner dann wieder einen sinnvollen Plan verfolgt.} ({I like} 20... Ba8 {more. gefällt mir besser.}) 21. b4 (21. Bg3 {is easily fended off with wehrt Schwarz locker mit} Ne5 {ab.}) 21... Ba8 {Inaccurate} 22. a4 {White takes the initiative on the queenside. Weiß entwicklelt Initiative am Damenflügel.} Rc8 23. Rb1 {OK} Qc7 24. a5 $5 {OK According to Anand a correct manoeuvre of the machine. Laut Anand ein korrektes Manöver des Rechners.} bxa5 {[#] OK} 25. b5 {OK} Bb7 {OK} (25... axb5 $2 26. Nxb5 Qd8 27. Nxd6 $16) 26. b6 {OK} (26. Nd4 {came strongly into consideration as well. kam auch stark in Betracht.}) 26... Qb8 {Inaccurate} 27. Ne3 {Mistake (1.0)} (27. Nd4 $16) 27... Nc5 {Inaccurate} 28. Qa2 {Best (Easy)} Nfd7 {Inaccurate} 29. Na4 {Best} Ne5 {Inaccurate} 30. Nc2 {good prophylaxis gute Prophylaxe} (30. Nxc5 $6 dxc5 31. Qxa5 Nc6 {is not, what White wants. ist nicht im weißen Sinn.}) 30... Ncd7 {Inaccurate} 31. Nd4 Red8 32. Kh1 {OK Such a computer-move always shows that the machine does not see a way to improve its position. So ein Computerzug zeigt immer, dass der Rechner nicht weiß, wie er seine Stellung verstärken soll.} Nc6 {Inaccurate} 33. Nxc6 Rxc6 34. Kg1 {Inaccurate} h6 {OK} 35. Qa3 {Junior fights again hard! Junior nimmt wieder Fahrt auf.} Rdc8 {Inaccurate Kasparov did not have much time left and had to play quickly. Kasparov hatte nun nicht mehr viel Zeit auf der Uhr und musste sich beeilen.} 36. Bg3 Bf8 $6 {OK now Bd8 is no option anymore. nun ist Ld8 keine Option mehr.} (36... Ne5 $4 37. Bxe5 dxe5 38. Qxe7 $18) (36... Kf8 $5 {Anand}) 37. Qc3 Ne5 {OK} 38. c5 {Inaccurate opens the floodgates. öffnet die Fluttore.} Nd7 {Best} 39. Qxa5 {Best (Easy)} Nxc5 {OK} 40. Nxc5 Rxc5 41. Qa4 {Inaccurate} R5c6 {OK} (41... a5 42. Bb5 {is very good for White as well. ist ebenfalls sehr gut für Weiß.}) 42. Bf2 (42. Bxa6 $2 Qa8 43. Bb5 Rxb6 {is wrong. ist falsch.}) 42... d5 43. Bxa6 {Best} Bc5 {Inaccurate} 44. Bxc5 {OK} Rxc5 45. Bxb7 {Inaccurate} (45. exd5 exd5 (45... Bxa6 46. Qxa6 exd5 47. b7 {and the b-pawn is a powerful force. und der b-Bauer ist eine Macht.}) 46. Bd3 {(Knaak) was much better. Now it is very questionable, if Black can survive. (Knaak) war viel besser. Es ist sehr fraglich, ob Schwarz sich nun halten kann.}) 45... Qxb7 46. exd5 exd5 47. Qa7 R5c7 {Inaccurate Kasparov sets a very deep trap. Kasparov stellt dem Rechner eine Art Falle, die dieser nicht durchschaut/durchschauen kann.} (47... Rb8 48. Re1 Rc7 {came into consideration as well. kam auch in Betracht.}) 48. Qxb7 $6 {OK} (48. Qa5 $5) 48... Rxb7 49. Rxd5 {Best this double rook ending is completely drawn. Dieses Doppelturmendspiel ist trotz Mehrbauern remis.} Rc6 $1 {OK} 50. Rdb5 {Best} h5 {OK Black has an impregnable fortress as the whote rook's are tied to the defence of the b-pawn. Deep Junior does not underastand the very nature of this conception and plays on. Schwarz hat eine uneinnehmbare Festung, weil sich die weißen Türme nicht vom b-Bauern lösen können. Der Rechner versteht die Natur dieses positionellen Remis nicht und spielt noch etwas weiter.