|Nov-14-03|| ||lordazol: 30 Rxc6 is ??blunder.After 29th move position shows there is no winning continuation for white.Fritz must be stop white pawns so king must came to the other side to stop them.
Leaving h pawn gives a clear win to black.Instead of Rxc6 Kc1 gives a draw changes to white.
Explaining the position is better than calculating next moves.Especially at the endgame. |
|Jul-28-04|| ||chess4games: Good endgame. Fritz dose not understand Kasparovs moves. But Kasparov dose understand Fritz's moves in this game. |
|Apr-19-05|| ||pencuse: 21. ... Rh8!! incredible move, almost impossible to find on board. Kasparov made a move computer can never understand, to make the computer computation completely useless in the endgame. |
|Apr-19-05|| ||RookFile: Well, I'm not sure I understand
31. Ra6 is useless, of course. Could
white still bail out with 31. Kc1?
|Apr-19-05|| ||Minor Piece Activity: 21...Rh8! is a clever "mysterious rook move." The plan, I think (I'm not GK), was to push the h pawn to h3, and then play Ra4 with the optimal position in the endgame. Fritz stops his plan with 24. h3, but cedes g3, giving black an outpost for the knight. |
|Apr-19-05|| ||beatgiant: White plays the ending planlessly. What about counterattacking, as in 21...Rh8 <22. Re4>, so that 22...h5 23. Ra4 etc. and I doubt that Black has anything. |
|Apr-19-05|| ||RookFile: But why do you need to play ...Rh8
to push the pawn to h3. Just push
the pawn to h3 if it's such a great
No, I'm not sure we understand ....Rh8.
|Apr-19-05|| ||Minor Piece Activity: Don't be silly. If you had read what I wrote more closely, you would have seen that the idea <might have been to play Ra4>, which is why you push the pawn to h3.|
I don't claim to completely understand Rh8, but there are two possibilities if you want to analyze the move.
(1) GK, in time trouble (this was a blitz game), quickly made his move, and there was no point to it. This makes a lot of sense, since the previous move Rb8 also looks sort of weird if he is just going to follow up with Rh8 anyway. Maybe he decided that Rb4 Re4 was just a clear draw and he had more than that.
(2) The other possibility then is to assume there was a reason he made his move. So I claim that maybe the idea was to advance the pawn to h3 to play Ra4. If successful, Rg2 would also be a strong plan later.
<beatgiant> Not an expert on the endgame, how do you think 23. Ra4 Re1 sacrificing the pawn is?
|Apr-20-05|| ||Minor Piece Activity: 23... h5 as well as 22...Rd8 in your line also look interesting, beat. (Btw I meant Rh4 was the plan, oops typo.) |
|Apr-20-05|| ||Minor Piece Activity: Here's an example of the line with 23. Ra4 Re1-- 24. Rxa7 Rd3 25. Rc7 Rh3 26. Rxc6 Rxh2 with a complicated game. I think black stands better despite his pawn deficit because he looks like he should queen first-- another strength of the h5 idea. |
|Apr-20-05|| ||RookFile: I agree that ...h3, ....Rg4, and
...Rg2 is an interesting plan for black. Thanks for the idea.
Hey, anything on the rook file is
fine with me!
|Apr-20-05|| ||beatgiant: <Minor Piece Activity>
On 22. Re4 h5 23. Ra4 Re8 24. Rxa7 Re3, it becomes a racing situation and White should simply push the a-pawn rather than going after the c-pawn: 25. a4 Rh3 26. a5 Rxh2 27. a6, and White is ahead in the race. |
|Apr-20-05|| ||Minor Piece Activity: Well beat, I don't think that GK can win this if black plays perfectly, but I still suspect that 21...Rh8! offers more fighting chances than other alternatives. Would would you rather play? 21- Rb4 Re4 would offer the machine no challenge even in 1992. 21....Rd8 isn't effective after 22. Re5 Rd4 23. Ne2 =|
On 24. Rxa7, 24..Re1+ Kd2 25. Rh1 Ne4 26. Rxh2+ Kd3.
