- About Us
- Chess Improvement
- Chess Puzzles
- Chess Research
- College Chess
- General News
- Home
- Major Tournaments
- News
- Polgar Events
- Privacy Policy
- Scholastic Chess
- SPICE / Webster
- Susan’s Personal Blog
- Track your order
- USA Chess
- Videos
- Women’s Chess
- Contact Us
- Daily News
- My Account
- Terms & Conditions
- Privacy Policy
One thing is for sure: White can win the queen for a rook:
1. Rd8+ seems better to me than Rc8+, because otherwise black could take the Rook on d6 in his third move.
… Kg7
2. Rd7+ Now the black king can go to either g8 or g6. Kf8 or h8 lose quickly to Rc8+ and Kh6 loses because of Kf5+ and a quick mate. Let´s have a look at Kg6 first:
… Kg6
3. Ke5+ Kg5
4. Rg7+ Kh4
5. Rc4+ Kh3
6. Rc3+ and the queen is lost while the black king cannot leave the h-file. I think that white should win here. Now what about 2. … Kg8?
3. Rc8+ Qf8
4. Rxf8+ Kxf8
5. Rxa7 and white should again be the winner.
Obviously, from a material standpoint, black is ahead, but the black king is vulnerable if white has a mating attack that can win immediately, or force black to give up the queen for one of the rooks. I would look first at the two starting rook checks to see if I could mate the black king:
1. Rc8 Kg7
2. Rd7
Here, I simply can’t make Rc7 work: [2.Rc7 Kg6 (Kg8 3.Rd8+-) 3.Ke5 (what else?) Kg5 4.Rg7 Kh4 5.Rd4 Kh3 6.Ke4 Nb3 7.Rd3 Kh2 and now what for white?]. Continuing:
2. …..Kg6
3. Rg8 Kh6
Now, I had a lot of hope for this line for white- I could certainly see the box the black king would be in in this line, but how to continue? The black king will find an escape hatch at g5 after Rh8+. The other option that I can see is to play Rd6 so that a king move by white to f7, for example will uncover a check. Let’s examine both to see if we get anywhere:
4. Rh8 Kg5
5. Rd5
Here, Rg7 or Rg8 don’t seem to lead anywhere. Continuing:
5. …..Kg4
6. Rhh5
I ran out of ideas. About all I think white can accomplish in this line is to try to set up a perpetual check using the connected rooks on the fifth rank. The main problem I can foresee in this drawing plan is that the black king will almost certainly be able to use white’s d-pawn as a shield at some point and/or have time to activate the knight on the next, or a subsequent move. Continuing:
6. …..Nb3
Even if black could find a double attack on the h5 rook and the king, I don’t think the endgame is a win for black since the white king will beat the knight to the protection of the a-pawn after black wins the two rooks for the queen. If black is to win this line, he must prepare to shield the king on the queenside and find a way to advance the a-pawn. In any case, the knight at b3 provides a shield for the queen at a3 so that the black king can safely cross the third rank while also attacking d2. Continuing:
7. Rdg5 Kf3
And, I just don’t see any reason to continue this. I can’t say for sure if black can win this, but it is clear that white cannot. Clearly, to win (and I have to believe white can actually win this, otherwise, why make it a puzzle). In my next comment, I will look white’s first move alternatives.
In my last comment, I examined the beginning or Rc8+, but the black king managed to escape the rooks/king net through g5. Let’s see if Rd8 changes things:
1. Rd8 Kg7
From here, there are basically two major lines-Rc7 and Rd7:
2. Rc7 Kg6 (Kh6 3.Rh8)
3. Rg8 Kh6
4. Rh8
Here, I don’t think Rg7 changes things for the good in any way, and actually looks a little worse to me for white. Continuing:
4. …..Kg5
5. Rg8 Kh6
At first, I thought Kh4 would end up the same place as the line of the previous comment, but I overlooked something- white controls c3 with the pawn: [Kh4?? 6.Rc4! Kh3 7.Rc3 Qc3 8.dc3 and white should be able to queen the pawn before black can engineer a stalemate of his king on h1, or use his knight to get the king off the h-file]. Of course, that win depends on the trapping of the black king on the h-file- if [Kf4?! 6.Rc4 Kf3 7.Rc3 Qc3 8.dc3 h4 looks drawish, but there might be a way for white to win]. So, black does best to just allow the perpetual by returning the king to h6. So, white can at least draw in this variation.