} 51. Kf2 Re6 $1 {OK assures that White's king remains out of play. sichert ab, dass der König nicht eingreifen kann.} 52. f4 g6 {OK} 53. Kg3 Kg7 {OK} 54. Kh4 Kh6 {OK} 55. R1b4 Rd6 {OK} 56. g3 {Inaccurate} f6 {threatens Rbxb6 Rxb6?? g5 followed by mate. droht Tbxb6 Txb6?? g5 nebst matt.} 57. g4 {OK} hxg4 {OK} 58. hxg4 Kg7 59. Rb3 (59. f5 g5+ {leads to nothing as führt zu nichts, weil} 60. Kh5 $4 {is refuted by an} Rb8 61. b7 Rh8# {scheitert.}) 59... Rc6 60. g5 f5 {OK} 61. Rb1 {Loses game: --- --- Missed win: --- --- Mistake: White=1 --- Inaccurate: White=9 Black=14 OK: White=10 Black=20 Best: White=6 Black=2 Strong: --- --- Brilliant: --- --- . Weighted Error Value: White=0.13/Black=0.13 Centipawn loss: w=12/b=11} 1/2-1/2
1. d4 Nf6 {The Deep Junior has changed the opening strategy. They obviously don't want to see g2-g4 against the Semi Slav again. Das Deep Junior Team hat offenbar die Strategie gewechselt und will kein Kasparovsches g2-g4 gegen Halbslawisch mehr sehen!} 2. c4 e6 3. Nc3 Bb4 {The Nimzoindian Die nimzoindische Eröffnung} 4. e3 O-O 5. Bd3 d5 {now a structu re from the Queen's gambit has arisen. nun geht das Spiel in damengambitartige Strukturen über.} 6. cxd5 {frees the Bc8, but gives White the opportunity to launch a majority attack with f3 followed by e4 or a minority attack (e.g. with b4-b5, if Black plays ...c6). Befreit den c8, gibt Weiß aber die Möglichkeit einen Majoritätsangriff mit f3 nebst e4 oder einen Minoritätsangriff zu starten (insbesondere mit b4-b5, falls Schwarz c6 spielt).} exd5 7. Nge2 {Stockfish dev-20230203-8d3457a9:} Re8 8. O-O Bd6 9. a3 {rather tame. wirkt recht zahm.} c6 $146 {OK} 10. Qc2 Bxh2+ {Mistake (1.0) A very courageous sacrifice by Deep Junior! Which other computer program would have played like this? Ein sehr mutiges Opfer von Deep Junior! Welches andere Computerprogramm hätte dies gewagt?} (10... Nbd7 $11) 11. Kxh2 $16 {This is neccessary as Die Annahme ist Pflicht, denn} (11. Kh1 $2 Bc7 {is just bad. kann es natürlich nicht sein.}) 11... Ng4+ {[#] Best} 12. Kg3 $1 {Strong only move einziger Zug} (12. Kg1 $2 Qh4 13. Rd1 Qxf2+ 14. Kh1 Rxe3 $19) (12. Kh1 $4 Qh4+ 13. Kg1 Qh2#) (12. Kh3 $4 Nxe3+ 13. Kh2 Nxc2 $19) 12... Qg5 {[#] Best} 13. f4 $1 {Strong} (13. Bxh7+ $6 Kh8 14. f4 Qh5 15. Bd3 Qh2+ 16. Kf3 Qh4 17. Ng3 Nh2+ 18. Kf2 Ng4+ 19. Kf3 $11) 13... Qh5 {OK} 14. Bd2 {OK} Qh2+ {OK} (14... Rxe3+ $2 15. Bxe3 Nxe3 16. Qd2 Nxf1+ 17. Rxf1 $16) 15. Kf3 Qh4 {OK} 16. Bxh7+ {Mistake (1.2) Kasparov contends himself with a draw and puts his hopes in the last game on Friday. The alternative 16.g3!? had probably too wild complications to try it against a machine: Kasparov fügt sich ins Remis und