< Earlier Kibitzing · PAGE 2 OF 2 ·
|Oct-08-05|| ||Calli: Gypsy> Yes we missed putting in Ra8+
38...Rb7 39.Kg2 Nc5 40.Ra8+ Ke7 41.Kf2 f5 42.Ke3 Ne6 is ok
|Oct-08-05|| ||Gypsy: < Calli: ... Yes we missed putting in Ra8+ > That explains it. <38...Rb7 39.Kg2 Nc5 40.Ra8+ Ke7> In this variation, White king needs to wait his turn. Perhaps <41.Rc8> with the threat of Bb5-c6-d5 and Rc8-c6.|
|Jun-26-10|| ||goodevans: "It was night. I went home and put my old house clothes on and set the chessmen out and mixed a drink and played over another Capablanca. It went fifty-nine moves. Beautiful, cold, remorseless chess, almost creepy in its silent implacability."|
<Raymond Chandler, The High Window>
This is one of two games that seems to fit the bill, the other being Capablanca vs J Corzo, 1901.
Comparing the two I would reckon this to be the less likely candidate, but does anyone know for sure?
|Jun-26-10|| ||Petrosianic: Assuming he was talking about any specific game at all, and didn't just pull the number 59 out of a hat. The description of play is completely generic.|
|Jun-26-10|| ||Pawn and Two: See Edward Winter's C.N. 5611, at http://chesshistory.com/winter/wint..., for his review of four Capablanca games that would meet the description given in Raymond Chandler's, "The High Window".|
|Jun-26-10|| ||sleepyirv: When I read "The High Window," I assumed it was this game from searching the database, it being a "serious" game more worthy of study than a sim.|
I was hoping this would come up during the Christmas contest, but no dice. Maybe Murray Chandler should get a GoTD involving Raymond Chandler and chess-playing detective?
|Jun-26-10|| ||Oceanlake: Why not 12...Rac8?|
|Jun-26-10|| ||Pawn and Two: At move 26, Capablanca probably considered both 26.Qc6 and 26.Qc7. |
click for larger view
After 25...b6, Fritz indicates the natural looking 26.Qc7 would give White a small advantage in the following variation: (.25) (23 ply) 26.Qc7 Qxc7 27.Qxc7 a5, (.23) (24 ply) 28.Bd5 Rd8 29.Bc4, (.42) (24 ply) 29...Bh5 30.Rb7 Bg4 31.Kf1 Rb8 32.Rc7, or (.42) (24 ply) 29...Kf8 30.Nc2 Bh5 Ne3 Nc5 32.Nf5.
However, after 26.Qc7, Black could play 26...Qd4!, which leads to an equal position after 27.Rc4 Qa1 28.Rc1, or 27.Rc4 Qa1 28.Rxg4 Qxe1+ 29.Bf1 Nc5 30.Qxa7 Nxd3, or 27.Bc6 Bh3 28.Bxd7 Bxd7 29.Qxa7 Be6 30.Qb7 Bd5, or 27.Qxa7 Qb2 28.Rc4 Bf5 29.Be4 Bh3 30.Rc2 Qa1 31.Re2 Rc8 32.Bc6.
Also, after 26.Qc7 Qd4!, (.26) (23 ply) 27.h3 Bf5 28.Nc2 Qb2 29.Ne3 Be6 30.Rc2 Qb1+, the position is again equal.
Perhaps Capablanca noticed some of the lines leading to equality after 26.Qc7 Qd4!, and decided to play 26.Qc6. However, 26.Qc6 actually gave Black a very small edge: (-.32) (20 ply) 26...Qc6, (-.24) (26 ply) 27.Rxc6 Nc5 28.Rc7 a5 29.Rc6 Rb8 30.Rd6 Kf8, or (-23) (26) 27.Bxc6 Rc8 28.Rc4 Nf6 29.Bb5 Ra8 30.Ra4 a5.
In the game continuation 26.Qc6 Qxc6 27.Rxc6, Horowitz missed the best move 27...Nc5, and after 27...Rd8, the position was equal.
|Jun-26-10|| ||Pawn and Two: Fritz confirms that 38...Kc8? was the losing move: (1.40) (25 ply) 38...Kc8?? 39.Ra8+ Kb7 40.Rg8.|
Black's best continuation, leading to a likely draw and a near equal position was: (.40) (25 ply) 38...f6! 39.Kf2 Kc8! 40.Ra8+ Kb7 41.Re8 Nb8 42.Ke3 Nc6 43.f4 exf4+ 44.Kxf4, (.30) (20 ply) 44...Nb4 45.Re2 Nc6.
The move 38...Rb7, would give White some advantage, (.46) 38...Rb7 39.Ra8+ Ke7 40.Kf2 Nc5 41.Ke3. Additional analysis is needed to determine Black's drawing chances after 38...Rb7.
|Jun-26-10|| ||Pawn and Two: <Oceanlake: Why not 12...Rac8?>|
Black had many good alternatives at move 12. A strong chess program will be able to confirm this.
