Chessboard Magic

Discuss anything you like about chess related matters in this forum.
Roger de Coverly
Posts: 21301
Joined: Tue Apr 15, 2008 2:51 pm

Re: Chessboard Magic

Post by Roger de Coverly » Tue Jun 04, 2013 2:50 pm

The position that White would like to reach is something like.



As to whether it can forced, the online Houdini analysis didn't seem to find a way.

(edit) but there's a defence with Rook on g4 as suggested by Simon Brown below.(/edit)

Last edited by Roger de Coverly on Tue Jun 04, 2013 4:06 pm, edited 2 times in total.

User avatar
Christopher Kreuzer
Posts: 8806
Joined: Fri Aug 06, 2010 2:34 am
Location: London

Re: Chessboard Magic

Post by Christopher Kreuzer » Tue Jun 04, 2013 3:26 pm

Roger de Coverly wrote:
Christopher Kreuzer wrote:
Anyway, the interesting position is here:



Black to play (obviously). Who would really play 1...Kb8 here? It is the best try apparently. Does it draw? Does White have a move where when Black captures the b6 pawn with the rook, it is a tablebase win for White? I think 2.b7 Rb6 may hold for Black, not sure.
I saw the drawing idea as being to ignore the b pawn and to avoid any position where White could capture on g6 without losing the pawn on f4. The point being that the RP v R ending is drawn, assuming correct play, with the Black king on b8. In the game when the King is on b6, the capture of the g pawn with the Rook is with check, so there's a vital extra tempo.
I don't think that is the point at all, as White can still win after 1...Kb8 with either 2.b7 or 2.Rf7 (after 2.Rf7, the tablebases say that 2...Rxb6 is lost). Other moves only draw. For example, 1...Kb8 2.Re7 Rxb6 3.Re6 Rb5+ is drawn, as is 1...Kb8 2.Rd7 Rxb6 3.Rd6 Rb5+. The White rook needs to comes to the third rank on f6, with the White king going to e6 and f7 to take the g6 pawn. The other way, with the White king going to f6 after Rb5+, allows Rf5+ to win White's f-pawn. That seems to be the crucial point here.

Can anyone turn all that into some variations that can be clicked through?

Simon Brown
Posts: 798
Joined: Wed Oct 08, 2008 8:38 pm
Location: Sevenoaks, Kent, if not in Costa Calida, Spain

Re: Chessboard Magic

Post by Simon Brown » Tue Jun 04, 2013 4:00 pm

Sorry, Geoff, I can't do the diagrams.

But I think the point is that 1..Kxb6 allows 2 Kf6 Rg3 3 Rg7 Rg4 4 Rxg6 Rxf4? 5 Ke5+ winning. With the K on b8 that doesn't work. I don't see how after 1...Kb8 either 2 Rf7 or 2 b7 win against a defence of Rg3 followed by Rg4, then shuffle the king. K can't get across in time to support the b pawn and I can't see how to win the g pawn without losing the f pawn. f5 with a Rg7 never works as gxf5 will protect the rook.

User avatar
Christopher Kreuzer
Posts: 8806
Joined: Fri Aug 06, 2010 2:34 am
Location: London

Re: Chessboard Magic

Post by Christopher Kreuzer » Tue Jun 04, 2013 4:56 pm

Is there a way to analyse these rook endings without getting a headache? :(
I keep thinking White's king can pop across in time - how does Black defend against that?

Roger de Coverly
Posts: 21301
Joined: Tue Apr 15, 2008 2:51 pm

Re: Chessboard Magic

Post by Roger de Coverly » Tue Jun 04, 2013 5:29 pm

Christopher Kreuzer wrote: I keep thinking White's king can pop across in time - how does Black defend against that?
In time for what? It's a standard piece of Rook and Pawn theory, that if you control the queening square as a defender, then you should draw. So the removal of the f and g pawns from the board is of no help to White. By contrast, the line we saw in the actual game where White captures the g6 pawn in exchange for the b6 pawn is winning because Black cannot get close enough to f8 to hold a draw.

Actually there's a potential idea. Leave the f and g pawns on the board and try to win something like this.



Black being able to play g5 is a tactical difficulty, but the main idea is to force the Rooks off. That's a draw if just the b pawn remains, but a win with the f and g pawns. So not all the standard defences will work.

Jonathan Bryant
Posts: 3452
Joined: Sun May 11, 2008 3:54 pm

Re: Chessboard Magic

Post by Jonathan Bryant » Tue Jun 04, 2013 5:31 pm

Christopher Kreuzer wrote:Is there a way to analyse these rook endings without getting a headache? :(
I keep thinking White's king can pop across in time - how does Black defend against that?

No that what makes them fun.


Do you mean something like

60 ... Kb8, 61 Kd6?

Can't Black then go something like

61 ... Rf3

and then if the kingside pawns come off Black just plonks his rook on the back rank and just sits there - as alluded to by Roger earlier - and if White tries

62 Kc6

then just check it away with

62 ... Rc3+




Rook endings just so tricky. Easy to be right for the wrong reasons. E.g. I'd originally thought

60 ... Kb8
61 Rd7 Rxb6
62 Rd6 Rb5+


was driving the king where it wanted to go. Then I concluded Black would draw after

63 Kf6 Rb4
64 Kg5 Kc7
65 Rxg6 Kd7
66 f5 Ke7


Checking with tablebase I see my final position is indeed drawn, but there's a missed win on the way and a startling obvious improvement for Black that avoids giving White even a sniff.


Anyhoo, good to see you coming over from the dark side on rook endings.