setzt seine Hoffnungen damit in die letzte Partie am Freitag. Er hat die Alternative 16.g3 wohl wegen ihrer taktischen Komplexität verworfen:} (16. Ng3 Nh2+ 17. Kf2 Ng4+ 18. Kf3 $11 {(Gulko)}) ({Is Black's compensation after Hat Schwarz nach} 16. g3 $5 {really sufficient? This is certainly not easy to say and we have to wait for more detailed analysis. I can just look at some interesting lines: wirklich genügend Spiel für die Figur? Es ist jedenfalls nicht leicht zu sehen und man wird genauere Analysen abwarten müssen. Ich gebe lediglich einige Varianten an, für eine erschöpfende Analyse reicht die Zeit nicht aus:} Qh2 (16... Nh2+ 17. Kf2 Ng4+ 18. Ke1 $16 {Gulko: I don't like this for Black at all.} Qh3 19. Nd1 $5) 17. Rae1 (17. Rh1 $4 Qf2#) (17. f5 $6 Nd7 $1 18. Kxg4 Qg2 19. e4 ({One possible line after Eine mögliche Variante nach} 19. Rh1 $6 {is: lautet:} Nf6+ 20. Kh4 h6 21. Rag1 Bxf5 22. e4 Qf3 23. g4 dxe4 24. Bc4 Qf2+ 25. Rg3 Bxg4 {with powerful attack. mit gewaltigen Angriff.}) 19... Nf6+ 20. Kf4 dxe4 21. Bxe4 Rxe4+ 22. Nxe4 Nd5+ 23. Ke5 Bxf5 {with great complications according to IM Nikolai Vlassov in Chess Today 822. mit großen Verwicklungen. IM Nikolai Vlassov hat diese fantastische Variante in Chess Today 822 angegeben.}) 17... f5 (17... Nxe3 $2 18. Bxe3 Qh5+ 19. Kg2 Bh3+ 20. Kf2 Bxf1 21. Rxf1 Qh2+ 22. Kf3 Qh5+ 23. g4 Qh3+ 24. Ng3 Rxe3+ 25. Kxe3 Qxg3+ 26. Kd2 $18) (17... Qh5 $2 18. Rh1 $18) 18. Bxf5 Qh5 19. Bxh7+ Qxh7 20. Qxh7+ Kxh7 21. Rh1+ Kg8 22. e4 {and White is better. und Weiß steht besser.}) (16. g3 $16) 16... Kh8 $11 {OK} 17. Ng3 {OK} Nh2+ {Best (Easy)} 18. Kf2 {Best (Easy)} Ng4+ {Best (Easy)} 19. Kf3 {Best (Easy)} Nh2+ {Best (Easy). Loses game: --- --- Missed win: --- --- Mistake: White=1 Black=1 Inaccurate: --- --- OK: White=2 Black=5 Best: White=2 Black=5 Strong: White=2 --- Brilliant: --- --- . Weighted Error Value: White=0.11/Black=0.11 Centipawn loss: w=9/b=9 Deep Junior has added a new, fascinating example to the long chapter of Bxh7+ (resp Bxh2+) Greek gift sacrifices! Deep Junior hat das lange Kapitel über die Läuferopfer auf h7 (bzw h2) mit dieser Partie um ein sehr interessantes Beispiel bereichert!} 1/2-1/2
1. e4 c5 2. Nf3 d6 3. d4 cxd4 4. Nxd4 Nf6 5. Nc3 a6 6. Be2 e5 {One of the rf main lines. 5...a6 was played to deny White's pieces the important b5 square. Eine der Najdorfhauptvarianten. 5...a6 wurde gespielt, um den weißen Kräften das wichtige Feld b5 zu nehmen.} 7. Nb3 Be7 {Stockfish dev-20230203-8d3457a9:} 8. O-O O-O 9. Kh1 Bd7 10. Be3 {OK} Bc6 {OK} 11. Bf3 {does not fit well to Kh1, I think. passt irgendwie nicht recht zu Kh1, finde ich.} Nbd7 12. a4 {OK} (12. Nd5 {can be answered with kann mit} Nxd5 13. exd5 Bb5 {beantwortet werden.}) 12... b6 (12... Nc5 $5) 13. Qd3 Bb7 14. h3 {I don't like this move too much (but Junior likes rook's pawn moves....) gefällt mir nicht sonderlich (Junior mag Randbauernzüge nun mal...)