Definitely not Mate in 14

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

Moderators: hgm, Rebel, chrisw

User avatar
MikeB
Posts: 4889
Joined: Thu Mar 09, 2006 6:34 am
Location: Pen Argyl, Pennsylvania

Re: Definitely not Mate in 14

Post by MikeB »

Dann Corbit wrote:Komodo agrees:

Code: Select all

     29	24:20	22,730,483,441	15,563,173	+M13	Rg8+ Kf2 Rg2+ Kf1 h4 Rxe5+ Be6 Ra5 h3 Ra7+ Bd7 Ra8 h2 Rh8 Bb5+ Ke1 Nd3+ Kd1 Bd7 e5 Be6 Re8+ Kxe8 f4 h1Q+
SF-McB_3.1 (dev) says mate in 12:

Code: Select all

dep	score	nodes	time	(not shown:  tbhits	knps	seldep)
 49	 #-12 	2.85G	1:32.75	Rg8+ Kf2 Rg2+ Kf1 Kf6 Ra6+ Kg7 Rd6 Bb5+ Ke1 h4 b4 Nd3+ Rxd3 Bxd3 f4 exf4 e5 h3 e6 h2 exf7 h1=Q# 
 48	 #-12 	2.56G	1:23.89	Rg8+ Kf2 Rg2+ Kf1 Kf6 Ra6+ Kg7 Rd6 Bb5+ Ke1 h4 b4 Nd3+ Rxd3 Bxd3 f4 exf4 e5 h3 e6 h2 exf7 h1=Q# 
 47	 #-12 	2.12G	1:09.81	Rg8+ Kf2 Rg2+ Kf1 Kf6 Ra6+ Kg7 Rd6 Bb5+ Ke1 h4 b4 Nd3+ Rxd3 Bxd3 f4 exf4 e5 h3 e6 h2 exf7 h1=Q# 
 46	 #-12 	1.89G	1:02.26	Rg8+ Kf2 Rg2+ Kf1 Kf6 Ra6+ Kg7 Rd6 Bb5+ Ke1 h4 b4 Nd3+ Rxd3 Bxd3 f4 exf4 e5 h3 e6 h2 exf7 h1=Q# 
 45	 #-12 	1.68G	0:55.47	Rg8+ Kf2 Rg2+ Kf1 Kf6 Ra6+ Kg7 Rd6 Bb5+ Ke1 h4 b4 Nd3+ Rxd3 Bxd3 f4 exf4 e5 h3 e6 h2 exf7 h1=Q# 
 44	 #-15 	1.45G	0:47.97	Rg8+ Kf2 Rg2+ Kf1 Bh3 Ra7+ Kf8 Ra8+ Kg7 Ke1 Nd3+ Kd1 Rb2 Rd8 Bf1 Rxd3 Bxd3 b4 h4 f4 h3 f5 h2 f6+ Kxf6 Kc1 Rc2+ Kd1 h1=Q#
Top
Dann Corbit
Posts: 12662
Joined: Wed Mar 08, 2006 8:57 pm
Location: Redmond, WA USA

Re: Definitely not Mate in 14

Post by Dann Corbit »

Definitely a mate in 13. Chest output is pretty well a proof that there is not shorter solution. You can also ask for a proof tree, but with a mate in 13 it would be truly enormous.

Code: Select all

Analysis of Q:\dm26.epd   
2017-11-10 3:49:10 PM Level: 3600 Seconds
Analyzing engine: ChestUCI_AutoTurbo