Maybe black can play ambitiously at this point, trying for 27. Rh1 Ng5 28. Nh6 a4 29. h4 and I think black is at least equal, maybe even better.
But the point though is that you can see that this line is a complicated race, much more complicated than any other line, and so the logical choice if you want to present problems to your opponent even if it's even theoretically.
|Apr-20-05|| ||beatgiant: <Minor Piece Activity>
Completely agreed. I have only criticized White's play, not Blacks's.|
On 22. Re4 h5 23. Ra4 Re8 24. Rxa7, your <24...Re1+> is a big improvement over the previously suggested <24...Re3>, and it leads to a close race.
If White is satisfied with a draw, at first glance 22. Re4 h5 23. Ra4 Re8 <24. Re4> looks pretty similar to 21...Rb4 22. Re4 which you said offered no challenge.
|Apr-21-05|| ||beatgiant: On move 31, White again doesn't realize this is a racing position and spends too much time going after the c-pawn. |
Even though Black has the advantage with two connected passers, can White still hold even now by pushing his own pawns?
It looks very tricky, but here's an initial example:
31. b4 g5 32. a4 g4 33. Rc4 g3 34. Rg4+ Kf6 35. b5 Ke5 36. a5 f5 38. Rg8 f4 39. b6 Rh7 40. c4 Kd4 41. Rg4.
I haven't found a convincing Black win yet.
|Apr-28-05|| ||Kingdumb: I'm surprised some of the stronger players did not see GK's genius in this game. He had this calculated right down to the ending you see in front of you. At this point it is a race that white cannot win and black has an easy victory with a rook and queen up.|
I have noticed in my own play against computers that they are very materialistic and don't take positional play into account as much. GK clearly took advantage of that fact in this game and exploited it magnificently.
|Apr-28-05|| ||RookFile: Kasparov played this ending quite well. Really a clever approach.|
|Apr-28-05|| ||Boomie: 20...♖b8 struck me as a little odd. I looked at ♖e8 instead. Although I distrust long Fritz lines, they are more valid in endgames.|
20... ♖e8 21. ♖e2 ♖xe2 22. ♘xe2 ♔h6 23. ♔d2 ♔h5 24. b4 ♔g4 25. c4
(25. ♔d3 h5 26. c4 h4 27. c5 ♘e7)
25... h5 26. c5 h4 27. b5 cxb5 28. c6 ♘d6 29. c7 ♔h3 30. ♔e3 ♔xh2 31. ♔f3 ♘c8 32. ♘d4 ♔h3 33. ♘xb5 f6 34. ♘d4 g5
35. fxg5 fxg5 36. ♘e6 g4+ 37. ♔f4 g3 38. ♘g5+ ♔g2 39. ♔g4 ♔h1 40. ♔xh4 g2 41. ♘f3 g1=♕ 42. ♘xg1 ♔xg1 43. ♔g4 ♔f2 44. ♔f5 ♔e3 45. ♔e5 a5 46. a4 ♘a7 47. ♔d5 ♔f4 48. ♔c5 ♔e5 49. ♔b6 ♘c8+ 50. ♔xa5 ♔d6 =
Amazingly GK knew this and tried to improve.
|Jun-23-05|| ||delterp: Mistake.
18... ♗xg2 is not check.
|Oct-07-13|| ||kontoleon: lol human win machine in 5 minute game!|
|Apr-30-17|| ||clement41: Wow, finding 21...Rh8!? (anti-engine play) OTB, and what's more in blitz is outstanding.
I'm wondering, is 10 g4 to restrict the Nh6, or 10 h4 to start a king-side attack (priyome against Nh6&g6 by black), any good?|
|May-01-17|| ||beatgiant: <clement41>
On either 10. g4 or 10. h4, what if Black replies 10...d6, opening a path for Bg4?