13 hard test positions July 2016

Discussion of anything and everything relating to chess playing software and machines.

Moderator: Ras

Vinvin
Posts: 5287
Joined: Thu Mar 09, 2006 9:40 am
Full name: Vincent Lejeune

13 hard test positions July 2016

Post by Vinvin »

Some new positions from a couple of months, verified but may be not enough :

1)
b4!
8 
7 
6 
5 
4 
3 
2 
1 
abcdefgh

2kr3r/1p3pb1/pBnqp2p/P2p2p1/8/2PB4/1P3PPP/R2Q1RK1 w - - bm b4;

2)
Winning moves : 1.a8=Q or =R



3)
How much time to find 16. Bg5! (after 15...d5?) with a very good advantage ( >0.50) ?
From the match Polgar - Spassky (1993) Round 8.

8 
7 
6 
5 
4 
3 
2 
1 
abcdefgh

r2qrbk1/1bpn1p1p/p4np1/1p1pp3/3PP3/1PP2NNP/P1B2PP1/R1BQR1K1 w - - 0 16



4)
From the Chessbase site, a position with a lot of zugzwang :


Sting-SF 7 finds the solution in 20 to 30 sec :

Code: Select all

FEN: q7/7R/k7/1p6/1p6/1P2B3/7K/8 w - - 0 1

Sting-sf-7-x64:
...
 44	00:29	 383 996k	12 909k	 0,00	1.Bd4 Qb8+ 2.Kh1 Qa8+ 3.Kg1 Qg8+ 4.Rg7 Qa8 5.Be3 Qb8 6.Bc5 Qa8 7.Rd7 Qg8+ 8.Kf1 Qb8 9.Kg2 Qg8+ 10.Kf1
 45+	00:30	 402 134k	13 031k	+0,08	1.Bd4
 45+	00:30	 402 151k	13 030k	+0,16	1.Bd4
 45+	00:31	 407 159k	13 023k	+13,77	1.Bd4
 45	00:32	 419 890k	13 021k	+12,52	1.Bd4 Qb8+ 2.Kh1 Qa8+ 3.Kg1 Qg8+ 4.Rg7 Qa8 5.Kf2 Qb8 6.Kf3 Qf8+ 7.Ke4 Qa8+ 8.Ke3 Qb8 9.Kd3 Qa8 10.Rf7 Qb8 11.Kc2 Qh2+ 12.Kb1 Qh1+ 13.Ka2 Qg2+ 14.Bf2 Qa8 15.Rg7 Qb8 16.Bg1 Qa8 17.Bd4 Qb8 18.Rf7 Qh2+ 19.Kb1
 46+	00:32	 420 989k	13 020k	+12,64	1.Bd4
 46+	00:32	 421 070k	13 018k	+12,76	1.Bd4
 46+	00:34	 449 832k	13 086k	+47,98	1.Bd4
 46	00:35	 459 170k	13 100k	+14,86	1.Bd4 Qb8+ 2.Kh1 Qa8+ 3.Kg1 Qg8+ 4.Rg7 Qa8 5.Kf2 Qb8 6.Ke3 Qe8+ 7.Kd3 Qa8 8.Rf7 Qb8 9.Kc2 Qh2+ 10.Bf2 Qb8 11.Kb1 Qa8 12.Ka2 Qb8 13.Rh7 Qa8 14.Kb2 Qb8 15.Bd4 Qa8 16.Ka1 Qb8 17.Rh6+ Ka5 18.Rh8 Qxh8 19.Bxh8 Kb6 20.Kb2 Kc6 21.Kc2 Kd5 22.Kd3 Ke6 23.Ke4 Kf7 24.Be5 Ke7 25.Kd5 Kd7 26.Bf4 Ke8 27.Ke4 Kf7 28.Kd4 Kg6 29.Kc5 Kf5 30.Bd6 Ke6 31.Kxb4 Kd7 32.Kxb5 Ke6 33.b4 Kd7 34.Ka6 Ke6 35.b5 Kf7 36.b6 Kg8 37.b7
 47+	00:35	 459 523k	13 096k	+14,98	1.Bd4
 47+	00:35	 459 656k	13 094k	+15,11	1.Bd4
 47+	00:35	 460 274k	13 075k	+61,43	1.Bd4
 47	00:35	 465 035k	13 058k	+77,08	1.Bd4 Qb8+ 2.Kh1 Qa8+ 3.Kg1 Qg8+ 4.Rg7 Qa8 5.Kf2 Qb8 6.Ke3 Qa8 7.Kd2 Qb8 8.Kd3 Qa8 9.Rf7 Qb8 10.Kc2 Qh2+ 11.Bf2 Qb8 12.Kb1 Qa8 13.Ka2 Qb8 14.Rh7 Qa8 15.Kb2 Qb8 16.Bd4 Qa8 17.Ka1 Qb8 18.Rh6+ Ka5 19.Rh8 Qxh8 20.Bxh8 Kb6 21.Kb2 
5)
Zugzwang too : 1B1k4/3p4/1Kp5/PpP2p2/1PqP1P2/8/1P6/8 w - - bm a6;


