|
< Earlier Kibitzing · PAGE 4 OF 4 ·
Later Kibitzing> |
Oct-28-09
 | | Jimfromprovidence: The analysis by <Random Visitor> shows that if 37... Rf6, then 38...Rxf5 is the best way to draw. But is it the only way to draw after 37...Rf6?
click for larger view In this position above, derived from 37...Rf6, white can penetrate and likely win with Ne7+, followed by Nc8. It's diffficult to say, though, if getting to such a position can be achieved by force. |
|
| Oct-28-09 | | YouRang: <turbo231: We are told "black to move" but black loses,does not even draw. Should one be told to move the color that loses to move first? What are we suppose to do help black lose!?> Sometimes puzzles involve moves that *should* have been played, but weren't, and that is the case today. Black wouldn't have lost had he found the better move, which is shown in the annotation for move 37: <37...Rxf5! => |
|
Oct-28-09
 | | Pawn and Two: Establishing a fortress by 37...Rxf5, and thereby securing the draw, seems clearly the best course for Black. If Black tries 37...Nf6, as noted by <David2009>, an unclear position, with some chances for White, could result after 38.Ne7 Rf4 39.Nc8. It is better for Black to take the draw with 37...Rxf5. If after 37...Nf6, White tries 38.h5, Black again should go for the draw with 38...Rxf5 39.exf5 Nf6 40.Kg2 Nxh5. If after 37...Nf6 38.h5, Black plays 38...Rf8?, he will have a lost position: 
click for larger viewHere is Fritz's analysis of the position after 38...Rf8?: White can now win with 39.Rf1!: (2.73) (30 ply) 39...Rf6 40.b6 a6 41.Ne7 Rxf1+ 42.Kxf1 Nf6 43.Ke2 Nxh5 44.Nc8 Nf4+ 45.Kf3 Nxg6 46.Nxd6 (3.37) (27 ply) 46...Nf8 47.Nf7+ Kg8 48.Nxe5 (4.70) (27 ply) 48...a5 49.a4 h5 50.d6 Ne6 51.Nc6 Nf8 52.Nxa5, and White is clearly winning. |
|
| Oct-28-09 | | AnalyzeThis: <RandomVisitor: 38.h5 Rxf5 39.exf5 Nf6 40.Kf2 Nxh5 41.Rh1 Nf6 42.Kf3 Kg8 43.a4 Kf8 44.a5 Ke7 45.Rh4 Kd7 46.Rh3 Ke7 00:31:08 100710kN > I guess the question is whather black is obligated to play 40...Nxh5. If he doesn't, isn't this just the same blockade as the main line? |
|
| Oct-28-09 | | muralman: Aha!! I first looked to my knight to attack, but it became obvious the white knight would have the drop on me in a counter attack. The only other piece worth considering was my rook. The only aggressive move was to fall on it's sword. Any other move was simply defensive, and that obviously led to a long tangle, the kind of puzzle I don't care for. So, looking at the sacrifice, I knew that would leave me with a knight going against a rook. An overview of the board gave me the notion I could lock out the other rook for a draw. I looked how I would handle the pawns to bring it to a draw, and settled for that. To my surprise, a note read the same, making me correct!! Ta da!!!!! I think. :) |
|
Oct-28-09
 | | chrisowen: I thought it was Rxf5 as its the only obvious hack. Why sack when you can shuffle though? The rook lifts only stalls the inevitable, black should have freestyled it kicking the minor piece off the board. Instead he tricks himself as Ng4 is not good enough to keep it up. |
|
| Oct-28-09 | | BOSTER: 37...Rxf5
38.exf5 Nf6 1/2-1/2
After move 38.Nf6 you need to realize that White can not break into black camp.
White can not do anything to break through. |
|
| Oct-28-09 | | A Karpov Fan: got it |
|
| Oct-28-09 | | MaczynskiPratten: I got this quicker than I would otherwise have done by remembering some of the comments on yesterday's puzzle about "when is a game drawn" showing blocked positions. Thanks kibitzers! |
|
| Oct-28-09 | | dumbgai: I would have played 56...Rf7+ hoping for gxf7 stalemate. Of course Larsen wouldn't fall for it but why not joke around if you're lost anyways? |
|
| Oct-28-09 | | turbo231: I played this game twice, Rybka vs Rybka, i gave black 20\22 white 20\20. Both games black moved rd7 same as Torre. White won both games no contest, it was not even close. Again i ask why have us move the lossing color first. CG must use Crafty 19.3 to test these games. Bad choice! |
|
| Oct-28-09 | | WhiteRook48: I knew it was 37...Rxf5! it has a fortress draw |
|
| Oct-28-09 | | turbo231: Just in case anyone is confused,that's 20 moves in 22 minutes for black and 20 moves in 20 minutes for white, I spent some time on it. |
|
| Oct-28-09 | | Phoenix: <turbo> 37...Rxf5 is a draw. White cannot break through and so cannot win. |
|
| Oct-28-09 | | Phoenix: To be more specific, 37...Rxf5 38.exf5 Nf6 (avoiding a possible f6 from White) and then Black just moves his king endlessly. If White goes for a3-a4-a5-a6, Black will play ...b6. If White instead tries b5-b6, Black goes ...a6. |
|
| Oct-28-09 | | turbo231: I take it all back Rykba made the wrong first move! No wonder Nakamura was able to beat Rybka. Rybka anylized that first move for over 4 minutes. I played the game again ME vs Rybka using rxf5 and i got a draw! I just kept moving my King back and forth and Rybka was blocked out! |
|
Oct-28-09
 | | ajk68: Why does black play 45...Ng4,?
