< Earlier Kibitzing · PAGE 2 OF 2 ·
|Jan-06-05|| ||Calli: 8.Qc3 is given in Skinner and Verhoeven's massive Alekhine book. The Bible! They cite 1) De Schaakwereld 1938, v3, p233-4 and 2) Chess 1938, v4, p126. Even Skinner and Verhoeven could be wrong, but its more likely that Euwe's tournament book contained the error that has propagated to the online databases.|
Have not looked at the ending in detail. In general terms ,however, black's rook has free range to harass the white pawns and king. White has to be aware of Rc2 and a Bishop move followed by a2. By contrast, White's rook must stay on the a-file. Secondly, the long range bishop both guards the pawn and the kingside and is superior to white's knight. Capablanca probably judged such an ending as drawish.
|Jan-06-05|| ||beatgiant: <Calli>:
<black's rook has free range...By contrast, White's rook must stay on the a-file> Black's rook is also somewhat constrained because he has to be aware of White's possible mating nets with Ng6 and Ra8.
<the long range bishop both guards the pawn and the kingside> But not if White gets his pawn to e5 blocking the diagonal.
I think <capanegra>'s find is very strong, and I'm tempted to look at the four pawns versus rook ending again.
|Jan-06-05|| ||Benzol: <capanegra> <Calli> <beatgiant>
The small tournament book printed by the British Chess Magazine has the White 8th move in this game as ♕c2-b3 and as <Calli> says Skinner & Verhoeven give 8.Qc3, so which one should you believe?
Sometimes I get annoyed about such things. Is there any other source that we could look at? |
|Jan-06-05|| ||acirce: chessbase.com gives 8.Qb3 and has no examples at all of 8.Qc3. Instead, 8.Qb3 was played in I A Horowitz vs Reshevsky, 1945 as well, for example|
I also find it odd that Capa wouldn't play 11.Nxe4 if the queen is on c3 - and that Alekhine would allow it. 10..Bb4 would be a bit annoying instead, for example, but several other moves would also be better. In fact 10..dxc4 seems to make little sense at all with the queen on c3. This is not proof but only the basis for a more or less good guess, so far.
|Jan-06-05|| ||Calli: But Horowitz follows Qb3 with Nc3. Capa's Nd2 makes more sense with 8.Qc3. He also does not seem too worried about Bb4 after 12.Qc3 in the game. AA wants to avoid Bg5 so he plays 12...h6, but then decides against 13...Bb4 anyway. Finally, if 8.Qb3 why not 10...Na5? |
|Jan-07-05|| ||Calli: Dutchbase at the Max Euwe site also gives Qc3 in their file. http://www.maxeuwe.nl/dutchbase/pag... |
<Benzol> Have you spotted any errors in Skinner and Verhoeven? I suspect that the Britsh reprint just copies the original tournament book.
|Jan-07-05|| ||beatgiant: You're going to laugh, but here's a quick sample of the ending of four pawns versus rook.|
41. h4 Kg6 42. g4 Kf6!? 43. Nxg7 Ke5 44. Nf5 Kd5 45. Nxh6 Kc4 46. g5 Kb3 47. h5 a2 48. Rxa2 Rxa2 49. Nf5 Kc4 50. g6 Ra7 51. h6 Kd5 52. g7 Ra8 53. h7 Bxg7 54. Nxg7 Rh8 followed by 55...Rxh7, and I believe the resulting ending should be drawn.
Obviously, there's a strong suspicion that White must have improvements over this line at various points by making more preparations before trying to cash in on the passed pawns. But White will always have to be careful to avoid Black's plan of trading his bishop for two pawns to reach a drawn ♔♘♙♙ versus ♔♖ ending, as happens in the sample line above.
|Jan-07-05|| ||tamar: I hadn't even thought of 41. h4 Kg6 42. g4 Kf6!? 43 Nxg7, giving up too early on the variation because White is gaining material, but of course it makes sense for Black to get an active King with his own pawn so close to queening. |
|Jan-07-05|| ||Calli: You have convinced me that 41.h4 was worth a shot. Perhaps Capablanca thought the 41.Nf4 was better, but didn't see the saving 48.Bxf2 in his calculations. |
BTW- the "New In Chess" database also gives 8.Qc3.
|Jan-07-05|| ||Benzol: <Calli> <<Benzol> Have you spotted any errors in Skinner and Verhoeven?>|
No I haven't yet but I seem to remember some reference that <ray keene> made to a game that was corrected here at <chessgames.com> that wasn't in or wasn't correct in Skinner & Verhoeven.
