Saturday, 22 September 2018

Study 157

Árpád Rusz
Probleemblad
2017

White wins

1. Nf3 Rh3 2. Ke4 Rxf3! 3. Rh6+ 3. Kxf3? Nh4+ -+ 3... Nh4! 3... Kg1 4. Kxf3 +- 4. Rxh4+ Kg2

5. Rh1!! Thematic try: 5. Rh2+? Kxh2 6. Kxf3 a5 7. Ke4 Kg3 8. Kd4 Kf4 = 5... Rc3 5... Kxh1 6. Kxf3 +- 6. Rc1 a5 7. Kd4 Rc8 8. c4 +-

Study 156

Árpád Rusz
Pat a Mat
2017

White wins

1. Kd7 e6 2. fxe6 2. f6+? Kg6 3. Ra1 Rb8 4. Ke7 (4. Rg1+ Kf5 =) 4... Rb7+ = 2... fxe6

3. Kxe6+ White is firing the first battery. 3... Kg6

4. Ra6! White is building the second battery. 4... Rb8 5. Ke7+ Kg7

6. Ra7! The third battery is on the board. 6. e6? Rb7+ 7. Kd6 Kf6 = chameleon echo, WCCT-10 (position B) 6... Kg6 7. e6 Kg7


cyclic zugzwang - WTM

8. Kd6+ White is firing the third battery. 8... Kf6 9. Kd7! 9. e7? Kf7 10. Kd7 Re8 11. Kd6 Rb8 positional draw 9... Kg7! 10. Ke7! 10. e7? Kf7 11. Ra1 Rb7+ 12. Kd8 Rb8+ 13. Kd7 Rb7+ perpetual check


cyclic zugzwang - BTM

10... Rb1 11. Ra8! White destroys the battery. Firing the battery again would be a mistake: 11. Ke8+? Kf6 12. e7 Rb8+ 13. Kd7 Kf7 = chameleon echo 11... Rb7+ 12. Kd6 Kf6 WCCT-10 (position A) 12... Rb6+ 13. Kd7 Rb7+ (13... Kf6 14. Rf8+ Ke5 15. e7 Rb7+ 16. Kd8 Kd6 17. Rf6+ +-) 14. Kc6 Re7 15. Kd6 Kf6 16. Rf8+ +- 13. Rf8+ Kg7 14. e7 +-

Note: The manoeuvre from moves 8 to 10 was discovered by O. Frink (1926)

Study 155

Árpád Rusz
Garcia-70 JT
2018
Special Honourable Mention

White wins

1. Rd3+!! rook sacrifice 1... Ke5 1... Kxd3 2. Rd1+ Kc4 3. Rxd6 Kc5 4. a7! +- rook sacrifice 2. Rxd6 Kxd6 3. a7 Rxe6+ 3... Ra4 4. e7! Kxe7 5. Rg8! Rxa7 6. Rg7+ +-

4. Rg6!! rook sacrifice 4... Rxg6+ 5. Kxh7 +-

Thematic try: 1. Rd1+? Ke5 2. Rxd6 Kxd6 3. a7 Ra4! 3... Rxe6+?

4. Rf6!! rook sacrifice 4... Rxf6+ 5. Kg7! +- 4. e7 Kxe7 =

Study 154

Árpád Rusz
Schach
2018

White wins

1. Nxa3+! (1. Nxc3+? Ka6 =) 1... Qxa3 (1... Ka6 2. Nc4 h2 3. Ra8+ Kb7 4. Ra5 +-) 2. Rb8+ Both dangerous white pawns are pinned but black unpins the d6 pawn and sets a trap: 2... Kc5! 3. Rxc3+! fork. It was very tempting to push the unpinned pawn: 3. d7? Qa7! Both pawns are pinned again. 4. Rc8+ Kd5 5. Rf1 h2 6. Ke8 Qxd7+! (6... Qe3+? 7. Kf8 Rxf7+! 8. Rxf7 Qh6+ 9. Kg8 Qg6+ 10. Rg7 Qe6+ 11. Kh7 Qf5+ 12. Kh8! Qxd7 13. Rxd7+ Ke6 14. Re8+! Kxd7 15. Re1 +-) 7. Kxd7 Rxf7+! 8. Rxf7 h1=Q = 3... Qxc3 4. Rc8+ Kd4 5. Rxc3 Rxf7+! (5... Kxc3 6. Ke8 +-)