Back at move 2, there was Rd7. From the top:
1. Rd8 Kg7
2. Rd7 Kg6 (Kh6 3.Kf5+-; Kg8 bel.)
And, now, I can finally see it!
3. Ke5 Kg5
4. Rg7 Kh4
5. Rc4 Kh3
6. Rc3 Qc3
7. dc3 a5 (h4 later)
8. c4 Nb3
And the question now is, how does white continue the c-pawn’s advance? Or should he try for a mate on the black king? Or is the ending where white forces black to exchange the knight for the pawn won? Actually looking at this shows me I was too quick to dismiss this sort of position as winning for white in the last variation. My instinct tells me this is a draw if the last white pawn is exchanged for the knight. What I am worried about is that when white gobbles up the a-pawn with the king, black has enough time to advance his king to h1 and the pawn to h2 engineering a stalemate, or a theoretical draw if the white rook gives up the g-file. The other option in that line is to do the exchange at c4, bring the king back to f3 to keep black’s king trapped, and use the rook to win at a1 or a2. Let’s look at both possibilities before we try to win this by queening the c-pawn, but let’s continue this in the next comment since I am going to run up against the length restriction for a single comment.
I don’t see how white is better.
In my previous comment, I had developed the following line:
1. Rd8 Kg7
2. Rd7 Kg6
3. Ke5 Kg5
4. Rg7 Kh4
5. Rc4 Kh3
6. Rc3 Qc3
7. dc3 a5
8. c4 Nb3
To summarize the position, black has two passed pawns and a knight. For the moment, white’s rook has the black king trapped on the h-file, but black has the white pawn blocked by the knight’s attack on c5. The question is- how does white win this without giving black a stalemate or a theoretical draw with the black king free of the h-file supporting the advance of the h-pawn at h2? First, I am going to look at the brute force exchange at c5, win the a-pawn with the king, and see if the white king has enough time to make it back to the king side before black forces stalemate, or the white rook off the g-file.
9. Kd5 a4 (h4 below)
10.c5 Nc5
11.Kc5 a3
Now, the pawn can’t be stopped by the white king. It takes white three moves to reach f3 with the king to pin black to the h-file, and one move to bring the rook to a7 or g1 to cover a1. I think the only issue is can the white king get close enough to the king and h-pawn to win that ending? And if the answer is yes, then how should the white king make the approach? I am forced to simply look at the lines because I can’t easily see all the issues in my head.
12.Kd4 a2
13.Ra7
Here, I looked at Rg1, but after Kh2, the black king gets off the h-file one move earlier than he does with Ra7:
13. …..h4 (Kg3 looks similar)
14. Ke5
An insight for me, I think. The black king is going to get off the h-file, that much is clear, and he is going to do it via g3 or g2 eventually. In order to stop the pawn, white is going to, at a minimum, have to check black from the a-file with the rook and be able to push him away from the h-pawn, or be able to push him into the first rank- for that to work, the white king needs to be able to face the black king from an effective opposition without allowing himself to be used as a shield- this means the white king is best to be on the fifth rank right now. Continuing:
14. …..Kg4
Trying to foil the plan of checking black with the rook while getting the white king in opposition. Won’t work, I think:
15. Ra4 Kg3 (Kg5 better? Don’t kn)
16. Kf5 h3 (a1Q different?)
17. Ra3
Basically a repeat of move 15, but one rank down. White wants to take at a2 with check. Continuing:
17. …..Kg2 (Kh4 below)
18. Ra2 Kg3
19. Ra3 Kg2
20. Kg4 h2
21. Ra2 Kg1 (Kh1 22.Kg3+-)
22. Kg3
And this sort of position has been a puzzle on this site more than once. If black queens the pawn, Ra1 is mate, so black either loses the pawn at h2, or underpromotes to a knight, but this is still lost as white enforces zugzwang on black:
23. …..h2(N)
24. Kf3 and the knight must be lost at g3 or f2, or black gets mated.
Back at move 17 in this line, there is a question of whether black is too compliant by playing to g2 rather than h4, but I am sure this makes no difference now:
17. …..Kh4
18. Kf4! h2 (or lose the pawn)
19. Ra1 Kh3 (Kh5 20.Rh1+-)
20. Kf3 Kh4 (or mate)
21. Kg2 wins the pawn.
So, I think white can win the ending that arises by forcing the exchange at c5 on move 10. The only doubt I have is what happens if black pushes the h-pawn instead of the a pawn at move 9. Continued in my next comment.
I don’t know if I am right, but…
1. Rd8+ Kg7
2. Rd7+ Kg6
(If 2… Kf8, Kg8 or Kh8, 3. Rc8+, winning. If 2… Kh6, 3. Kf5+, winning)
3. Ke5+ Kg5
4. Rg7+ Kh4
5. Rc4+ Kh3
6. Rc3+ Qxc3
7. bxc3
Black’s pawn “a” is hanging, while the Knight in “a1″ is very bad and the }”h” pawn is too far from h1.
White must be winning because of the c pawn.
In my previous comment, I showed the line:
In my previous comment, I had developed the following line:
1. Rd8 Kg7
2. Rd7 Kg6
3. Ke5 Kg5
4. Rg7 Kh4
5. Rc4 Kh3
6. Rc3 Qc3
7. dc3 a5
8. c4 Nb3
9. Kd5
In the previous comment, I had analyzed the ending that arises after black pushes the a-pawn, and I hope I had shown how that was lost for black as the white king has enough time to come at the black king from opposition, or one square off opposition, and use the rook on the a-file to check and push black back, and eventually win the a-pawn with check, then continue pushing black back forcing an underpromotion at h1 followed by zugzwang losing the new knight. I want to see what happens when black pushes the h-pawn at move 9 instead to see if it is any better for black:
9. …..h4
10. c5 Nc5
11. Kc5 Kh2 (a4 below)
12. Kd4
It takes white 9 moves to win the a-pawn with the king and bring the king back as far as f3. It only takes black 3 moves to put his king on h1 and the pawn on h2- that will be be a draw for sure. So, white must use a plan similar to the one I outlined in the previous comment- bring the king back to the king side immediately, and let the rook deal with the a-pawn as necessary. Continuing:
12. …..h3
Here, a4 is going to transpose to the line in the previous comment, so we can ignore it. I am just interested in proving that rushing the h-pawn changes nothing for black. Continuing:
13. Ke3
An obvious move based on the previous analysis. Ke4 should win as well. Continuing:
13. …..Kh1
14. Kf2
We should be able to see where this is all going by now:
15. …..a4 (h2 16.Re7 a4 17.Re1#)
16. Rg3 Kh2 (h2 17.Re3+-)
17. Ra3 with mate to follow on the next move. So, clearly, pushing the h-pawn at move 9 is even worse.
I am going to go back to before the knight was forced off the board to see if white had any shorter wins, but I am doubting it right now.
white is better because he can win the queen! very nice!
white is better by winning the black queen:
1. Rd8+ Kg7
2. Rd7+ Kg6 (else backrankmate)
3. Ke5+ Kg5
4. Rg7+ Kh4
5. Rc4+ Kh3
6. Rc3+ Qxc3
7. dxc3
next capture the a-pawn.
greets, jan
it is not very easy to see in this case… but we can, at least change a rook for the queen… Rd8+ Kg7 Rd7+ Kg6 Ke5+ Kg5 Rg7+ Kh4 Rc4+ Kh3 Rc3+ Qxc3+ dxc3 … after this point, white can find a way to crown his c pawn…
White wins.