6)
Stalemate theme, finds a draw score :


7)
Mate with the knight after some zugzwang


8)
Let's block the black king than bring the white king :


9)
Black is winning after 1...Bxh4! : 8/p7/5k2/pPp1p3/2P3Kb/8/8/3B4 w - - 0 2


10)
Draw combination by stalemate :


11)
Rxf5!
8 
7 
6 
5 
4 
3 
2 
1 
abcdefgh

4rb2/1p1q1ppk/p1n1p2p/2p1Pn2/3PNR2/2P2K2/P2QBP2/6R1 w - - bm Rxf5;

12)
White is winning after 1. Bxh7+! Kxh7 2. Ng5+! Kg8 3. Qe2! g6 4. Rh3!
8 
7 
6 
5 
4 
3 
2 
1 
abcdefgh

r1b2rk1/2q1bppp/pp2p3/2npP3/1n3P1P/2NBBN2/PPPQ2P1/1K1R3R w - - bm Bxh7;

13)
2 possible winning moves here : Qxd2! or Ng4!, hard for Stockfish.
8 
7 
6 
5 
4 
3 
2 
1 
abcdefgh

r4rk1/pb3pp1/1p1qpR2/2ppN3/3P4/2P4P/PP1pQ1P1/3R2K1 w - - 0 19
User avatar
Nordlandia
Posts: 2822
Joined: Fri Sep 25, 2015 9:38 pm
Location: Sortland, Norway

Re: 13 hard test positions July 2016

Post by Nordlandia »


Code: Select all

[Event "5.c Farago MT Revista de Romana de Sah "]
[Site "?"]
[Date "1971.??.??"]
[Round "?"]
[White "Ruszczynski=P"]
[Black "(=0323.11e1d4)"]
[Result "1/2-1/2"]
[SetUp "1"]
[FEN "8/4B3/4p3/8/2rk1P2/2n5/2B5/4K3 w - - 0 1"]
[PlyCount "19"]
[EventDate "1971.??.??"]

{Revista de Romana de Sah/11 1973 EG#02122 @3: Ulrichsen=J EG=175 1/2009} 1. f5
Nd5 2. Bb3 Rc7 3. Bxd5 exf5 $1 (3... exd5 4. f6 Ke5 (4... Ke3 5. Kd1) 5. f7) 4.
Bd8 (4. Bd6 $2 Rd7 5. Be6 Rxd6 6. Bxf5 Ke3 7. Bh3 Rh6 8. Bg2 Ra6 $1) 4... Rd7
5. Be6 Rxd8 6. Bxf5 Ke3 $1 7. Bh3 $1 Rh8 (7... Rd2 $1 {<cook JU>}) 8. Bg2 Rd8
9. Bh3 Rh8 10. Bg2 1/2-1/2


Code: Select all

[Event "3.hm New Statesman {and Nation}#82 @3"]
[Site "?"]
[Date "1973.??.??"]
[Round "?"]
[White "Kovalenko=V"]
[Black "(+0170.11g8d3)"]
[Result "1-0"]
[SetUp "1"]
[FEN "6K1/4p3/3b4/8/b7/2Bk4/7P/2R5 w - - 0 1"]
[PlyCount "17"]
[EventDate "1973.??.??"]

