PDA

View Full Version : Instructive Rook endgame



Jesper Norgaard
17-07-2009, 04:41 PM
Another fisherman's story - it was this big and I almost got it! I had White and match point in this position. White to move and win:

=2r5/4k1p1/2P1p3/p3K2p/P1R4P/8/6P1/8 w - - 0 49

Hint zugzwang could be an element! This will span the current Rook endgame, pawn endgame, and Queen endgame!

Spiny Norman
18-07-2009, 06:55 AM
Interesting position. I've only spent about 2 minutes looking at it, so the following is likely to be completely wrong, as it is based on the ravings of a 1500-odd who is just working off general principles, not concrete variations:

* White has an advantage consisting of
-- an advanced passed pawn that is not blockaded by Black's king (whereas Black's passed pawn is not nearly so far up the board and is blockaded by White's king)
-- there are pawns on both sides of the board which are undefended, meaning that one could be 'picked off' to create a material advantage
-- if Black exchanges rooks I think he has a lost king+pawn endgame

I would probably play c7 right away, to lock Black's rook down, then follow up with Rc5 and prepare to win material.

Having said that (i.e. this is what I would do) ... I would be almost certain that this is not the idea you have in mind, as it seems rather obvious from the position given.

Desmond
18-07-2009, 07:40 AM
Maybe what you're getting at is that the immediate c7 is inferior to waiting first then playing it, for example


1. Rc5 (1. c7 Kd7 2. g3 Rxc7 3. Rxc7+ Kxc7 4. Kxe6 Kc6 5. Kf7 Kc5 6. Kxg7 Kb4
7. Kg6 Kxa4 8. Kxh5 Kb4 9. g4 a4 10. g5 a3 11. g6 a2 12. g7 a1=Q 13. g8=Q) 1...
Rc7 2. Rc3 g6 3. Rc5 Rc8 4. c7 Kd7 5. Kf6 Rxc7 6. Rxc7+ Kxc7 7. Kxe6 Kc6 8. Kf6
Kc5 9. Kxg6 Kb4 10. Kxh5 Kxa4 11. Kg6 Kb3 12. h5 a4 13. h6 a3 14. h7 a2 15.
h8=Q

Spiny Norman
18-07-2009, 10:23 AM
Interesting. I obviously hadn't calculated much, as I didn't see that Black has just enough time to get his queen and end up in a drawn Q+PvQ position

MichaelBaron
18-07-2009, 10:51 AM
Rc5 followed up by Rxa5 and Ra7+ looks convincing enough. White's king is far more active.

Jesper Norgaard
18-07-2009, 07:21 PM
Very interesting tries! I can see you have really gotten the little grey cells working!
I think as a team you are perhaps close to a solution ...



I would probably play c7 right away, to lock Black's rook down, then follow up with Rc5 and prepare to win material.
Yes but as Boris points out after 1.c7 Kd7 White already has to answer to the c7-pawn, not leaving much time for Rc5


Maybe what you're getting at is that the immediate c7 is inferior to waiting first then playing it, for example


1. Rc5 (1. c7 Kd7 2. g3 Rxc7 3. Rxc7+ Kxc7 4. Kxe6 Kc6 5. Kf7 Kc5 6. Kxg7 Kb4
7. Kg6 Kxa4 8. Kxh5 Kb4 9. g4 a4 10. g5 a3 11. g6 a2 12. g7 a1=Q 13. g8=Q) 1...
Rc7 2. Rc3 g6 3. Rc5 Rc8 4. c7 Kd7 5. Kf6 Rxc7 6. Rxc7+ Kxc7 7. Kxe6 Kc6 8. Kf6
Kc5 9. Kxg6 Kb4 10. Kxh5 Kxa4 11. Kg6 Kb3 12. h5 a4 13. h6 a3 14. h7 a2 15.
h8=Q

Right! Although in your main line 2...g6 is a serious weakening that invites your later refutation 5.Kf6 - I would play 2...Rc8


Interesting. I obviously hadn't calculated much, as I didn't see that Black has just enough time to get his queen and end up in a drawn Q+PvQ position

Yes as Shakespeare would have put the question - to be or not to be a winning Queen endgame!


