Page 6 of 6
Re: Funny mate in 16!
Posted: Thu Feb 18, 2016 1:44 pm
by bnst
[d]8/1p1p1p1p/3P1P2/pp5P/kp6/1p4P1/1P4P1/2K5 w - - 0 1
Now this is a well known theme and a computer solves it in no time. It is fun though to solve it yourself since it has some tricky variations. But the really funny part is that if you decide that only knight promotions are legal you can solve it in one move less
Best regards
Andreas
Re: Funny mate in 6
Posted: Thu Feb 18, 2016 1:47 pm
by lech
Dann Corbit wrote:Jesús Muñoz made a nice summary here:
http://www.talkchess.com/forum/viewtopi ... 52&t=59276
SOLVED THE POSITION (10 engines):
As far as solvers, first the dedicated mate finders (not surprising):
Code: Select all
Seconds Engine
0.009 ChestUCI (version?)
0.330 SF Matefinder (by Ellis)
1.000 Chest 3.19 (command line version) ---> the exact solving time is not reported
1.321 SF Matefinder (other version)
then the general purpose engines:
Code: Select all
Seconds Engine
0.424 Houdini 4 Pro (Tactical Mode)
0.476 Critter 1.6a
1.930 Sting 6
2.020 Houdini 4 Pro
7.657 Strelka 6
49.625 Gull 3
I only gave a minute each, so I guess others can solve it.
I notice that all of the solvers except sting are at least influenced by Ivanhoe or Rybka family tree. I did not study Sting enough to know if there are any Ivanhoe/Rybka techniques in it like the huge material imbalance table.
Sting uses my original ideas only, with exception that is based on Stockfish 2.1.1 and it doesn't help in question to solve any tough positions.
Re: Funny mate in 16!
Posted: Thu Feb 18, 2016 2:03 pm
by zullil
bnst wrote:[d]8/1p1p1p1p/3P1P2/pp5P/kp6/1p4P1/1P4P1/2K5 w - - 0 1
If you decide that only knight promotions are legal you can solve it in one move less
Best regards
Andreas
Cute observation. White can simply promote to a knight for a third time and then ignore the black passer, since it can only promote without check.
Re: Funny mate in 6
Posted: Thu Feb 18, 2016 2:05 pm
by rabbits23
Chessmaster 10 announces mates fairly quickly
Allan
Time Depth Score Positions Moves
1:24 7/16 Mate06 98863429 1.ˆh1 d4 2.ˆa1 d3 3.ˆa3 bxa3+ 4.‹a1
a2 5.†a3 ‹xf6 6.†b2#
1:30 8/17 Mate06 105144356 1.ˆh1 d4 2.ˆa1 d3 3.ˆa3 bxa3+ 4.‹a1
a2 5.†a3 ‹xf6 6.†b2#
1:33 9/18 Mate06 108103379 1.ˆh1 d4 2.ˆa1 d3 3.ˆa3 bxa3+ 4.‹a1
a2 5.†a3 ‹xf6 6.†b2#
1:35 10/19 Mate06 109936361 1.ˆh1 d4 2.ˆa1 d3 3.ˆa3 bxa3+ 4.‹a1
a2 5.†a3 ‹xf6 6.†b2#
1:38 11/20 Mate06 111521713 1.ˆh1 d4 2.ˆa1 d3 3.ˆa3 bxa3+ 4.‹a1
a2 5.†a3 ‹xf6 6.†b2#
Re: Funny mate in 6
Posted: Thu Feb 18, 2016 2:47 pm
by mar
As usual, it's null-move related.
Disabling nmp makes this a matter of 1-2 seconds, there's a reason why I have a nullmove checkbox in engine params, no need for fake lobotomized SF version
Re: Funny mate in 16!
Posted: Thu Feb 18, 2016 2:55 pm
by bnst
zullil wrote:bnst wrote:[d]8/1p1p1p1p/3P1P2/pp5P/kp6/1p4P1/1P4P1/2K5 w - - 0 1
If you decide that only knight promotions are legal you can solve it in one move less
Best regards
Andreas
Cute observation. White can simply promote to a knight for a third time and then ignore the black passer, since it can only promote without check.