{EG#02248 @3: Veitch=W EG=41 7/1975} 1. h4 Bb3+ (1... Ba3 2. Ra1 Bb3+ 3. Kg7
Kxc3 4. Rxa3 e5 5. h5 Kb2 6. Rxb3+ Kxb3 7. h6 e4 8. h7 e3 9. h8=Q) (1... Bf4 $1
{<cook WV>} 2. Rf1 e5) 2. Kg7 (2. Kf8 $2 e5+ 3. Kg7 Ba3 4. Ra1 Kxc3 5. Rxa3 e4
6. h5 e3 7. h6 e2 8. Ra1 Bc2) (2. Kh8 $2 Ba3 3. Ra1 Kxc3 4. Rxa3 e5 5. h5 (5.
Ra5 e4 6. Re5 Bc2 7. h5 Kd2 8. h6 e3) 5... e4 6. h6 e3 7. h7 e2 8. Ra1 Bc2 9.
Re1 Kd2) 2... Ba3 (2... Bf4 3. Rf1 Kxc3 4. Rxf4 Bc2 5. h5 e5 6. Rf3+ Kd2 (6...
Kd4 7. h6 Be4 8. Rg3 Bc2 9. Kf6 Bh7 10. Rg7) 7. h6 Be4 8. Rg3) 3. Ra1 Bd6 4.
Bb2 Kc2 5. Bd4 Kd3 6. Rb1 (6. Bf2 $2 Be5+) 6... Bc2 (6... Kxd4 7. Rxb3 e5 8.
Rb6 Bc7 (8... Kc5 9. Rb1 e4 10. h5 e3 11. Kg6 Bg3 (11... Bf8 12. Kf5 Kd4 13.
Kf4 Kd3 14. Kf3 Bh6 15. Rd1+) (11... Kd4 12. Rd1+ Ke5 13. h6) (11... Kc4 12.
Rc1+ Kb3 13. h6) 12. h6 e2 13. h7) 9. Rc6 Bb8 10. h5 e4 11. Re6 Be5+ 12. Rxe5
Kxe5 13. h6 e3 14. h7 e2 15. h8=Q e1=Q 16. Qe8+) 7. Rc1 (7. Rb2 $2 Ba4 8. Ra2
Bb3 9. Rb2 Ba4 10. Rg2 Bc6 11. Rg4 Bf3) 7... Ba3 (7... Kxd4 8. Rxc2 e5 9. Rd2+
Kc5 10. h5 e4 11. Kg6 e3 12. Rd1) 8. Ra1 Bd6 (8... Kxd4 9. Rxa3 e5 10. Kf6 e4
11. Kg5 e3 12. Kf4 e2 13. Re3) 9. Bb2 1-0

peter
Posts: 3393
Joined: Sat Feb 16, 2008 7:38 am
Full name: Peter Martan

Re: 13 hard test positions July 2016

Post by peter »

Vinvin wrote: 5)
Zugzwang too : 1B1k4/3p4/1Kp5/PpP2p2/1PqP1P2/8/1P6/8 w - - bm a6;
Very tricky!
Found it in YACPDB
http://www.yacpdb.org/#search/MUIxazQzc ... MS8xLzA=/1
with a slightly different line of solution from O.Bláthy, in which the longer way with one more black pawn- move is shown.
1. a6 Kc8 2. a7 Qa2 3. Bd6 Qa4 4. Be7 Qa2 5. Bh4 Qa4
6. b3 Qa3 7. Bg3 Qa2 8. Bf2 Qa1 9. Be1 Qa2 10. Bd2 Qa1
11. Bc1 d5 12. Be3
In this one (or in your line above by 9...d5 instead of Qa2), you can fool even engines with switched off or very little nullmove by default like komodo-NM-off or Sting for quite a while yet.
Even if the eval already remains a winning one, much to much shuffling still then.
:)
Peter.
Vinvin
Posts: 5287
Joined: Thu Mar 09, 2006 9:40 am
Full name: Vincent Lejeune