1. Rd8+ Kg7
2. Rd7+ Kg6
If the King moves to the back rank, 3. Rc8#. If 2….Kh6 3. Kf5 mates next move.
3. Ke5+ Kg5
4. Rg7+ Kh4
5. Rc4+ Kh3
6. Rc3+ wins the Queen.
I wrote my response without a board, and now I am wondering if my first checks were with the d Rook or the c Rook. Obviously, the d Rook is intended.
1.Rd8+ Kg7 2.Rd7+Kg6 3.Ke5+ Kg5 4.Rg7+ Kh4 5.Rc4+ Kh3 6.Rc3+ Qxc3 7.dxc3 and I believe White quickly rules this out
For those who found the way to win the queen for a rook, this isn’t necessarily the end of the story. Now, if you found a mate with the rooks, then bravo, because I have completely missed it.
This looks like a draw to me. I can’t find anything better for white than a perpetual.
1.Rd8+ Kg7
2.Rd7+ Kg6
3.Kd5+ Kg5
4.Rg7+ Kf5
5.Rf7+ Kg4
6.Kg7+ Kf5+ (not 6…Kf4 7.Rf6# and 6…Kh4 7.Rc4 Kh3
8.Rc3+ wins the queen).
So it’s a draw I guess.
Man, everyone here thinks that white can win the rook for a queen, but they failed to realized that black doesn’t have to play 4….Kh4. Instead 4….Kf5 draws.
henryK,
Your third move for white is an error. 3.Ke5 prevents the line you mention from arising.
And for my own mea culpa, you can ignore every move I analyzed after white takes the black queen at c3, and black plays Nb3 at move 7. White wins directly with 8.Ra7, and probably wins very easily with 8.Kd5, too. The truth is that I didn’t even consider either of these plans, and I don’t know why- they are both obvious to me this morning. Indeed, I can confirm for sure that 8.Ra7 wins in that line because my Nalimov Tablebase says so. I can’t confirm that 8.Kd5 wins without analyzing a bit (the tablebase is for six or fewer men on the board), but keeping the c4 square clear gives the white king a clear path to attack the knight without allowing the a-pawn to advance too far.
At move 7, black probably does best to simply play a5 first followed by Nb3. Then, it isn’t so clear to me right now if white can win by bringing the rook to bear on the queenside. From the top:
1. Rd8 Kg7
2. Rd7 Kg6
3. Ke5 Kg5
4. Rg7 Kh4
5. Rc4 Kh3
6. Rc3 Qc3
7. dc3 a5
The line might continue with:
8. Ra7 Nb3 (protecting a5)
And, if white persists with the standard R vs N+P technique of switching the attack between the two opponents to force the knight to give way, this could end up in a draw:
9. Rb7 a4 (protecting the knight)
10.Ra7
Switching back to the attack on the pawn:
10. …..Kg3!
And, now, if white takes at a4
11.Ra4 h4 is a draw according the Nalimov Tablebase (http://www.k4it.de/index.php?topic=egtb&lang=en).
There may still be a way for white to win after his 10th move, but it isn’t obvious to me. It might be better to delay c4- keep the black king trapped by the rook, and bring the white king through c4 and d3 (the knight is on c2 in the line I am think of) to deal with the knight and pawn. However, this can be very difficult to do in some lines. For example, from the line above ending at move 7:
8. Kd5 Nc2
9. Kc4 a4 (cutting off b3)
10.Kd3 Na1
From here, I can see two major plans for white- bring the white king back to f3 (white has managed to close the distance a bit) and trap the black king on the h-file with the king, and then use the rook against the knight and pawn, with, or without advancing the c-pawn. Or start advancing the c-pawn at move 11. Let’s just look at the first idea in a general line:
11.Ke3 a3
12.Kf3 Nb3
13.Ra7 Nd2
14.Kf4 Nc4
15.Ra4 Nb6
16.Ra3 and the Nalimov tablebase confirms this as a win for white.
In any case, I think I am done with this problem. Very interesting endgame after the queen for rook exchange.