And the computers won't help you with that one
Re: Funny mate in 6
Posted: Thu Feb 18, 2016 2:56 pm
by mar
mar wrote:no need for fake lobotomized SF version
To make myself clear: I don't mean Sting, it's an awesome project and I use it to analyze difficult positions, I guess it's obvious now what I mean (those who fork latest SF and make random 10 line changes)
Re: Funny mate in 16!
Posted: Thu Feb 18, 2016 3:14 pm
by BBauer
My modified SF
Code: Select all
FEN: 8/1p1p1p1p/3P1P2/pp5P/kp6/1p4P1/1P4P1/2K5 w - - 0 1
St-mod-16-02-08:
8/8 00:00 473 158k +5,82 g3-g4 h7-h6 g4-g5 b7-b6 g5-g6 f7xg6 f6-f7 g6xh5
9/9 00:00 699 233k +5,82 g3-g4 h7-h6 g4-g5 b7-b6 g5-g6 f7xg6 f6-f7 g6xh5 f7-f8Q
10/15 00:00 4k 902k +5,53 g3-g4 h7-h6 g4-g5 h6xg5 h5-h6 g5-g4 h6-h7 b7-b6 h7-h8N g4-g3 Nh8-g6 f7xg6 f6-f7
11/18 00:00 7k 1.129k +5,45 g3-g4 h7-h6 g4-g5 h6xg5 h5-h6 g5-g4 h6-h7 b7-b6 h7-h8N g4-g3 Nh8-g6 f7xg6 Kc1-b1
12/18 00:00 9k 1.256k +0,14 g3-g4 h7-h6 g4-g5 h6xg5 h5-h6 g5-g4 h6-h7 b7-b6 h7-h8N g4-g3 Nh8-g6 f7xg6 f6-f7 g6-g5 f7-f8N g5-g4
13/18 00:00 9k 1.343k +4,72 g3-g4 h7-h6 g4-g5 h6xg5 h5-h6 g5-g4 h6-h7 b7-b6 h7-h8N g4-g3 Nh8-g6 f7xg6 f6-f7 g6-g5 f7-f8N g5-g4 Nf8-e6
14/21 00:00 10k 1.386k +4,84 g3-g4 h7-h6 g4-g5 h6xg5 h5-h6 g5-g4 h6-h7 b7-b6 h7-h8N g4-g3 Nh8-g6 f7xg6 f6-f7 g6-g5 f7-f8N g5-g4 Nf8-e6 d7xe6 d6-d7
15/21 00:00 10k 1.265k +5,62 g3-g4 h7-h6 g4-g5 h6xg5 h5-h6 g5-g4 h6-h7 b7-b6 h7-h8N g4-g3 Nh8-g6 f7xg6 f6-f7 g6-g5 f7-f8N g5-g4 Nf8-e6 d7xe6 d6-d7 e6-e5 d7-d8Q
16/26 00:00 11k 1.222k +6,56 g3-g4 h7-h6 g4-g5 h6xg5 h5-h6 g5-g4 h6-h7 b7-b6 h7-h8N g4-g3 Nh8-g6 f7xg6 f6-f7 g6-g5 f7-f8N g5-g4 Nf8-e6 d7xe6 d6-d7 e6-e5 d7-d8Q e5-e4 Qd8xb6
17/29 00:00 12k 1.155k +7,01 g3-g4 h7-h6 g4-g5 h6xg5 h5-h6 g5-g4 h6-h7 b7-b6 h7-h8N g4-g3 Nh8-g6 f7xg6 f6-f7 g6-g5 f7-f8N g5-g4 Nf8-e6 d7xe6 d6-d7 e6-e5 d7-d8Q e5-e4 Qd8xb6 e4-e3 Qb6-c7
18/32 00:00 20k 1.123k +M16 g3-g4 h7-h6 g4-g5 h6xg5 h5-h6 g5-g4 h6-h7 b7-b6 h7-h8N g4-g3 Nh8-g6 f7xg6 f6-f7 g6-g5 f7-f8N g5-g4 Nf8-e6 d7xe6 d6-d7
