|
< Earlier Kibitzing · PAGE 3 OF 3 ·
Later Kibitzing> |
| Aug-28-05 | | Koster: I also think 54. Rxb2 wins. I could well be overlooking something but the key problem for black is the that the unavoidable Rxg3 leaves his rook badly placed. It needs to be on the queenside to keep the white king from advancing after the white rook drives the black king back. |
|
Aug-28-05
 | | Sneaky: <sharpnova: . i only considered rxb2... i checked it with fritz.. the evaluation is steadis a clearcut win and probably faster than the text move. period.> 54.Rxb2? throws the win away. Period. I've been playing against my computer and I find it fairly easy to draw with Black against 54.Rxb2. One such game went like this SigmaChess6 vs Sneaky
54. Rxb2<?> Ra3 55. Rg2 Ra5 56. Rc2 <56.g4 fxg4+ 57.Kxg4 => Ra3 57. Kxf5 Rxg3 58. Rc7+ Kf8 59. Kf6 Kg8 60.f5 Ra3 <and now Black draws with checks from the side. The game continued:> 61. Rc8+ Kh7 62. Re8 Ra6+ 63. Kf7 Ra7+ 64. Re7 Ra6 65. Kf8+ Kh8 66. Kf7 Kh7 67. f6 Ra1 68. Rd7 Rb1 69. Ke7 Rb8 70. f7 Kg7 <and now we reach a classic drawn book position.> Where does your Fritz say that White went wrong above? |
|
| Aug-28-05 | | 50 Quatloos Newcomer: Why debate? Run this bounder through Fritz or Shredder. Isnt there a famous name for the end of this position, where the side with the pawn puts his rook on the 5th rank while boxing out the enemy king? Its the only way to avoid the rook checks and still queen the pawn...? |
|
| Aug-28-05 | | aginis: <50 Q newcomer> thats when the white king is in front of the pawn if the black king is in front of the pawn its a draw.
if you need convincing try a few positions with the nalimov tablebase
http://www.lokasoft.com/tbweb.htm |
|
Aug-28-05
| | Sneaky: <Fritz8: 54. Rxb2 Ra3 55. Rb7+ Ke8> instead of ...Ke8, how about ...Ke6 instead? What can White do? 56.g4? fxg4 just leads to basic draws as far as I can see. 56.Rb6+ Kf7 57.Rb7+ Ke6 just repeats position, so perhaps 57.Kh4 (this looks like a move Fritz would approve of going by the line above) and now ...Rc3! and I'll keep eyeing that g pawn until you either push it or go after my f pawn with your King. The Black King is in the 'safety zone' in front of the White pawn, so = |
|
| Aug-28-05 | | sfm: <billcrutcher: Fritz 8: 54. Rxb2 Ra3 55. Rb7+ Ke8> Huh?? What's the king doing down there? It is a losing move! 55.-,Ke6 is the move to be played. Black is now shooting at the g-pawn and White can no longer cover it by playing Rg2. So the white king must cover it from h4 and black covers his f-pawn with his king and nobody gets further. But with the black king at last rank he can not get back to cover the f-pawn. So he must take his rook away from the third rank which is fatal as your variation shows. BTW - all these discussions show how good a puzzle this really is. :) |
|
Aug-28-05
| | Sneaky: <sfm> Thank you, a voice of sanity in the jungle! <Why debate? Run this bounder through Fritz or Shredder.> We did, but they gave the wrong answer! Now the debate is between people who think that Fritz is never wrong and people who actually want to see the win demonstrated. <billcrutcher> Why did you remove your message? It's OK if your Fritz was sort of patzerish, that's not your fault. |
|
| Aug-28-05 | | aginis: <sharponova> you talk a lot but you never offer anything intelligent.
yesterday you claimed to have solved an intricate position in <10 nanoseconds> , and the fact that everyone else was analysing variations in great depth with little success didn't seem to bother you.
today you've been told flat out you're wrong and you still insist that <Rxb2 wins...stupid puzzle> and i don't buy your arrogant <sorry> either. I challenge you to offer an ATTEMPT at winning after 54.Rxb2 Ra3.