 1 Rg8+;                
    Searching move: Rd8-g8
    Best move (ChestUCI_AutoTurbo): Rd8-g8
    Identical moves! Found in: 41:11
   FEN: 3r4/3bkp2/8/R3p2p/4Pn2/1P3PK1/8/8 b - -   (5+7)
   Position-Analysis:  C0/R0/K4/P3/X15   B:6/31
   AutoTurbo-Search for Special-Mate [C1/R0/K0/P0/X0] in 13 ...  (Hash=2044MB)
   AutoTurbo-Search for Special-Mate [C0/R1/K0/P0/X0] in 13 ...  (Hash=2044MB)
   AutoTurbo-Search for Special-Mate [C0/R2/K0/P0/X0] in 13 ...  (Hash=2044MB)
   AutoTurbo-Search for Special-Mate [C0/R-2/K0/P0/X0] in 13 ...  (Hash=2044MB)
   AutoTurbo-Search for Special-Mate [C0/R3/K0/P0/X0] in 13 ...  (Hash=2044MB)
   AutoTurbo-Search for Special-Mate [C0/R-3/K0/P0/X0] in 13 ...  (Hash=2044MB)
   AutoTurbo-Search for Special-Mate [C0/R0/K1/P0/X0] in 13 ...  (Hash=2044MB)
   AutoTurbo-Search for Special-Mate [C0/R0/K2/P0/X0] in 13 ...  (Hash=2044MB)
   AutoTurbo-Search for Special-Mate [C0/R0/K3/P0/X0] in 13 ...  (Hash=2044MB)
   AutoTurbo-Search for Special-Mate [C0/R0/K4/P0/X0] in 13 ...  (Hash=2044MB)
    13/13	41:11	2,724,064,996	1,200,372	+M13	Rg8+
    13/13	41:11	2,983,735,614	1,211,228	+M13	Rg8+ Kf2 Rg2+ Kf1 h4 Rxe5+ Kf6 Rh5 h3 e5+ Ke7 Rh7 Bb5+ Ke1 h2 e6 Nd3+ Kd1 Bc6 Rxf7+ Kxe6 Re7+ Kxe7 f4 h1Q+
   AutoTurbo-Search for Special-Mate [C0/R0/K4/P4/X0] in 13 ...  (Hash=2044MB)
   2017-11-10 4:49:13 PM, Time for this analysis: 01:00:00, Rated time: 41:11

1 of 1 matching moves
2017-11-10 4:49:14 PM, Total time: 1:00:04 AM
Rated time: 41:11 = 2471 Seconds
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.
Top
User avatar
MikeB
Posts: 4889
Joined: Thu Mar 09, 2006 6:34 am
Location: Pen Argyl, Pennsylvania

Re: Definitely not Mate in 14

Post by MikeB »

Dann Corbit wrote:Definitely a mate in 13. Chest output is pretty well a proof that there is not shorter solution. You can also ask for a proof tree, but with a mate in 13 it would be truly enormous.

Code: Select all

Analysis of Q:\dm26.epd   
2017-11-10 3:49:10 PM Level: 3600 Seconds
Analyzing engine: ChestUCI_AutoTurbo

 1 Rg8+;                
    Searching move: Rd8-g8
    Best move (ChestUCI_AutoTurbo): Rd8-g8
    Identical moves! Found in: 41:11
   FEN: 3r4/3bkp2/8/R3p2p/4Pn2/1P3PK1/8/8 b - -   (5+7)
   Position-Analysis:  C0/R0/K4/P3/X15   B:6/31
   AutoTurbo-Search for Special-Mate [C1/R0/K0/P0/X0] in 13 ...  (Hash=2044MB)
   AutoTurbo-Search for Special-Mate [C0/R1/K0/P0/X0] in 13 ...  (Hash=2044MB)
   AutoTurbo-Search for Special-Mate [C0/R2/K0/P0/X0] in 13 ...  (Hash=2044MB)
   AutoTurbo-Search for Special-Mate [C0/R-2/K0/P0/X0] in 13 ...  (Hash=2044MB)
   AutoTurbo-Search for Special-Mate [C0/R3/K0/P0/X0] in 13 ...  (Hash=2044MB)
   AutoTurbo-Search for Special-Mate [C0/R-3/K0/P0/X0] in 13 ...  (Hash=2044MB)
   AutoTurbo-Search for Special-Mate [C0/R0/K1/P0/X0] in 13 ...  (Hash=2044MB)
   AutoTurbo-Search for Special-Mate [C0/R0/K2/P0/X0] in 13 ...  (Hash=2044MB)
   AutoTurbo-Search for Special-Mate [C0/R0/K3/P0/X0] in 13 ...  (Hash=2044MB)
   AutoTurbo-Search for Special-Mate [C0/R0/K4/P0/X0] in 13 ...  (Hash=2044MB)
    13/13	41:11	2,724,064,996	1,200,372	+M13	Rg8+
    13/13	41:11	2,983,735,614	1,211,228	+M13	Rg8+ Kf2 Rg2+ Kf1 h4 Rxe5+ Kf6 Rh5 h3 e5+ Ke7 Rh7 Bb5+ Ke1 h2 e6 Nd3+ Kd1 Bc6 Rxf7+ Kxe6 Re7+ Kxe7 f4 h1Q+
   AutoTurbo-Search for Special-Mate [C0/R0/K4/P4/X0] in 13 ...  (Hash=2044MB)
   2017-11-10 4:49:13 PM, Time for this analysis: 01:00:00, Rated time: 41:11