} Rc8 (14... Nc5 $5 {came into consideration again. kam erneut in Betracht.}) 15. Rad1 h6 {OK} 16. Rfe1 Qc7 {OK} 17. g3 Rfd8 18. Kh2 Re8 {both sides have reached their ideal set ups and it is not easy for them to make progress. Beide Seiten haben ihre Idealaufstellungen eingenommen und es ist für sie nicht leicht, Fortschritte zu machen.} 19. Re2 {Inaccurate} Qc4 {OK} 20. Qxc4 Rxc4 21. Nd2 {OK} Rc7 {OK} 22. Bg2 Rec8 23. Nb3 {OK} Rxc3 {Kasparov combined this typical Sicilian exchange sacrifice with a draw offer. Kasparov verband dieses typische sizilianische Qualitätsopfer mit einem Remisangebot.} 24. bxc3 Bxe4 $5 {OK} (24... Nxe4 {was the riskier alternative: war die riskantere Alternative:} 25. Bxb6 Nxg3 26. Rxe5 (26. fxg3 $6 Bxg2 27. Kxg2 Nxb6) 26... Nxe5 27. Bxb7 Rb8 28. Na5 Ne2 29. Ra1) 25. Bc1 Bxg2 {OK} 26. Kxg2 Rxc3 27. Ba3 Ne8 $1 {Strong} (27... Nc5 $6 {is answered by ist wegen} 28. a5 {nicht empfehlenswert.}) 28. f4 {Loses game: --- --- Missed win: --- --- Mistake: --- --- Inaccurate: White=1 --- OK: White=4 Black=7 Best: --- --- Strong: --- Black=1 Brilliant: --- --- . Weighted Error Value: White=0.08/Black=0.05 Centipawn loss: w=7/b=5 Draw offer om DJ team, which is accepted by Kasparov to avopid further risk. He started very well, but could not use his chances in games 2 and 3. So man and machine seems to be of equal strength at the moment, but their main qualities are on completely different fields. This fascinating theme will remain on the agenda! Remisangebot vom Deep Junior Team, was Kasparov annimmt und damit jedes weitere Risiko vermeidet. Er hatte Deep Junior zu beginn des Matches gut im Griff, konnte seine Chancen aber nicht verwerten. Mensch und Machine scheinen zur Zeit ungefähr gleich stark zu sein, ob wohl ihre jeweiligen Stärken auf ganz anderen Feldern liegen. Dieses faszinierende Thema dürfte uns also noch weiter beschäftigen!} 1/2-1/2
Subject: Re: The Year Computer Chess Overtook The Very Best Human Chess Play. Thu Feb 09, 2023 12:36 am
Match 9 GM Kramnik vs Deep Fritz 2006
Comp Deep Fritz 10: 10/17/3/2/6/17 => Average=0.09 Kramnik Vladimir: 30/26/5/3/5/18 => Average=0.15
This is the first time in the history of the computer vs human chess matches. That a commercial computer chess engine scored a centipawn loss of less then .12. Showing the computers have reached a playing ability above all human chess players.
Last edited by mwyoung on Thu Feb 09, 2023 1:00 am; edited 2 times in total
mwyoung
Posts : 880 Join date : 2020-11-25 Location : USA
Subject: Re: The Year Computer Chess Overtook The Very Best Human Chess Play. Thu Feb 09, 2023 12:53 am
Match 10 GM Adams vs Hydra 2005.
In 2005 Hydra was the first private chess engine to score below .12 CP in a computer vs human chess match.
Home 
} (12. c4 $14) 12... Qxe7 $11 13. f3 {Inaccurate} d5 {OK} 14. Nd2 {Mistake (0.