How does white make progress if he doesn't play this? I don't have a chess engine in front of me. |
|
| Oct-28-09 | | Manic: <ajk68> Someone said earlier that 45...anything else 46.Rg5 with 47.Ng3 winning the pawn |
|
| Oct-28-09 | | ComboKal: <Once> Your line does force a draw with the blockade, but I disagree that black should play for a draw at this point in the game. White has four weak pawns to defend on the 4th rank. , and black plays <40. ...Nf6> anyway, while saving the rook. It was the unfortunate <45. ...Ng4 46. Rxg4> oversight that lost the game. |
|
| Oct-28-09 | | ComboKal: My oversight! (A common theme in my games!) Black plays <45. ...Ng4> to try to save the h pawn, which would be lost after Rg5 and Ng3. Maybe playing for the draw was the best idea after all. My apologies <Once>. |
|
| Oct-28-09 | | psmith: <turbo231>
This is the kind of position that even very strong programs like Rybka can't quite understand. The kind of position where seeing the consequences (which are in some sense obvious to humans) requires what is for computers impossibly long calculation.But these kinds of positions are relatively rare, so strong programs will beat strong humans most of the time these days. |
|
| Oct-28-09 | | RandomVisitor: After 21.Rxe3 black might be able to equalize with the clever move 21...h5: 1: Bent Larsen - Eugenio Torre, Brussels 1987
click for larger viewAnalysis by Rybka 3 :
<[+0.11] d=25 21...h5> 22.Rb3 g6 23.Ne3 Rab8 24.Ra1 Kg8 25.a4 Qa5 26.g4 Qd8 27.gxh5 Qxh4 28.Qg4 Qxg4+ 29.Nxg4 Kh8 30.hxg6 fxg6 31.Kg2 Ng5 32.Re1 Kg7 33.Rg3 03:16:22 811751kN |
|
| Oct-28-09 | | CHESSTTCAMPS: Material is even in this closed endgame position, with white having a space and mobility advantage. I spent a while examining 37... Rf6, with the idea of 38.h5 Rxf5 39.exf5 Nf6, winning the h5 pawn at no risk, because white has no entry points with his rook. In this line, white can try to win with 38.Ne7!??
Protects the g6 pawn, puts the black king in a straitjacket (creating back rank mate problems), and the knight is untouchable. Now if A) 38... Rf4 39.Rf1 Rxf1?+ 40.Kxf1 Nf6 41.Nc8 Nxe4 42.Ke2! Now black's problems becomes apparent - white's knight is behind in the race to capture pawns, but it doesn't matter - black's d-pawn must fall, white's protected d-pawn stands tall, black's fixed pawns are vulnerable, and the white king position and other white pawns are superior. Kg8 43.Ke3 Ng3 44.Nxd6 Kf8 45.Kf2! Nh5 (Ke7? 46.Nc8+! Kd7 47.Kxb3 Kxc8 h5 is a won K&P ending) 46.Nxb7 Nf4 47.Nxc5 Nxg6 48.a4! Ke7 49.a5 Kd6 50.b6! axb6 51.a6 and white promotes the a-pawn! A.1) 39...Rxe4 (Rxh4)?? Rf8#
However, there's a problem:
A.2) 39...Rg4+! 40.Kh2 Nf6! (not Rxh4+? 41.Kg3 Rf4 42.Rxf4 exf4 43.Kxf4 is winning) and black seems to be better. Another trap try for white is
A.3) 39.Kg2 Rxh4?? 40.Rf1 Nf6 41.Rxf6! gxf6 42.g7+ Kxg7 43.Nf5+ winning. However,
A.3.1) 39... Nf6 again looks fine for black.
B) 38... a6 39.bxa6 bxa6 40.Rb1 leaves no good defense to the threat Rb8. C) 38... b6 39.a4! Rf4 (otherwise a5 wins anyway) 40.a5! bxa5 41.Ra1 Rxe4 (no neewd to die hungry) 42.Rxa5 Rxc4 43.Rxa7 Rg4+ 44.Kh2 Rxh4+ (Rxg6 45.Nxg6+ is also hopeless) 45.Kg3 Rf4 46.Ra8 Rf8 47.b6 wins. In view of A.2 and A.3.1, I have to conclude that after 37... Rf6?!, the safest response is 38.h5 Rxf5 39.exf5 Nf6 40.Kg2 Nxh5, after which black has no losing chances, but white can protect the f-pawn with his king, so I don't see winning chances for black either. Also, from the diagram black can avoid any risks by playing an 37....Rxf5 38.exf5 Nf6 and white can not force any entry points for K or R, therefore the position should be drawn. Time to see what I've missed... |
|
| Oct-28-09 | | TheBish: Larsen vs E Torre, 1987 Black to play (37...?) "Medium/Easy"
White has an advantage after 37...Rf6 38. Ne7!, so, being "draw week", I'm going to opt for 37...Rxf5! 38. exf5 Nf6, when White has no way to break through, as Black will manage to blockade the queenside pawns if White advances. |
|
| Oct-29-09 | | Summerfruit: Material is even.
Black can't win, and the easiest way to draw is an exhange sacfifice: 39...Rxf5 40.exf5 Nf6
White can't make any progress in the center or on the kingside. The queenside can easily be blocked by black. |
|
 |
|
|
< Earlier Kibitzing · PAGE 4 OF 4 ·
Later Kibitzing> |
Kibitzing
sac: 46.Rxg4