1 of 1 matching moves
2017-11-10 4:49:14 PM, Total time: 1:00:04 AM
Rated time: 41:11 = 2471 Seconds
You better ask for the proof - if you play the line SF-McB displayed and back track with K-11.2 , K-11.2 also says mate in 12 . Unlikely for K11.2 to announce mate-in-12 unless it was true.

Image
Top
User avatar
MikeB
Posts: 4889
Joined: Thu Mar 09, 2006 6:34 am
Location: Pen Argyl, Pennsylvania

Re: Definitely not Mate in 14

Post by MikeB »

It appears that in SF-McB line , the third move Kf6 is a quiet move that chest and other engines are not factoring in correctly. If you force that move, K-11.2 will see the mate-in-12. Perhaps you need to try a parameter tweak in chest.
Top
User avatar
MikeB
Posts: 4889
Joined: Thu Mar 09, 2006 6:34 am
Location: Pen Argyl, Pennsylvania

Re: Definitely not Mate in 14

Post by MikeB »

from this position , which is after 3.Kf6 = white's 4 best moves all lead to mate in 9, so that pretty well confirms that it is mate--in 12 from the start position

K-11.2 analysis

[d]8/3b1p2/5k2/R3p2p/4Pn2/1P3P2/6r1/5K2 w - - 5 4

Code: Select all

dep	score	nodes	time	(not shown:  tbhits	knps	seldep)
 36	   #-9 	4.31G	2:28.44	Ra8 Kg7 Ra5 h4 Rxe5 h3 Rg5+ Rxg5 Kf2 h2 Ke3 h1=Q Kd2 Qxf3 Kc1 Rg2 e5 Qf1# 
 36	   #-9 	4.31G	2:28.44	Ra6+ Kg7 Rd6 Bb5+ Ke1 h4 Kd1 Bd3 Rxd3 Nxd3 f4 h3 f5 h2 f6+ Kxf6 b4 h1=Q# 
 36	   #-9 	4.31G	2:28.44	Ke1 h4 Ra6+ Be6 Ra8 Re2+ Kf1 Bh3+ Kg1 Re1+ Kh2 Bg2 Ra6+ Kg7 Rg6+ fxg6 b4 Rh1# 
 36	   #-9 	4.31G	2:28.44	Rd5 h4 Rd6+ Kg7 Ke1 Bb5 Kd1 Bd3 Rxd3 Nxd3 f4 h3 f5 h2 f6+ Kxf6 b4 h1=Q#
Top
Dann Corbit
Posts: 12662
Joined: Wed Mar 08, 2006 8:57 pm
Location: Redmond, WA USA

Re: Definitely not Mate in 14

Post by Dann Corbit »

MikeB wrote:from this position , which is after 3.Kf6 = white's 4 best moves all lead to mate in 9, so that pretty well confirms that it is mate--in 12 from the start position