6. Ke6!! White is very polite and lets Black to be the first to take the rook. Thematic try: 6. Kxf7? Kxc3 mutual zugzwang 7. d7 h2 mutual zugzwang 8. d8=Q h1=Q mutual zugzwang 9. d4 Qb7+ 10. Ke6 (10. Qe7 Qd5+ perpetual check 11. Qe6 Qb7+ 12. Qe7 Qd5+ perpetual check) 10... Qe4+ = The e4-square was no longer controlled. 6... Kxc3 7. Kxf7 mutual zugzwang

7... h2 8. d7 mutual zugzwang 8... h1=Q Black is also let to be the first to promote the pawn. 9. d8=Q mutual zugzwang

9... Qb7+ Now we see the reason for white's eccentric play: the black queen has to leave its ideal square. 9... Qf3+ 10. Qf6+ +- 10. Ke6! Contrary to the thematic try, now there is no check on e4. 10... Qh1 The black queen returns to h1 but something has changed: the white king is protecting the d5 square! 11. Qa5+! Kxd3 12. Qd5+ +- fork

Study 153

Árpád Rusz
Tarasiuk-50 JT
2017
Special Honourable Mention

White wins

1. Qd4+ e5 2. Nxe5 f6! 3. Nxf6 b1=Q+ 3... h1=Q 4. Nf7# 4. Ka7 Qgb3 4... Qh7 5. Nxh7 Kxh7 6. Qxd7+ Kxh6 7. Ng4+ Kg5 8. Nxh2 +-

5. Qb2!!


Sandwich Theme

5... Qb7+! Domino Theme 5... Q1xb2 6. Ng6#; 5... Q3xb2 6. Nf7#; 5... h1=Q 6. Qxb3 Qxb3 7. Ng6# 6. Qxb7 Qxb7+ 7. Kxb7 h1=Q+ 8. Kb6! The king returns to its initial square after drawing a triangle on the board. 8. Ka7? Qxh6 9. Nf7+ Kg7 10. Nxh6 Kxf6 11. Kb6 a4 =; 8. Kc7? Qh2 =; 8. Kc8? Qa8+ =; 8. c6? Qxc6+! 9. Nxc6 dxc6 10. Kxc6 a4 = 8... Qxh6 9. Nf7+ +-

Thursday, 13 September 2018

7-man Mutual Zugzwangs (KNPP vs KNP)

1,766,467 mutual zugzwangs have been found in this endgame. (The side with two pawns wins only if the other side is on move.) That is almost twice the number of mutual zugzwangs from all endgames up to six pieces! You can download them in two parts from here:
KNPPvKNP_part1 (wKa1-d4)
KNPPvKNP_part2 (wKa5-d8)

Here's sample with mutually attacking undefended knights:


mutual zugzwang

In the following position white has a nice winning move leading to a mutual zugzwang position:

1. Nd6!! mutual zugzwang (1. d6? Nc6 =) 1... Ng6 2. Nf5(c8) Kc4 3. Nxe7! Nf4 4. Nf5 (or 4. Nc8 Nxd5 5. Nb6+! +-) 4... Nxd5 5. Ne3+! Nxe3 6. e7 +-

There are also 88,879 mutual zugzwangs where the side with one pawn wins:
KNPvKNPP

An example:

1... Kb4 2. Nc7 Nb6 3. Nd5+! +-

The position remains a mutual zugzwang even if we complete the a-file with a black pawn on a2:


An 8-man mutual zugzwang

Monday, 10 September 2018

7-man Mutual Zugzwangs (KRPP vs KRP)

I have checked this tablebase with SEE, and I have found 119707 mutual zugzwangs where the side with 2 pawns wins only if the other side is to move. Here you can download them in a pgn file:

KRPPvKRP

Here's an interesting one with mutually attacking rooks, both of them being undefended:


mutual zugzwang(!!)

There are also 9030 mutual zugzwangs where the side with 1 pawn wins:

KRPvKRPP

The 70 full point mutual zugzwangs are in both lists, so they are counted twice. Here is a separate list of them:

Full Point Mutual Zugzwangs

Analysing more than a hundred thousand positions isn't feasible but I have developed a method to identify all positions where there are corresponding squares between two rooks. An example:

1. Rh2! Re5 2. Rh1! Rf5 3. Rh4! Rg5 4. Rh3! Rd5 5. Rh2 positional draw

I cannot yet show publicly the two studies I have composed also featuring 4-4 corresponding squares. The best of them is a win study where white must visit every corresponding square before winning! It seems that is unique in this tablebase. Please, don't rediscover it... :)