Re: 13 hard test positions July 2016

Post by Vinvin »

peter wrote:
Vinvin wrote: 5)
Zugzwang too : 1B1k4/3p4/1Kp5/PpP2p2/1PqP1P2/8/1P6/8 w - - bm a6;
Very tricky!
Found it in YACPDB
http://www.yacpdb.org/#search/MUIxazQzc ... MS8xLzA=/1
with a slightly different line of solution from O.Bláthy, in which the longer way with one more black pawn- move is shown.
1. a6 Kc8 2. a7 Qa2 3. Bd6 Qa4 4. Be7 Qa2 5. Bh4 Qa4
6. b3 Qa3 7. Bg3 Qa2 8. Bf2 Qa1 9. Be1 Qa2 10. Bd2 Qa1
11. Bc1 d5 12. Be3
In this one (or in your line above by 9...d5 instead of Qa2), you can fool even engines with switched off or very little nullmove by default like komodo-NM-off or Sting for quite a while yet.
Even if the eval already remains a winning one, much to much shuffling still then.
:)
Yes, I didn't chose the more exact line but the most showy way.
But the ideas still : when the queen goes on a2, the bishop has to go on d2.
Then, if Qa3 -> Bc3 , controlling a1, forces Qa2, then Bb2 prevents Qa1 and Qa3, so white wins !
And, if Qa1 -> Bc1 , controlling a3, forces Qa2, then Bb2 prevents Qa1 and Qa3, so white wins !
peter
Posts: 3393
Joined: Sat Feb 16, 2008 7:38 am
Full name: Peter Martan

Re: 13 hard test positions July 2016

Post by peter »

Vinvin wrote:[Yes, I didn't chose the more exact line but the most showy way.
But the ideas still : when the queen goes on a2, the bishop has to go on d2.
Then, if Qa3 -> Bc3 , controlling a1, forces Qa2, then Bb2 prevents Qa1 and Qa3, so white wins !
And, if Qa1 -> Bc1 , controlling a3, forces Qa2, then Bb2 prevents Qa1 and Qa3, so white wins !
Of course, but the most fun against the engines with the one more black pawn-move is to see them fail again and again to find the shortest triangle-maneuver of the white Bishop to get the black Queen in the right moment at the right square before starting the final cut-off. The minimum-example of Zugzwang, going triangle, normally with King, here with Bishop, still a weakness of all engines.
:)
Peter.
MikeGL
Posts: 1010
Joined: Thu Sep 01, 2011 2:49 pm

Re: 13 hard test positions July 2016

Post by MikeGL »

Thanks for this puzzle compilation.

I haven't checked with other engines but critter1.6 32bit seems to solve
one of these difficult puzzles in less than 10 seconds on my old machine.


8 
7 
6 
5 
4 
3 
2 
1 
abcdefgh

4rb2/1p1q1ppk/p1n1p2p/2p1Pn2/3PNR2/2P2K2/P2QBP2/6R1 w - - bm Rxf5;


1.Rxf5
first run found it in 3 seconds.
second run 11 seconds.
third run in 5 seconds.

critter 1.6 32bit
OS 32bit, dual core, hash 256MB
Dann Corbit
Posts: 12777
Joined: Wed Mar 08, 2006 8:57 pm
Location: Redmond, WA USA

Re: 13 hard test positions July 2016

Post by Dann Corbit »