K-11.2 analysis

[d]8/3b1p2/5k2/R3p2p/4Pn2/1P3P2/6r1/5K2 w - - 5 4

Code: Select all

dep	score	nodes	time	(not shown:  tbhits	knps	seldep)
 36	   #-9 	4.31G	2:28.44	Ra8 Kg7 Ra5 h4 Rxe5 h3 Rg5+ Rxg5 Kf2 h2 Ke3 h1=Q Kd2 Qxf3 Kc1 Rg2 e5 Qf1# 
 36	   #-9 	4.31G	2:28.44	Ra6+ Kg7 Rd6 Bb5+ Ke1 h4 Kd1 Bd3 Rxd3 Nxd3 f4 h3 f5 h2 f6+ Kxf6 b4 h1=Q# 
 36	   #-9 	4.31G	2:28.44	Ke1 h4 Ra6+ Be6 Ra8 Re2+ Kf1 Bh3+ Kg1 Re1+ Kh2 Bg2 Ra6+ Kg7 Rg6+ fxg6 b4 Rh1# 
 36	   #-9 	4.31G	2:28.44	Rd5 h4 Rd6+ Kg7 Ke1 Bb5 Kd1 Bd3 Rxd3 Nxd3 f4 h3 f5 h2 f6+ Kxf6 b4 h1=Q#
I am working backwards with the command line brute force version:

Code: Select all

Q:\>chest319 -b -M 8096 -z12 mlist.epd
8/5Pk1/8/8/1P3p2/3b4/6rp/4K3 b - - acn 2; acs 0; bm h1=Q# h1=R#; ce 32766; dm 1; fmvn 12; hmvc 0; pv h1=Q#;
8/5pk1/4P3/8/1P3p2/3b3p/6r1/4K3 b - - acn 71; acs 0; bm h2; ce 32764; dm 2; fmvn 11; hmvc 0; pv h2 e7 h1=Q#;
8/5pk1/8/4P3/1P3p1p/3b4/6r1/4K3 b - - acn 1412; acs 0; bm h3; ce 32762; dm 3; fmvn 10; hmvc 0; pv h3 e6 h2 e7 h1=Q#;
8/5pk1/8/4p3/1P2PP1p/3b4/6r1/4K3 b - - acn 14967; acs 0; bm Kh6 Kh7 Kh8 exf4 h3; ce 32760; dm 4; fmvn 9; hmvc 0; pv h3 f5 h2 f6+ Kh7 b5 h1=Q#;
8/5pk1/8/1b2p3/1P2P2p/3R1P2/6r1/4K3 b - - acn 26648; acs 0; bm Bxd3; ce 32758; dm 5; fmvn 8; hmvc 0; pv Bxd3 f4 h3 f5 h2 f6+ Kh7 b5 h1=Q#;
8/5pk1/3R4/1b2p3/1P2Pn1p/5P2/6r1/4K3 b - - acn 1022997; acs 1; bm Bd3 Nd3+; ce 32756; dm 6; fmvn 7; hmvc 0; pv Nd3+ Rxd3 Bxd3 f4 h3 f5 h2 f6+ Kh7 b5 h1=Q#;
8/5pk1/3R4/1b2p2p/4Pn2/1P3P2/6r1/4K3 b - - acn 8260752; acs 7; bm h4; ce 32754; dm 7; fmvn 6; hmvc 10; pv h4 b4 Nd3+ Rxd3 Bxd3 f4 h3 f5 h2 f6+ Kh7 b5 h1=Q#;
8/3b1pk1/3R4/4p2p/4Pn2/1P3P2/6r1/5K2 b - - acn 30596512; acs 26; bm Bb5+ h4; ce 32752; dm 8; fmvn 5; hmvc 8; pv Bb5+ Ke1 h4 b4 Nd3+ Rxd3 Bxd3 f4 h3 f5 h2 f6+ Kh7 b5 h1=Q#;
8/3b1p2/R4k2/4p2p/4Pn2/1P3P2/6r1/5K2 b - - acn 91896075; acs 68; bm Kg7; ce 32750; dm 9; fmvn 4; hmvc 6; pv Kg7 Rd6 Bb5+ Ke1 h4 b4 Nd3+ Rxd3 Bxd3 f4 h3 f5 h2 f6+ Kh7 b5 h1=Q#;
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.
Top
Dann Corbit
Posts: 12662
Joined: Wed Mar 08, 2006 8:57 pm
Location: Redmond, WA USA

