Chess.com's Weekly Study: May 1st 2016

DailyFun

Hello Chess.com!

Starting with 2016, we will be posting a Weekly Study, courtesy of Yochanan Afek, Grandmaster for composing endgame studies. These challenging positions are designed to stimulate (and improve upon!) your creativity, depth of calculation, pattern recognition and pure imagination.

G. Zakhodyakin, 1978

Enjoy!

 

DailyFun

Make sure you check out last week's puzzle, too!

https://www.chess.com/forum/view/endgames/chess-com-s-weekly-study-april-24th-2016

muhsin_ibn_al_azeez
how it become a draw?
muhsin_ibn_al_azeez

I mean, for example, 6.Rb1 can be replayed with Qxb1 and still white have a decisive advantage.

PJsStudio

Very interesting composition that I didn't understand while I was going over it. Basically white sets up a shield around his own king that blacks king cannot approach. After black queens, white checks the black King back and forth endlessly until the black King must cross the fifth rank and then exchanges rook for Queen with a theoretical draw of knight vs rook.

This is very impressive.

fightingbob
DJsStudio wrote:

Very interesting composition that I didn't understand while I was going over it. Basically white sets up a shield around his own king that blacks king cannot approach. After black queens, white checks the black King back and forth endlessly until the black King must cross the fifth rank and then exchanges rook for Queen with a theoretical draw of knight vs rook.

This is very impressive.

I agree with you DJ, it is impressive and interesting.  It has a quasi-domination theme that ends in perpetual check or, as you noted, a theoretical draw of rook against knight (though I actually won this match up a long time ago but shouldn't have).

After the initial check with the knight, it was obvious Black's king had to stay along the second rank after each check because anything else allows White to get to the first rank and capture the promoted queen.  I got all the way to 4.Rf5+, which was pretty obvious, and knew 5.Rg5+ wouldn't cut it because of ...Kh3.  I was stumped.

I looked at 5.Rb5 but thought what good does that do because as I check the king he moves along the h1-a8 diagonal and eventually out of trouble.  I even looked for a stalemate, but quickly dismissed that possibility.

After a short break, I returned to notice the peculiar geometry of Black's pieces and realized the queen could only move along rank and file.  That had to be part of the solution, especially since she was constrained by her own rook.  Then it became clear I had to threaten the queen first with capture and get her to move.  After that the position played itself.

I thoroughly enjoyed this one not only because I enjoy domination themes but also because I learned something quite practical.

Best,

Bob

psychottico

I don't understand why when white moves 5.Rb5 is black moving h1=Q shouldn't they move BRf6. It seems like the winning move to me, please let me know if I'm wrong.

n9531l
psychottico wrote:

I don't understand why when white moves 5.Rb5 is black moving h1=Q shouldn't they move BRf6. It seems like the winning move to me, please let me know if I'm wrong.

Then White could play Rh5 and can't be stopped from trading rook for pawn, all that's needed for the draw.

greenpointjerzy

why not 1...kd3 and then if 2 Rd5 Ke4 to stop checks? I don't see why the black king moves towards its own pawn. The rook defends it and the white knight can't take it if he can't fork king and pawn. Later, that king just impairs the mobility of his new queen.

caveatcanis

why not 1...kd3 and then if 2 Rd5 Ke4 to stop checks?

3. Rd1 h1=Q 4. Rxh1 Rxh1 would be a drawn R v N endgame.


n9531l

Just so you won't get the idea that a knight always draws against a rook, here's a position to consider. If you have trouble finding White's mate in 37, your computer should be able to help.

White to play and win



fightingbob
n9531l wrote:

Just so you won't get the idea that a knight always draws against a rook, here's a position to consider. If you have trouble finding White's mate in 37, your computer should be able to help.

White to play and win

Bob, you should have included the 37 most efficient moves -- no need for variations -- right within the diagram so we could play them out.  There is a way of doing that.

Best,
Bob

n9531l

That's more work than I wanted to do for the small number of people that would want to see it. But if you think the demand is there, you could give the moves. There is a way of doing that.

fightingbob
n9531l wrote:

That's more work than I wanted to do for the small number of people that would want to see it. But if you think the demand is there, you could give the moves. There is a way of doing that.

Okay, Bob, I just did.  Now watch me get questions about all the quicker mates.  Folks, go to Shredder's 6-Piece Endgame Database, set up the pieces, and work through it to your own satisfaction.

Actually, it's quite instructive; sometimes you threaten to corral the king, other times you threaten to corral or capture the knight. The general concept is to keep the two separated or uncoordinated when together, and keep the king moving toward the corner.  Interestingly, there are situations where almost any White move will do and others where only one move by White will keep the win alive.

Buon Appetito!

Best,
Bob

 



n9531l

Thanks, Bob. I knew it was going to be quite a bit of work.

n9531l

By the way, here's a suggestion for showing a winning line. When a move by White is the only one to keep the win, put a ! after it (Nunn convention). This gives a loose indication of the difficulty of the winning procedure.

crossfire125

Very good!! I didn't find it , of course...