Monday, 3 September 2018

Study 152

Árpád Rusz
RCS blog
2018

White wins

1. g4! 1st knight sacrifice 1... hxg3 e.p. 1... Bxe4 2. a6 +- 2. Nxg3+ Kg2

3. Nh1!! 2nd knight sacrifice 3. Nh5? Bc4! (3... Kxh3? 4. Nf4+ +- echo) 4. h4 Kh3 = 3... Be2 3... Kxh3 4. Nf2+ +- echo; 3... Kxh1 4. h4 Kg2 5. h5 Kf3 6. h6 Kf4 7. a6! Kg5 8. a7 +- 4. h4 Kh3! 4... Kxh1 5. Ke7 Kg2 6. Kf6 Kg3 7. Kg5 +- 5. h5! Bxh5 6. Nf2+ Kg3 7. Ne4+


Rundlauf

7... Kf4 8. a6! 3rd knight sacrifice 8... Kxe4 9. a7 +-

Sunday, 26 August 2018

7-man Mutual Zugzwangs (K + 3P vs K + 2P)

All 7-man Syzygy tablebases have been generated by Bojun Guo! It took 5 months on two high-end computers with more than 1TB memory each. Because of the dependencies, the K + 3P vs K + 2P tablebase was built last. I have checked this tablebase with SEE, and I have found 173990 mutual zugzwangs where the side with 3 pawns wins only if the other side is to move. Here you can download the list in a pgn file:

KPPPvKPP

Believe it or not, this is a mutual zugzwang. No, the h5 pawn is not misplaced!


mutual zugzwang

There are also 36955 mutual zugzwangs where the side with 2 pawns wins:

KPPvKPPP

The 6790 full point mutual zugzwangs are in both lists, so they are counted twice. Here is a separate list of them:

Full Point Mutual Zugzwangs

In total, there are 204155 unique mutual zugzwangs in the K + 3P vs K + 2P endgame.

Tuesday, 31 July 2018

An example of a Syzygy Endgame Explorer search


White wins

The only move to win is 1. Qh1!!
This position was found with the help of SEE using the following commands:

[Tablebase "KQPkrr"]
[Output "out.pgn"]

[Position "1"]
[Eval "win"]
[SideToMove "white"]
[Piece2 "a8"]
[Piece5 "rank1"]
[Piece6 "fileH"]
[OnlyMove "yes"]
[Move "Qa8h1"]
[Capture "no"]
[Check "no"]

[Position "2"]
[AttackedByBlack "h1"]
[NotAttackedByWhite "h1"]

I was searching for a position where the only winning move for white is Qa8-h1, a silent move without capture or check. Two black rooks should protect h1, so the queen move should be a sacrifice on a double protected corner square! It turned out that I was lucky: there exists a position like that!
I have succeeded to turn this position into a nice study which will be published soon.

Friday, 13 July 2018

A tempo move which mirrors the position

Sometimes you wish you could pass the move to your opponent without making the slightest change in the position. Of course, the chess rules forbid this. But in the following position discovered with the help of SEE, you can "break" the rules!


White wins

1. Kf4! (zugzwang) (1. Nbc5? Nb6! =; 1. Nbd6? Nc7! =) 1... Nb6 2. Nbd6 +- (zugzwang) 1... Nc7 2. Nbc5 +- (zugzwang)

After the key we had practically the same position as before but with black to move!

Here are the other positions discovered so far:




The following positions feature more exotic material:



SEE also confirms my suspicion that the only position with similar properties having fewer than 7 pieces is one discovered by me last year:


White wins

1. Bh4! +- (zugzwang)

Friday, 15 June 2018

Syzygy Endgame Explorer (SEE)

Syzygy Endgame Explorer (SEE) is a software written by me for mining the newly generated 7-man endgame tablebases. In my previous post, I've presented the first study based on this mutual zugzwang found by SEE:

The tool can not only find mutual zugzwangs but can also identify positions where interesting moves are the only winning ones (like when a piece moves to a corner square or it is sacrificed). The software is still in the very early phases of development but my goal is to write a tool similar to CQL which searches tablebases instead of pgn files.

For example, this is a position found by SEE:

The only winning move for white is a surprising move to the corner square: 1.Ka1! (tempo). Black is in zugzwang, although not a mutual one. White could reverse its key by moving back to b2.

Making a full search in the tablebase is very time-consuming. Instead, I did searches by adding some constraints to the position of pieces. The following mutual zugzwang was found by restricting the knights to the long diagonal. It is interesting to see how two knights are immobilised in opposite corners:

The next position is a mutual zugzwang unless black has still kept the right to castle! If black can still castle, it saves the draw by 1...0-0! This position was found by restricting the black king to e8 and the black rook to h8.