I can hold this draw against your computer using nalimov to hold the ♖♙v♖ endgame after exchanging pawns. |
|
| Aug-28-05 | | billcrutcher: I don't agree with <sharpnova> that this is a stupid puzzle, but to answer <Sneaky>, here is what my Fritz recommends: 54. Rxb2 Ra3 55. Rb7+
At this point, the top four responses Fritz considers for black are 55. ... Kg8
55. ... Ke8
55. ... Kf8
55. ... Ke6
To take the last option first, this transposes directly into the line with 54. Rb7+ first. Fritz gives the best evaluation to 55. ... Kg8, with the following continuation: 55...Kg8 56.Kh4 Ra5 57.Re7 Ra1 58.Re3 Kf7 59.Kg5 Ra6 60.Kxf5 and White should be able to advance the two connected, passed pawns. The other two options seem no better:
55...Kf8 56.Kf6 Kg8 57.Rg7+ Kf8 58.Rg5 Ra6+ 59.Kxf5 and again, the two pawns should win. 55...Ke8 56. Kh4 Ra5 57. Rb3 Kf7 58. Kg5 Rd5 59. Re3 Rc5 60. Re5 Rc6 61. Rxf5+ and again, the two pawns should win. So, <Sneaky>, can you improve on Fritz's handling of Black's options in response to 55. Rb7+? |
|
| Aug-28-05 | | billcrutcher: <sfm> I deleted my previous message because I wanted to give Fritz's analysis to more than just 55. ... Ke8. And now, I have made a mistake in that post, regarding the 55. ... Ke6 option that you advocate. Fritz still gives the advantage to White in this line, too. Here is how Fritz continues: 55...Ke6 56.g4 fxg4 57.f5+ Kd6 58.Kxg4 Ke5 59.Rb6 Ra8 60.Re6+ Kd5 61.Kh5 Black's king is out of position and cannot block the pawn. Am I missing something? What position did you use when consulting the tablebases? |
|
Aug-28-05
| | Sneaky: The more I look at it, the more I start to think that <54. Rxb2 Ra3 55. Rb7+ Ke6> loses for Black. The variation that really changes my mind is 56.g4 fxg4 57.f5 Ke5 58.Re7+! Kd6 59.Re6+ K-any 60.Kxg4  |
|
Aug-28-05
| | Sneaky: If it's true that 54.Rxb2 also wins, it's not really a flaw in the puzzle, because the winning idea of Rb7+!! is still required. It's amazing that it would work a move late, when Black gets his rook on a3 "for free", but that f5+ line discussed above works out a lot like the game. Definitely not a stupid puzzle, I learned a tremendous amount today. |
|
| Aug-28-05 | | snowie1: This one is deep! But for my level, I would go with; 54 Rxb2...Ra3 55 Kxf5
Rxg3 56 Rb7+...Kf8 57 Kf6... Now, white has R & p to go with black's K in trouble and only a R. Black cannot even draw without a colossal blunder from white. |
|
Aug-28-05
| | Sneaky: snowie--I would probably handle the position just like you describe in a real game, and I'd screw it up also. As explained below that line really draws. I checked in tablebase just to make 100% sure--sure enough, it's a dead draw. To help you understand here is the final position of the line I played with my computer: Black rook at b8, Black king at g7, White rook at d7, White King at e7, White pawn at f7. Yes, it looks weird, all those pieces piled up on the 7th. There is no way for White to win this, even though the pawn made it all the way to the 7th rank there is no way to keep going. |
|
| Aug-28-05 | | MaxxLange: That's a nice looking finesse - White can advance his King without being checked on the ranks after Black's 54...Ke6. |
|
| Aug-28-05 | | Boomie: Using the Nalimov tablebase at
http://www.lokasoft.com/tbweb.htm, I set up the position after 54. ♖xb2 ♖a3 55. ♖b7+ ♔e6 56. g4 fxg4 57. f5+ ♔d6 58. ♔xg4. The result was M30, mate in 30 for white. So 55...♔e6 loses. When I tried the same position with the black king on f8, Nalimov said it's a draw. Whether white plays 56. ♔xf5 ♖xg3 or the g4 line, black gets a position with the king in front of the pawn and the rook behind. We should know or learn from this, that this is a draw.(<billcrutcher> in your analysis of 55...♔f8 you give 56. ♔f6. However the white king is on g4 here.) But what about 54. ♖b7+ ♔f8 55. ♖xb2 ♖a3? 56. ♔xf5 ♖xg3 is a draw. 56. ♖g2 ♖a5 57. g4 fxg4 is a draw. There must be something for white after 56. ♖g2 ♖a5 but I can't find it. Somebody help out please. Anyway one thing us Fritzy types should know by now, the computer is useless in certain endgames. |
|
| Aug-28-05 | | sfm: Wish I had a chess board here, but I haven't had one for 15 years. So all analysis is done in my (apparently) imperfect head.