At 20 ply, my Fritz 12 indicated the following: (-.08) 12...Bg4; (-.05) 12...Be6; (.03) 12...Rfc8; (.07) 12...Rac8; (.12) 12...Rfd8; or (.31) 12...Rad8.
The losing move in this game was not made until 38...Kc8??.
|Jun-27-10|| ||beatgiant: <Pawn and Two>
Thanks for this analysis.
Ordinary humans like me looking at it briefly might reject 38...f6 because of 39. Bxd7 Rxd7 40. Rxb6 Rd1+ 41. Kf2 Rd2+ 42. Ke3 Rxa2 43. Rb7, with an apparently large advantage (Black's king cut off, White's king has good activity, White already has a passed pawn). What did Fritz find in this case?
|Jun-27-10|| ||Pawn and Two: <beatgiant> After 38...f6 39.Bxd7 Rxd7 40.Rxb6, Fritz indicates the position is equal: (.00) (24 ply) 40...Rd1+ 41.Kg2 Rd2+ 42.Kh3 Rxa2 43.Rb8+ Ke7 44.Rb7+ Kf8 45.c5 Rc2.|
|Jun-27-10|| ||Pawn and Two: At move 36, the position was nearly equal, with Fritz indicating only a slight edge for White: (.18) (21 ply) 36.Nc4 Rc7 37.Ra6.|
At Black's 37th move Fritz preferred: (.18) (21 ply) 37...Bd5 38.Bxd7 Bxc4 39.dxc4 Rxd7 40.Rxb6 Rd1+, (.19) (24 ply) 41.Kf2 Rd2+ 42.Ke3 Rxa2 43.Rb7 Rxh2 44.Rxf7 Rb2 45.Rxg7 Rxb3+ 46.Ke4, (.12) (24 ply) 46...Rc3, with a near equal position.
After the game continuation 37...Bxc4 38.dxc4, 38...f6! was Black's best continuation, with good drawing chances.
Capablanca allowed Horowitz no additional chances after 38...Kc8??.
|Jun-28-10|| ||beatgiant: <Pawn and Two>
Maybe Fritz has it worked out to a draw, but Black still looks under torture after 38...f6 39. Bxd7 Rxd7 40. Rxb6 Rd1+ 41. Kg2 Rd2+ 42. Kh3 Rxa2 43. Rb8+ Ke7 44. Rb7+ Kf8 <45. Kg4>, and now 45...Rxh2 46. Kf5. Even if a draw is possible, I think Black would have to be very careful.
|Jun-28-10|| ||Pawn and Two: <beatgiant> That was an interesting find! As you have noted, in the line 38...f6 39.Bxd7 Rxd7 40.Rxb6 Rd1+ 41.Kg2 Rd2+ 43.Kh3 Rxa2 43.Rb8+ Ke7 44.Rb7+ Kf8 45.Kg4 Rxh2 46.Kf5, Black has to play very carefully to hold the draw.|
After 46.Kf5, an additional 26 ply search by Fritz indicates an equal position after 46...Rc2. Other moves give White an advantage. Your chess program probably also confirms this analysis.
|Jun-28-10|| ||Pawn and Two: <beatgiant> In your last variation, one interesting way the game could conclude in a draw after 46.Kf5 Rc2, is by 47.Ke6 Kg8 48.Rb5 a4 49.Rb8+ Kh7 50.Kf7 axb3 51.Rg8 Rxc4 52.Rxg7+.|
|Jun-29-10|| ||beatgiant: <Pawn and Two>
Thanks again. The critical line is the race where White runs the c pawn and Black's rook marauds the kingside. Probably White wins a rook but loses all kingside pawns, followed by a fight between a rook and four connected passed pawns.
It might start with 38...f6 39.Bxd7 Rxd7 40.Rxb6 Rd1+ 41.Kg2 Rd2+ 43.Kh3 Rxa2 43.Rb8+ Ke7 44.Rb7+ Kf8 45.Kg4 Rxh2 46.Kf5 Rc2 47. Ke6 Kg8 <48. Kd5> Rc3 49. c5 Rxf3 50. c6.
click for larger view
If Fritz on deep analysis finds a draw, it probably is drawn, but White was winning in the few lines I briefly looked at by hand.
|Jun-29-10|| ||beatgiant: Maybe Black can improve on my above line with <48...Rd2+> so if 49. Ke6 Rc2 repeats, or 49. Kc6 blocks his own c-pawn, so Black gains a move and continues 49...Rd3 etc. Even then, it still looks to me like White is well ahead in the ensuing race.|
|Jun-29-10|| ||Pawn and Two: <beatgiant> It should be noted in your variation that in addition to 44...Kf8, Black can keep equality with (.00) (24 ply) 44...Ke6 45.Rxg7 Rb2 46.Ra7 Rxb3, or (.00) (24 ply) 44...Kd6 45.Rxg7 Rb2 46.Ra7 Rxb3.|
Your suggested continuation: 44...Kf8 45.Kg4 Rxh2 46.Kf5 Rc2 47.Ke6 Kg8 48.Kd5 Rc3? 49.c5 Rxf3?, is as you have noted, very inferior for Black.