An interesting mutual zugzwang:

There are 13 mutual zugzwangs with the same configuration of pieces in the upper left corner (with the same material):

rQ6/P2k4/8/8/8/5q2/4R3/3K4 w - - 0 1
rQ6/P7/8/8/R7/2q4k/8/3K4 w - - 0 1
rQ6/P7/8/q7/1R6/8/8/k3K3 w - - 0 1
rQ6/P2k4/5q2/4R3/8/8/1K6/8 w - - 0 1
rQ6/P2k4/8/4R3/8/5q2/2K5/8 w - - 0 1
rQ6/P2k4/2q5/4R3/8/1K6/8/8 w - - 0 1
rQ6/P7/8/3q4/R7/4K2k/8/8 w - - 0 1
rQ6/P4R2/8/8/4K3/8/6k1/7q w - - 0 1
rQ6/P7/4R1K1/8/6k1/8/6q1/8 w - - 0 1
rQ6/P7/4R2K/8/8/5q1k/8/8 w - - 0 1
rQ6/P3K3/2q5/4R3/k7/8/8/8 w - - 0 1
rQ6/P3K3/2q5/4R3/2k5/8/8/8 w - - 0 1
rQ6/P2k2K1/8/3q4/6R1/8/8/8 w - - 0 1

My underpromotion code also seems to work. A sample position:

1. c8=R!! 1. c8=Q? Ra2+! 2. Kxa2 Qxb3+ 3. Ka1 Qc3+! 4. Qxc3 stalemate 1... Rxc8 1... Ra2+ 2. Kxa2 Qxb3+ 3. Ka1 Qc3+ 4. Rxc3 +- 2. Qa6+ Kxb3 3. Qa2+ +-

Study 151

Árpád Rusz
Internet (RCS blog)
2018

White wins

The stipulation seems wrong: shouldn't white only fight for a draw?
1. Ba5+ Kc8 1... Kd7 2. Nf6+ +- 2. Nc5! 2. Nf6? Qg7 3. Bf3 Qc7! 4. Bg4+ Kd8 = 2... Qg7 3. Bf3! Thematic try: 3. Be4? Qf7 4. Bg2 Qh7 5. Bf3 Qg7 mutual zugzwang 6. Be4 Qf7 positional draw


mutual zugzwang

Now the black queen cannot move: it has to keep an eye on both the b7 and g4 squares (Focal Theme). The knight moves lead also to a disaster:

a. 3... Nf7 4. Bb7#
b. 3... Ng6 4. Bg4+ +-

Watch this study on a dynamic board! Click here!

This study features one of the first mutual zugzwangs found by Syzygy Endgame Explorer (SEE). More details in the next post.

Friday, 1 June 2018

Study 150

Árpád Rusz
Jirtdan Ty
2018
Special Commendation

White wins

1. d4 Nf3 1... Ne2 2. d5 Nc3 3. d6! Nxa4 4. d7 +- 2. d5 Ne5 2... Nd4 3. Kf6 +- 3. d6 3. Ke6? Nc4 = 3... Kg5 In order to win, white has to reach the same position but with black to move.


cyclic zugzwang - WTM

4. Ke6 Nd3 5. Bb5 5. d7? Nc5+ =; 5. Bc2? Nc5+ 6. Kd5 Nd7 = 5... Nf4+ 6. Kf7! Thematic try: It is too early for 6. Kd7?! (Position X') 6... Ng6 7. Bd3? Ne5+ 8. Ke6 Nxd3 9. d7 Nc5+ = 6... Ng6 7. Ba4! 7. d7? Ne5+ =; 7. Bc6? Ne5+ = 7... Ne5+ 7... Kf5 8. Bc2+ +- 8. Ke7! White has managed to transfer the move to black. While the white king made a triangle manoeuvre (Ke7-e6-f7-e7, 3 moves) and the bishop a switchback (Ba4-b5-a4, 2 moves), the black knight walked on a diamond-shaped path (Ne5-d3-f4-g6-e5, 4 moves). These manoeuvres result in a tempo loss by white.


cyclic zugzwang - BTM

8... Ng6+ 8... Kf4 9. Kf6! (9. Ke6? Nd3 positional draw 10. Kd5 Ne5 11. Ke6 Nd3 positional draw) 9... Ke4 10. Ke6 Nd3 11. Bc2 +- pin 9. Ke6 Nf4+

10. Kd7! (Position X) 10... Kf6 10... Ng6 11. Bc2 Ne5+ 12. Ke6 +- Now d3 is not available for the knight. Please, compare it with the thematic try. 11. Kc8! Nd5 12. Bb3! 12. d7? Nb6+ = 12... Ke5! 12... Ke6 13. d7+- pin 13. Kd7! Switchback 13... Nf6+ 14. Ke7 Kf5 15. Ba2(c4,f7) tempo 15... Ke5 16. Be6 +-

Watch this study on a dynamic board! Click here!

Related study: Study 73