BUT the final verdict is that the 54.Rxb2-gang were right, that black is still lost. I will have to find some convincing excuse for overlooking the simple 56.g4 which clearly wins. Hadn't missed that over the board - guaranteed. :-) Hilarious that all the clever considerations regarding the first move was not needed after all, and that the blunt and primitive capture on b2 in first move also wins. Chessgames - more "stupid" puzzles like these, please! =)) |
|
| Aug-28-05 | | Boomie: I think I've found the way for white against 54...♔f8. This combined with my previous post proves that 54. ♖b7+ is the only winning move. The key idea of the puzzle is 56. ♖g2 followed by ♖h2 when the black rook defends f5. Playing 54. ♖xb2 deprives white of this maneuver. <chessgames.com>, very nice puzzle. 54. ♖b7+ ♔f8 55. ♖xb2 ♖a3 56. ♖g2 ♖a5 57. ♔f6 ♔g8
(57... ♖b5 58. ♖h2 ♔g8 59. ♖h5 ♖b3 60. ♖xf5 ♖xg3 61. ♖g5+) (57... ♔e8 58. ♖h2 ♖a3 59. ♖h8+ ♔d7 60. ♖g8 ♖a5 61. ♖g5) 58. ♖h2 ♖a3 59. ♖h3 ♖a5 60. ♖h5 ♖a3 61. ♖xf5 ♖xg3 62. ♖g5+ ♖xg5 63. fxg5 and the pawn reaches the 6th rank with the opposition. |
|
Aug-28-05
| | Sneaky: Boomie, that line might work but I think this is simpler: 54.Rb7+ Kf8 55.Rxb2 Ra3 56.Kf6! (threatening mate) Ke8 57.Rb8+ Kd7 58. Rg8 and White will soon be two pawns to the good. |
|
| Aug-28-05 | | sfm: I hope that no one here doubts that when I wrote -
"Huh?? What's the king doing down there? It is a losing move! 55.-,Ke6 is the move to be played..." "...BUT the final verdict is that the 54.Rxb2-gang were right, that black is still lost..." - it was, naturally, only to provoke and inspire others to find the right solution (which I had seen all the time, of course). |
|
| Aug-29-05 | | Boomie: <Sneaky> Nice find. White can also play ♔f6 before ♖xb2. Another try for black here is ♔g8. The position then transposes to the "main" line. All these lines win for white. 54...♔f8 55. ♔f6 ♔g8 56. ♖b8+
(56. ♔g6 ♔f8 57. ♖b8+ ♔e7 58. ♖xb2 ♖a3 59. ♖g2 ♔e6
(59...♖a5 60. ♖h2 ♔e6 61. ♖e2+ ♔d7 62. ♔f6 ♔d8 63. ♖e5 ♖a3 64. ♔xf5 ♖xg3) 60. ♖e2+ ♔d7 61. ♔xf5 ♖xg3)
56...♔h7 57. ♖xb2 ♖a3 58. ♖h2+ ♔g8 59. ♖h3 |
|
| Aug-29-05 | | Giearth: So 54.♖xb2 draws (from <Boomie>'s findings) against correct defense by black. Once again, nice puzzle <chessgames.com>! |
|
| Aug-30-05 | | billcrutcher: <Giearth> and <Boomie> - I don't think you have made a satisfactory case that 54. Rxb2 draws against correct defense by black. <Boomie> - when you wrote "(<billcrutcher> in your analysis of 55... Kf8 you give 56. Kf6. However the white king is on g4 here.)" Surely if you take a second look, you will see I was right. In the sequence I give, the White king is indeed on g5. That's why it's not an option for the black king to move to f6 to block the pawn after 55. Rb7+. I invite you to look at it again. And <Bonnie>, your analysis using the Nalimov tablebases is less than convincing. You wrote: "When I tried the same position with the black king on f8, Nalimov said it's a draw. Whether white plays 56. Kxf5 Rxg3 or the g4 line, black gets a position with the king in front of the pawn and the rook behind. We should know or learn from this, that this is a draw."
However, white's best continuation after 55. ... Kf8 is neither the g4 line nor the Kxf5 alternative you propose. Rather, white's response is 56. Kf6, threatening a back rank mate in one. To avoid this, black either moves his king to e8 or f8, or defends with the black rook, with either Ra8 or Ra6+. With either king move, white answers with 57. Rg7, defending the g3 pawn, followed by 58. Rg5 and the f5 pawn falls. With either rook move, white simply captures the f5 pawn and has two pawns and the rook against just the rook. My money says Nalimov will not report this position as a draw. Unfortunately, the website you gave only goes through 5-piece endings, and there are still six pieces on the board. What say you? |
|
| Sep-04-05 | | Boomie: <billcrutcher> I apologize for my gaff on 56. ♔f6. I can't find a saving line for black after ♔f6. I also missed the maneuver ♖-b3-e3-e5, for which black has no defense. Too bad about the Nalimov only going to 5 pieces. Nice work. |
|
| Feb-21-20 | | rogl: < chessgames.com: The intended point of this puzzle is that the immediate 54.Rxb2 only draws, while the counterintuitive 54.Rb7+ actually wins. For example, 54.Rxb2 Ra3! 55.Rg2 (55.Kxf5 Rxg3 =) Ra5! =.
We received this position "on good authority" and we'll be the first to admit that we don't fully understand it ourselves. As the day progresses perhaps the stronger players here can help us analyze the details.> I stumbled upon this old Sunday puzzle. Now close to 15 years later we plug the position at 54.? into 7-men tablebase and learn that both 54.Rb7+ and 54.Rxb2 win. I guess it's a little bit sad that immediate answers like this bereave us from many interesting discussions like the one in this thread. |
|
 |
|
|
< Earlier Kibitzing · PAGE 3 OF 3 ·
Later Kibitzing> |
Kibitzing