What score did Critter give to the position?
Taking ideas is not a vice, it is a virtue. We have another word for this. It is called learning.
But sharing ideas is an even greater virtue. We have another word for this. It is called teaching.
MikeGL
Posts: 1010
Joined: Thu Sep 01, 2011 2:49 pm

Re: 13 hard test positions July 2016

Post by MikeGL »

Dann Corbit wrote:What score did Critter give to the position?
I tried to run it again just to get a screen grab. Below are the two runs which i did just now.
I am sure it can be easily replicated by anyone with a copy of critter1.6

4th run (with screengrab):
Image



5th run (with screengrab):
Image
Vinvin
Posts: 5287
Joined: Thu Mar 09, 2006 9:40 am
Full name: Vincent Lejeune

Re: 13 hard test positions July 2016

Post by Vinvin »

MikeGL wrote:Thanks for this puzzle compilation.

I haven't checked with other engines but critter1.6 32bit seems to solve
one of these difficult puzzles in less than 10 seconds on my old machine.


8 
7 
6 
5 
4 
3 
2 
1 
abcdefgh

4rb2/1p1q1ppk/p1n1p2p/2p1Pn2/3PNR2/2P2K2/P2QBP2/6R1 w - - bm Rxf5;


1.Rxf5
first run found it in 3 seconds.
second run 11 seconds.
third run in 5 seconds.

critter 1.6 32bit
OS 32bit, dual core, hash 256MB
Yes, a lot of engines solve this quickly, mainly Stockfish has trouble here.
peter
Posts: 3393
Joined: Sat Feb 16, 2008 7:38 am
Full name: Peter Martan

Re: 13 hard test positions July 2016

Post by peter »

Vinvin wrote: 13)
2 possible winning moves here : Qxd2! or Ng4!, hard for Stockfish.
[d]r4rk1/pb3pp1/1p1qpR2/2ppN3/3P4/2P4P/PP1pQ1P1/3R2K1 w - - 0 19
Really not easy to say, which is better.
That's after some Forward- Backward of both lines in 2 MV and depth 38:

r4rk1/pb3pp1/1p1qpR2/2ppN3/3P4/2P4P/PP1pQ1P1/3R2K1 w - - 0 1

Analysis by Stockfish 240716 64 POPCNT:

1. +- (3.28): 1.Qxd2 Ba6 2.Ng4 Rfc8 3.Qg5 Kf8 4.Rf2 Rc7 5.Qg6 cxd4 6.cxd4 Rb8 7.Qh7 Ke8 8.Qxg7 Kd7 9.Rxf7+ Kc8 10.Nf6 Qd8 11.Rxc7+ Qxc7 12.Qg8+ Kb7 13.Qxe6 Rf8 14.Nxd5 Qf7 15.Qxf7+ Rxf7 16.Nc3 Kc7 17.g4 Kd6 18.Kg2 Bb7+ 19.d5 Bc8 20.Kg3 Rg7 21.Nb5+ Ke5 22.Rc1 a6 23.Rxc8 axb5 24.h4 Kxd5 25.h5

2. +- (2.27): 1.Ng4 Rfc8 2.Qxd2 Kf8 3.Rxf7+ Kxf7 4.Rf1+ Ke8 5.Qg5 Ba6 6.Qh5+ Kd7 7.Ne5+ Kc7 8.Rf7+ Kb8 9.Nd7+ Kc7 10.Nxc5+ Kc6 11.Nxa6 b5 12.Qe2 Kb6 13.Nc5 Rc7 14.Rxc7 Qxc7 15.Qxe6+ Qc6 16.Qf7 a5 17.Qxg7 Qc7 18.Qf6+ Qc6 19.Qe7 Qc7 20.Nd7+ Ka7 21.Qe6 Rd8 22.Nc5 Rd6 23.Qe5 b4 24.cxb4 axb4 25.g4 Rc6 26.Qe3 Qg7 27.Kg2 Rh6 28.a3 Rh7 29.axb4 Qf6 30.g5 Qf5 31.Qa3+ Kb8
Peter.