And while we're talking about 4 pawns versus a Rook in the ending, have a look at Euwe vs Capablanca, 1929
|Jan-07-05|| ||acirce: <No I haven't yet but I seem to remember some reference that <ray keene> made to a game that was corrected here at <chessgames.com> that wasn't in or wasn't correct in Skinner & Verhoeven.>|
Keres vs Alekhine, 1943
|Jan-07-05|| ||Calli: Yes, its true that Keres-Alekhine was incorrect. This however, was explanable by the fact that the Keres book was published after Skinner/Verhoeven. They certainly saw seen this game (Capa/AA) had been published both ways because the tournament book is in the bibliography. So how did they choose? |
|Jan-08-05|| ||Benzol: <acirce> Thanks mate. :)
I couldn't remember what game it was. |
|Sep-08-05|| ||LIFE Master AJ: A magnificent struggle, the ending after 48...Bxf2!; 49.Kxf2, Rb2; 50.Rxa3, (Necessary - sooner or later.) 50...RxN/c2+; 51.Kg1, Ke6; is probably nothing more than a draw.|
|May-14-08|| ||RookFile: This was a really nice game, I enjoyed playing over it.|
|Oct-17-08|| ||Whitehat1963: So, what happens if white continues with 49. Nxa3?|
|Oct-18-08|| ||CapablancaFan: <Whitehat1963: So, what happens if white continues with 49. Nxa3?> You're joking right? If Capa was stupid enough to play something like that. 49...Ra1! (pinning the knight) would come like a shot.|
|Oct-18-08|| ||Benzol: <CapablancaFan> What happans after 49.♘xa3 ♖a1 50.♔xf2?|
|Oct-18-08|| ||CapablancaFan: <Benzol> Oh, snap, I wasn't paying attention, I feel so embarrassed. :-)|
|Oct-19-08|| ||Benzol: 49.♘xa3 ♖b2 50.♘c4 is probably drawn.|
|Oct-28-08|| ||Whitehat1963: <Benzol: 49. Nxa3 Rb2 50. Nc4 is probably drawn>|
With the aid of the chesslab.com computer, I think I found a lot more life left in this position. As you say, it's probably drawn, but white definitely holds the advantage, and black would have to be very careful. I'm surprised Capa didn't press on, superb endgame master that he was:
49. Nxa3 Rb2 50. Nc4 Rc2 51. Nd6+ Ke6 52 Nf5 Bc5+ (a move I don't understand) 53. Kh3 Bg1 54. Nxh6 Rxh2+ 55. Kg4 Bc5 56. Re8+! Kd7 57. Re5 Bd6 58. Nf5 Be5 59. Kxg5 Rg2 60. g4 Rg1 61. Nd4+ Bxd4 62. Rxd4 Ke5 63. Ra4 Rg2. And white has a pair of disconnected passed pawns that might prove too difficult to promote.
It would be interesting to see what something like Rybka 3 makes of the final position.
|Oct-28-08|| ||Whitehat1963: Yes, I just tested it on Jester, and 49. Nxa3 Rb2 50. Nc4 Rc2 51. Nd6+ Ke6 52. Nf5 Bc5+ leads not to a draw, but to a win for white every time!|
Here's the sequence we played:
Can someone with Rybka please analyze 49. Nxa3 and slide the analysis forward a bit as well? I think this is a clear-cut win for white!
|Oct-28-08|| ||Pawn and Two: <Whitehat1963> After 49.Nxa3 Rb2 50.Nc4 Rc2 51.Nd6+ Ke6 52.Nf5 g4!, the position will be a draw. Fritz found this line very quickly.|
|Oct-28-08|| ||Whitehat1963: 52...g4!, huh? I'll have to take a look. Thanks.|
|Aug-24-09|| ||WhiteRook48: someone analyze 52....Be1+|
< Earlier Kibitzing · PAGE 2 OF 2 ·