19/32 00:00 20k 1.138k +M16 g3-g4 h7-h6 g4-g5 h6xg5 h5-h6 g5-g4 h6-h7 b7-b6 h7-h8N g4-g3 Nh8-g6 f7xg6 f6-f7 g6-g5 f7-f8N g5-g4 Nf8-e6 d7xe6 d6-d7
20/32 00:00 21k 1.097k +M16 g3-g4 h7-h6 g4-g5 h6xg5 h5-h6 g5-g4 h6-h7 b7-b6 h7-h8N g4-g3 Nh8-g6 f7xg6 f6-f7 g6-g5 f7-f8N g5-g4 Nf8-e6 d7xe6 d6-d7 e6-e5
21/32 00:00 21k 1.115k +M16 g3-g4 h7-h6 g4-g5 h6xg5 h5-h6 g5-g4 h6-h7 b7-b6 h7-h8N g4-g3 Nh8-g6 f7xg6 f6-f7 g6-g5 f7-f8N g5-g4 Nf8-e6 d7xe6 d6-d7 e6-e5 d7-d8Q
22/32 00:00 22k 1.133k +M16 g3-g4 h7-h6 g4-g5 h6xg5 h5-h6 g5-g4 h6-h7 b7-b6 h7-h8N g4-g3 Nh8-g6 f7xg6 f6-f7 g6-g5 f7-f8N g5-g4 Nf8-e6 d7xe6 d6-d7 e6-e5 d7-d8Q e5-e4
23/32 00:00 22k 1.117k +M16 g3-g4 h7-h6 g4-g5 h6xg5 h5-h6 g5-g4 h6-h7 b7-b6 h7-h8N g4-g3 Nh8-g6 f7xg6 f6-f7 g6-g5 f7-f8N g5-g4 Nf8-e6 d7xe6 d6-d7 e6-e5 d7-d8Q e5-e4 Qd8-h4
So the computer helps
Kind regards
Bernhard
Re: Funny mate in 16!
Posted: Thu Feb 18, 2016 3:20 pm
by bnst
It solves the mate in 16 easily but I was refering to the solution in one move less when you add the limitation that only knight promotions are legal. I modified my own program to solve it that way though so in a way you are right.
Regards
Andreas
Re: Funny mate in 6
Posted: Fri Feb 19, 2016 12:34 am
by MikeB
Current Developmental Crafty 25.1a M09 finds almost instantly
Code: Select all
Black(1): n1N3br/2p1Bpkr/1pP2R1b/pP1p1PpR/Pp4P1/1P6/1K1P4/8 w - - 0 1
White(1): go
time surplus 0.00 time limit 15.00 (15.00)
depth time score variation (1)
starting thread 1 2 3 4 5 6 7 8 9 10 11 <done>
28-> 1.03/15.00 6.90 1. Rh1 d4 2. Nd6 cxd6 3. Rxd6 f6 4. Bxf6+
Kf7 5. Bxd4 Nc7 6. Rd7+ Ke8 7. Bxh8 Rxd7
8. cxd7+ Kxd7 9. Rxh6 Nd5 10. Bd4 Nf4
11. Bxb6 Nd3+ 12. Kc2 Ne1+ 13. Kb1 Bxb3
14. Rh7+ Kd6 15. Bxa5 Nc2 16. b6 Bxa4
17. b7 Na3+ 18. Ka2
29 1.60/15.00 6.39 1. Rh1 d4 2. Nd6 cxd6 3. Rxd6 f6 4. Bxf6+
Kf7 5. Bxd4 Nc7 6. Rd7+ Ke8 7. Bxh8 Rxd7
8. cxd7+ Kxd7 9. Rxh6 Nd5 10. Bd4 Nf4
11. Bxb6 Nd3+ 12. Kc2 Ne1+ 13. Kb1 Bxb3