Rc5 followed up by Rxa5 and Ra7+ looks convincing enough. White's king is far more active.

I am not fully convinced, and anyhow this is not the intended (and easier) solution.


1. Rc5 Rc7 2. Rxa5 Rxc6 3. Ra7+ Kf8 4. a5 Kg8 5. a6 Kh7 6. Kd4 Kg6 7. Ra8 Rc2 8. a7
Ra2

It seems quite difficult to prove the a-pawn, when the White King gets close to a7 he will be haunted with Checks from the Black Rook. Any improvement on that line, Michael?

Ian CCC
18-07-2009, 11:38 PM
Following the hints, and the previous analysis, I would suggest 1.Rc5 Rc7 2.Rc3 Rc8 3.c7 Kd7 4.Rg3 then if 4..Rxc7 5.Rxb7+ , or if 4...Kxc7 5.Rc3+ followed in both cases by swapping rooks and getting a won K&P endgame.

Jesper Norgaard
19-07-2009, 03:29 AM
Following the hints, and the previous analysis, I would suggest 1.Rc5 Rc7 2.Rc3 Rc8 3.c7 Kd7 4.Rg3 then if 4..Rxc7 5.Rxb7+ , or if 4...Kxc7 5.Rc3+ followed in both cases by swapping rooks and getting a won K&P endgame.
The question is, how do you win after

1.Rc5 Rc7 2.Rc3 Rc8 3.c7 Kd7 4.Rg3 Rg8? it looks extraordinarily passive for Black but still I would like to know the continuation ...

Desmond
19-07-2009, 07:54 AM
The question is, how do you win after

1.Rc5 Rc7 2.Rc3 Rc8 3.c7 Kd7 4.Rg3 Rg8? it looks extraordinarily passive for Black but still I would like to know the continuation ...OK if you are worried about that then play Rc3 when the black rook is on c8, and meet ...Rc7 with Rg3.

Oepty
19-07-2009, 08:42 AM
The question is, how do you win after

1.Rc5 Rc7 2.Rc3 Rc8 3.c7 Kd7 4.Rg3 Rg8? it looks extraordinarily passive for Black but still I would like to know the continuation ...

Make it 1.Rc5 Rc7 2.Rc3 Rc8 3.c7 Kd7 4.Re3+ Ke7 (4... Kxc7 5 Rc3+ loses) 5. Rg3 g6 (5 Rg8 loses to 6. Rxg7 and 5 Kf7 loses to 6. Kd6) 6. Rxg6 or Rc3 both win.

Scott

Ian CCC
19-07-2009, 09:52 AM
I think you've got it, Scott. Well done.

Jesper Norgaard
19-07-2009, 07:10 PM
I think you've got it, Scott. Well done.
Thanks all for the effort, Scott nailed the final spikes to the cuffin, and after 5.Rg3 Black is pretty much bankrupt. He has to go to the pawn ending as pointed out by Boris, where he will not even be a contender because h8Q stops a1Q. Such a geometrically simple solution, but still pretty difficult to find. I hope you enjoyed it all ...

In the game what happened was I played 1. c7 Kd7 2.Rd4+!? Ke7! and I realized I had ben caught napping, so 3.Rc4 Kd7 4.Rc3 Rxc7 followed and the Queen ending where White has an h-pawn is drawn with best play (proven by Nalimov tablebases). The pawns are of more value towards the centre and of less value towards the rims in Queen endings. a- or h-pawn should be draw, b- or g-pawn has some winning chances, c- and f- pawns are pretty advantageous, and d- or e-pawn is winning unles there is an outright perpetual check.