Jan-14-05
 | | beatgiant: I suspect Black can still draw by move 40. It would be a book draw (4 pawns vs. 3 on the same side) except for the presense of the a-pawn and c-pawn, and Black can play to engineer a trade of these pawns by advancing the a-pawn. But Black must not allow the White king to attack his pawns as in the game. Instead of 40...Re1 41. Kf5, I suggest 40...Rf1. For example, 40...Rf1 41. f3 a4 42. Kg4 Rg1 43. f4 g6 44. Kf3 Rc1 45. c4 a3 and White is now forced to trade pawns by 46. Ra2 Rxc4 47. Rxa3, with a book draw. |
|
| Jan-14-05 | | drukenknight: MOve 40? Bg. dont sell yourself short. 44...Rxg3 seems obvious, but I wonder if we cant save him later than that....hmm. |
|
Jan-14-05
| | beatgiant: <44...Rxg3 seems obvious>
It's an obvious try, but it's not clear that it draws after 44...Rxg3 45. Ra7 a3 46. e6 fxe6+ 47. Kxe6 Re3+ 48. Kd5 Rd3+ 49. Kc6, and White runs the c-pawn. |
|
| Jan-14-05 | | euripides: There seems to be a draw by repetition unclaimed on move 55 -bafflingly, since the position is by then so clearly better for White. I think white allows, and Black avoids, 44..Rxg3 because Black decides he needs to keep the white king from the d file. 43 ..Re1 would not have achieved this bjective because of 44 Ra7. But I'm not sure whether this is right. R+4 vc R+3 on one side is not always drawn and can be very difficult: Kasparov lost this ending once. The ideal defensive set-up is meant to be pawns on f5, g6 and h5, but Black cannot achieve this here. So the question would be whether with pawns on f7, g6 and h5 black could stop white making progress e.g. if White puts his King on h4 (e.g. 44 Kh4 in beat's line) and the rook on the 3rd or seventh rank. |
|
| Jan-14-05 | | euripides: <beat> my last post was written before yours appeared - we make the same point about the d file. |
|
Jan-14-05
| | beatgiant: <euripides>
<There seems to be a draw by repetition unclaimed on move 55 -bafflingly>This kind of problem has been discussed in other old games. It seems the old repetition rule was different - they counted recurrences of moves instead of positions. |
|
| Jan-14-05 | | euripides: Looking closer at the ending: the R+4P vs. R+3P ending that could have arisen here is very like the ending Kasparov lost against Piket, but it is in principle drawable, according to Muller and Lamprecht (example 6.106, internet rapid game, not in the chessgames database). The trick, which eluded Kasparov, is very precise handling of the defensive rook if White apporaches the pawns from the front. |
|
Jan-14-05
| | beatgiant: <euripides>
Thanks for that insight. I also haven't found a win for White yet with 40...Rf1 41. Ke3 Re1+ 42. Kd4 Rf1 43. f4 Rf3, and Black counterattacks the kingside. |
|
| Dec-10-05 | | aw1988: "All rook endings are drawn" - Tarrasch |
|
Dec-27-07
 | | Honza Cervenka: <aw1988: "All rook endings are drawn" - Tarrasch> Tarrasch never said or wrote anything like that. It's Tartakower's aphorism. |
|
| Dec-27-07 | | RookFile: You should explain this to Dennis Monokroussos, then. http://chessmind.powerblogs.com/pos... |
|
Dec-29-07
| | Honza Cervenka: Yes, I know that. But Dennis Monokroussos just repeats here a mistake made before him by many other authors, included Soltis or Dvoretsky. But still it is the mistake probably caused by confusion of two somewhat similar names of these two great players and the fact that both made some pronouncements related to Rook endings (maybe plus bad habit of chess writers sometimes just to copy down writings of their colleagues and predecessors without any serious research and control). In fact, Tarrasch formulated this rule for Rook endings (and not only for them) for the first time in his book Dreihundert Schachpartien within the comment of one game: "The Rook should be placed behind the passed Pawn, either yours or your opponent's." He also immediately added to that: "Except when it is incorrect to do so.":-) |
|
| Dec-18-10 | | Dr. Siggy: You find a thorough and very instructive analysis of this R+PP vs R+PP endgame (after 37... Rxa4)
click for larger viewin Dr. Tarrasch's great classic "The Game of Chess", english transl., London 1935, pages 85-89. In the 'Praeceptor Germaniae seu Mundi' own words, it "[...] admirably demonstrates how difficult it is in a Rook ending to turn the superiority of a pawn into a win. White has, as well as an advantage in material, the far superior position, and yet right to the very end victory hangs by a thread." |
|
Aug-31-12
 | | Chessical: Swapping off Rooks on move 23 did Schlechter no favours. Instead, Tarrasch's pieces spring to life; <23...Bf5> would have been better. I believe that Schlechter would have drawn if he had taken the <g> pawn on move 44. 
click for larger viewThe <c> pawn cannot be supported to Queening without Black promoting his <a> pawn first with a draw. For instance: <44...Rxg3> 45. Ra7 a3 46. Ke4 (or <46. e6> fxe6+ 47.Kxe6 Re3+ 48. Kd5 Rd3+ 49. Kc6 g6 50. c5 Ke8 51. Kb6 Rb3+ 52. Kc6 =) 46... Rc3 47. Kd5 Rd3+ 48. Kc6 g6 
click for larger view<49. c5> Ke8 50. Ra8+ Ke7 51. Kb6 Rb3+ 52. Kc7 f6 53.exf6+ Kxf6 54. c6 Ke6 55. Kc8 Ke7 56. Ra7+ Kd6 57. c7 Ke7 58. Rb7 Rxb7 59. Kxb7 a2 60. c8=Q a1=Q 61. Qc5+ = |
|
| Aug-30-14 | | Dr. Siggy: According to Dr. Tarrasch, after 44. ... ♖xg3, "White's King comes up to the Queen's Bishop pawn via e4 and conducts it on to queen." I may be wrong but, after 44. ... ♖xg3 45. ♔e4 ♖c3 46. c5 a3 47. ♔d5, Black's game seems as good as lost... |
|
Jul-17-20
| | Honza Cervenka: <Dr. Siggy: According to Dr. Tarrasch, after 44. ... ♖xg3, "White's King comes up to the Queen's Bishop pawn via e4 and conducts it on to queen." I may be wrong but, after 44. ... ♖xg3 45. ♔e4 ♖c3 46. c5 a3 47. ♔d5, Black's game seems as good as lost...> Well, after 44...Rxg3 45. Ke4 Rc3 46. c5 a3 47. Kd5 Rd3+ 48. Kc6 Rf3 49. Ra7 Rxf4 50. Rxa3 Rxh4 51. Ra6 Ke7 52. Kb7 Rb4+ 53. Rb6 Rc4 54. c6 Kd8 55. e6 fxe6 56. Rb1 h5 57. Rd1+ Ke7 58. c7 g5 59. c8=Q Rxc8 60. Kxc8 h4 the position is tablebase draw. |
|
Kibitzing