14. Rh7+ Kc8 15. Rc7+ Kb8 16. f6 Nf3
17. f7 Nxd2+ 18. Kc1 Bxf7 19. Rxf7
29-> 1.70/15.00 6.39 1. Rh1 d4 2. Nd6 cxd6 3. Rxd6 f6 4. Bxf6+
Kf7 5. Bxd4 Nc7 6. Rd7+ Ke8 7. Bxh8 Rxd7
8. cxd7+ Kxd7 9. Rxh6 Nd5 10. Bd4 Nf4
11. Bxb6 Nd3+ 12. Kc2 Ne1+ 13. Kb1 Bxb3
14. Rh7+ Kc8 15. Rc7+ Kb8 16. f6 Nf3
17. f7 Nxd2+ 18. Kc1 Bxf7 19. Rxf7
30 2.72/15.00 8.68 1. Rh1 d4 2. Re1 d3 3. Nd6 cxd6 4. Rxd6 f6
5. Bxf6+ Kf8 6. Rd8+ Kf7 7. Bxh8 Rxh8
8. Rxa8 Bg7+ 9. Kb1 Be5 10. Rxe5 Rh1+
11. Kb2 Rd1 12. Rd5 Rxd2+ 13. Kc1 Rc2+
14. Kd1 Kg7 15. Rd7+ Bf7 16. Raa7 Kf6
17. Rxf7+ Ke5 18. Rad7 Rc3 19. Rfe7+ Kf4
20. Kd2
30-> 2.83/15.00 8.68 1. Rh1 d4 2. Re1 d3 3. Nd6 cxd6 4. Rxd6 f6
5. Bxf6+ Kf8 6. Rd8+ Kf7 7. Bxh8 Rxh8
8. Rxa8 Bg7+ 9. Kb1 Be5 10. Rxe5 Rh1+
11. Kb2 Rd1 12. Rd5 Rxd2+ 13. Kc1 Rc2+
14. Kd1 Kg7 15. Rd7+ Bf7 16. Raa7 Kf6
17. Rxf7+ Ke5 18. Rad7 Rc3 19. Rfe7+ Kf4
20. Kd2
31 3.06/15.00 9.12 1. Rh1 d4 2. Re1 d3 3. Nd6 cxd6 4. Rxd6 f6
5. Bxf6+ Kf8 6. Rd8+ Kf7 7. Bxh8 Rxh8
8. Rxa8 Bg7+ 9. Kb1 Be5 10. Rxe5 Rh1+
11. Kb2 Rd1 12. Rd5 Rxd2+ 13. Kc1 Rc2+
14. Kd1 Kg7 15. Rd7+ Bf7 16. Raa7 Kf6
17. Rxf7+ Ke5 18. Rad7 Rc5 19. Rxd3 Kf4
20. Rd4+ Ke5
31-> 3.23/15.00 9.12 1. Rh1 d4 2. Re1 d3 3. Nd6 cxd6 4. Rxd6 f6
5. Bxf6+ Kf8 6. Rd8+ Kf7 7. Bxh8 Rxh8
8. Rxa8 Bg7+ 9. Kb1 Be5 10. Rxe5 Rh1+
11. Kb2 Rd1 12. Rd5 Rxd2+ 13. Kc1 Rc2+
14. Kd1 Kg7 15. Rd7+ Bf7 16. Raa7 Kf6
17. Rxf7+ Ke5 18. Rad7 Rc5 19. Rxd3 Kf4
20. Rd4+ Ke5
32 3.74/15.00 Mat08 1. Rh1 d4 2. Rc1 d3 3. Nd6 cxd6 4. c7 Nxc7
5. Rxc7 d5 6. Rxb6 f6 7. Bxf6+ Kf8 8. Rb8#
32-> 3.77/15.00 Mat08 1. Rh1 d4 2. Rc1 d3 3. Nd6 cxd6 4. c7 Nxc7
5. Rxc7 d5 6. Rxb6 f6 7. Bxf6+ Kf8 8. Rb8#
time=3.77(71%) nodes=173140793(173.1M) fh1=86% pred=0 nps=45.9M
chk=6.7M qchk=10.3M fp=73.6M mcp=4.6M 50move=1
LMReductions: 1/2.0M 2/1.4M 3/1.3M 4/727.7K 5/160.5K 6/13.3K
7/472
null-move (R): 3/6.0M 4/1.1M 5/74.5K 6/5.1K 7/205 8/1
splits=119.5K(58.5K) aborts=8.5K joins=318.7K data=25%(31%)
mate in 8 moves.