Fritz indicates the correct continuation after 48.Kd5, is 48...Rd2+ 49.Ke6 Rc2, with an equal position.
Your latest suggestion: 48.Kd5 Rd2+ 49.Ke6 Rd3?, also gives Black an inferior position. Instead of 49...Rd3?, Black should play 49...Rc2, with an equal position.
|Jun-29-10|| ||beatgiant: <Pawn and Two>
My suggestion was 48. Kd5 Rd2+ <49. Kc6>. Then I don't see an alternative to 49...Rd3 as 49...Rc2 50. c5 looks similar to the other lines above where Black is losing.
As for the earlier points, I had been assuming Fritz thought 44...Kf8 was best. Otherwise, there's no reason for White to release Black's king from the back rank, and 43. Rb8+ in the lines above may not be White's best.
|Jun-29-10|| ||Pawn and Two: <beatgiant> At move 43, Fritz indicates 43.Rb8+ is White's best choice, with an equal evaluation (.00) (24 ply). Fritz's second choice is: (-.11) (24 ply) 43.Rb7 Rb2 44.Rxg7 Rxb3.|
For a complete evaluation of the position, considerable additional analysis would be required for the move 43.Rb7, and for the moves 44...Ke6 and 44...Kd6, as alternatives to 44...Kf8.
After your suggestion 48.Kd5 Rd2+ 49.Kc6, Fritz indicates:(.00) (26 ply) 49...Rd3 50.c5 f5. Further analysis indicates a drawn endgame of rook vs pawns.
What program do you use for your analysis? Your variations appear to be at least as good as Fritz's.
|Jul-15-10|| ||beatgiant: <Pawn and Two>
<What program do you use for analysis?>
I analyze by hand, sometimes checking with gnuchess. Judging by the posts, my strength is probably one level below some pro kibitzers here like <tamar> or <Gypsy>.
The engines seem to have improved substantially since I first started posting on this forum a few years back. It's becoming almost impossible for me to rebut any engine-backed evaluation claims, even very unintuitive ones.
Maybe it's time for me to upgrade to a recent Windows freeware...or maybe it's time to retire to the <Beer> page to indulge in a pastime where humans still reign supreme.
|Jul-15-10|| ||keypusher: <beatgiant>
I have very much appreciated your analysis over the years, and I hope you keep it up, with or without an engine.
|Jul-18-10|| ||beatgiant: <keypusher> Thanks. We now return to the quixotic quest for truth.|
After 38...f6 39. Bxd7 Rxd7 40. Rxb6 Rd1+ 41. Kg2 Rd2+ 42. Kh3 Rxa2 43. Rb8+ Ke7 44. Rb7+ <Kd6>, I did not find any winning chances for White. For example, 45. g4 g6 46. Kg3 f5 etc.
That conclusion renders moot the whole analysis after <44...Kf8>, since Black has no need to risk it. For the record, I wasn't able to figure out how Black can save himself after 44...Kf8 45. Kg4 Rxh2 46. Kf5 Rc2 47. Ke6 Kg8 48. Kd5 Rd2+ 49. Kc6 Rd3 50. c5 f5 <51. Kc7>.
I'll post later about the other open questions later if time permits.
|Oct-13-10|| ||Pawn and Two: <beatgiant> Our analysis indicates White has no winning chances in the line, 38...f6 39.Bxd7 Rxd7 40.Rxb6 Rd1+ 41.Kg2 Rd2+ 43.Kh3 Rxa2 43.Rb8+ Ke7 44.Rb7+, after either 44...Ke6, or 44...Kd6.|
In this line, Fritz also indicates that 44...Kf8 provides an equal position.
In your note of 07/18/10, <For the record, I wasn't able to figure out how Black can save himself after 44...Kf8 45.Kg4 Rxh2 46.Kf5 Rc2 47.Ke6 Kg8 48.Kd5 Rd2+ 49.Kc6 Rd3 50.c5 f5 51.Kc7.>
What does your analysis show after 51.Kc7? Analysis by Fritz suggests that Black has at least three moves providing equality: (.00) (26 ply) 51...e4 52.fxe4 fxe4 53.c6 e3 54.Rb5 Kf7 55.Rxa5, (.00) (25 ply) 55...Rxb3.
Another choice for Black is: (.00) (26 ply) 51...Rc3 52.c6 e4 53.fxe4 fxe4, (.00) (25 ply) 54.Kd6 e3.
A third choice for Black is: (.00) (26 ply) 51...f4 52.gxf4 exf4 53.c6 Rxf3 54.Rb5 Kf7 55.Kb7 Rc3 56.c7 Ke6 57.c8Q Rxc8 58.Kxc8 g5 59.Rxa5 g4, (.00) (25 ply) 60.b4 f3 61.Ra1 g3 62.b5 f2.
< Earlier Kibitzing · PAGE 2 OF 2 ·