(edit: crossposted with Roger)

User avatar
Christopher Kreuzer
Posts: 8806
Joined: Fri Aug 06, 2010 2:34 am
Location: London

Re: Chessboard Magic

Post by Christopher Kreuzer » Tue Jun 04, 2013 5:47 pm

Jonathan Bryant wrote:
Christopher Kreuzer wrote:Is there a way to analyse these rook endings without getting a headache? :(
I keep thinking White's king can pop across in time - how does Black defend against that?

No that what makes them fun.


Do you mean something like

60 ... Kb8, 61 Kd6?
No, I meant after 61. Rf7 first, and possibly after a Rf6 at some point.
Jonathan Bryant wrote: Rook endings just so tricky. Easy to be right for the wrong reasons. E.g. I'd originally thought

60 ... Kb8
61 Rd7 Rxb6
62 Rd6 Rb5+


was driving the king where it wanted to go. Then I concluded Black would draw after

63 Kf6 Rb4
64 Kg5 Kc7
65 Rxg6 Kd7
66 f5 Ke7


Checking with tablebase I see my final position is indeed drawn, but there's a missed win on the way and a startling obvious improvement for Black that avoids giving White even a sniff.
63...Rf5+ is so startlingly obvious that I had already pointed it out earlier in the thread.
Jonathan Bryant wrote:Anyhoo, good to see you coming over from the dark side on rook endings.
I wasn't ever really on the dark side, but I don't think the analysis in this thread has really been top-notch (obviously including myself in that).

User avatar
Christopher Kreuzer
Posts: 8806
Joined: Fri Aug 06, 2010 2:34 am
Location: London

Re: Chessboard Magic

Post by Christopher Kreuzer » Tue Jun 04, 2013 5:52 pm

Try this position, with White to move:


Jonathan Bryant
Posts: 3452
Joined: Sun May 11, 2008 3:54 pm

Re: Chessboard Magic

Post by Jonathan Bryant » Tue Jun 04, 2013 5:59 pm

Christopher Kreuzer wrote: 63...Rf5+ is so startlingly obvious that I had already pointed it out earlier in the thread.
Oh yes, but that was after I'd looked at the thread and was posted while I was away in the coffee shop looking at the position.

As for the dark side reference - what about that thread (about the candidates' I think) when you said something like watching super GMs play out rook endings was really dull?? Anyway, I'm only teasing.

Off to look at the latest position now. We're assuming

60 ... Kb8
61 Rf7 Rg3


right? I'm not sure I'd want my rook on g3 there, but I'll go with it for now.


J

User avatar
Christopher Kreuzer
Posts: 8806
Joined: Fri Aug 06, 2010 2:34 am
Location: London

Re: Chessboard Magic

Post by Christopher Kreuzer » Tue Jun 04, 2013 6:01 pm

Roger de Coverly wrote:Actually there's a potential idea. Leave the f and g pawns on the board and try to win something like this.



Black being able to play g5 is a tactical difficulty, but the main idea is to force the Rooks off. That's a draw if just the b pawn remains, but a win with the f and g pawns. So not all the standard defences will work.
One trick after an incorrect g5 from Black is to throw in a rook check first, and then push f5.

User avatar
Christopher Kreuzer
Posts: 8806
Joined: Fri Aug 06, 2010 2:34 am
Location: London

Re: Chessboard Magic

Post by Christopher Kreuzer » Tue Jun 04, 2013 6:02 pm

Jonathan Bryant wrote:
Off to look at the latest position now. We're assuming

60 ... Kb8
61 Rf7 Rg3


right? I'm not sure I'd want my rook on g3 there, but I'll go with it for now.
Feel free to play another move...

Jonathan Bryant
Posts: 3452
Joined: Sun May 11, 2008 3:54 pm

Re: Chessboard Magic

Post by Jonathan Bryant » Tue Jun 04, 2013 6:44 pm

Christopher Kreuzer wrote:
Jonathan Bryant wrote:
Off to look at the latest position now. We're assuming

60 ... Kb8
61 Rf7 Rg3


right? I'm not sure I'd want my rook on g3 there, but I'll go with it for now.
Feel free to play another move...

Possibly it makes no difference if after 62 Rf6 you check from e3 anyway. Still, I've got to be at the club in half an hour so I'll have to leave it there for now. I don't see anything definite yet.

Your plan of cutting back to get behind the g-pawn is rather annoying and while I had hopes of taking on b6 with check with the rook already on f6 that doesn't appear to hold.

Geoff Chandler
Posts: 3486
Joined: Mon Jul 06, 2009 1:36 pm
Location: Under Cover
Contact:

Re: Chessboard Magic

Post by Geoff Chandler » Tue Jun 04, 2013 10:41 pm

I turn my back for a few hours and my thread has been hijacked by an endgame. :cry:

Why you guys make such a fuss over Rook and Pawn endings is beyond me.

How hard can they be? What is the problem?
Rooks move up and down and sideways. Pawns can only go forward.

It's nothing but a version of Sudoku!

Matt Fletcher
Posts: 271
Joined: Wed Mar 09, 2011 9:42 pm

Re: Chessboard Magic

Post by Matt Fletcher » Tue Jun 04, 2013 10:43 pm

I think this summarises the discussions so far on the original position (plus a couple of my comments based on looking at the tablebases) - but please correct me if I'm wrong:



It's a bit difficult to follow because there are a lot of options at move 2. Not sure if it's possible to tidy these up.

Geoff Chandler
Posts: 3486
Joined: Mon Jul 06, 2009 1:36 pm
Location: Under Cover
Contact:

Re: Chessboard Magic

Post by Geoff Chandler » Tue Jun 04, 2013 10:56 pm

Hi Matt

Glad to see someone is keeping us to date with diagrams/PGN's.

I put this position from above into that site that does endgames.

http://chessok.com/?page_id=361



It said within seconds this a draw.

Post Reply