Re: Definitely not Mate in 14

Post by Dann Corbit »

dm 10 from here:
[d]8/3bkp2/8/R3p2p/4Pn2/1P3P2/6r1/5K2 b - - acn 991774154; acs 844; bm Kf6; ce 32748; dm 10; fmvn 3; hmvc 4; pv Kf6 Ra6+ Kg7 Rd6 Bb5+ Ke1 h4 b4 Nd3+ Rxd3 Bxd3 f4 h3 f5 h2 f6+ Kh7 b5 h1=Q#;
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.
Top
User avatar
MikeB
Posts: 4889
Joined: Thu Mar 09, 2006 6:34 am
Location: Pen Argyl, Pennsylvania

Re: Definitely not Mate in 14

Post by MikeB »

Dann Corbit wrote:dm 10 from here:
[d]8/3bkp2/8/R3p2p/4Pn2/1P3P2/6r1/5K2 b - - acn 991774154; acs 844; bm Kf6; ce 32748; dm 10; fmvn 3; hmvc 4; pv Kf6 Ra6+ Kg7 Rd6 Bb5+ Ke1 h4 b4 Nd3+ Rxd3 Bxd3 f4 h3 f5 h2 f6+ Kh7 b5 h1=Q#;
Well if that's the case you don't need a program in brute force mode to tell you it's mate a 12. The first two moves, Rg8+ and Rg2+ are very forcing. Not a lot of options for white - surprised chest didn't see this. Kf6 was the key.
Top
Dann Corbit
Posts: 12662
Joined: Wed Mar 08, 2006 8:57 pm
Location: Redmond, WA USA

Re: Definitely not Mate in 14

Post by Dann Corbit »

MikeB wrote:
Dann Corbit wrote:dm 10 from here:
[d]8/3bkp2/8/R3p2p/4Pn2/1P3P2/6r1/5K2 b - - acn 991774154; acs 844; bm Kf6; ce 32748; dm 10; fmvn 3; hmvc 4; pv Kf6 Ra6+ Kg7 Rd6 Bb5+ Ke1 h4 b4 Nd3+ Rxd3 Bxd3 f4 h3 f5 h2 f6+ Kh7 b5 h1=Q#;
Well if that's the case you don't need a program in brute force mode to tell you it's mate a 12. The first two moves, Rg8+ and Rg2+ are very forcing. Not a lot of options for white - surprised chest didn't see this. Kf6 was the key.
I will also have to consider if white missed the best response (e.g. delay the mate one move by throwing away a rook, etc.)
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.
Top
Dann Corbit
Posts: 12662
Joined: Wed Mar 08, 2006 8:57 pm
Location: Redmond, WA USA

Re: Definitely not Mate in 14

Post by Dann Corbit »

MikeB wrote:
Dann Corbit wrote:dm 10 from here:
[d]8/3bkp2/8/R3p2p/4Pn2/1P3P2/6r1/5K2 b - - acn 991774154; acs 844; bm Kf6; ce 32748; dm 10; fmvn 3; hmvc 4; pv Kf6 Ra6+ Kg7 Rd6 Bb5+ Ke1 h4 b4 Nd3+ Rxd3 Bxd3 f4 h3 f5 h2 f6+ Kh7 b5 h1=Q#;
Well if that's the case you don't need a program in brute force mode to tell you it's mate a 12. The first two moves, Rg8+ and Rg2+ are very forcing. Not a lot of options for white - surprised chest didn't see this. Kf6 was the key.
You are definitely correct. This is enough to definitely prove your case:

Code: Select all

8/5Pk1/8/8/1P3p2/3b4/6rp/4K3 b - - acn 2; acs 0; bm h1=Q# h1=R#; ce 32766; dm 1; pv h1=Q#;
8/5pk1/4P3/8/1P3p2/3b4/6rp/4K3 w - - acn 20; acs 0; ce -32765; pv exf7 h1=Q#;
8/5pk1/4P3/8/1P3p2/3b3p/6r1/4K3 b - - acn 71; acs 0; bm h2; ce 32764; dm 2; pv h2 e7 h1=Q#;
8/5pk1/8/4P3/1P3p2/3b3p/6r1/4K3 w - - acn 81; acs 0; ce -32763; pv e6 h2 e7 h1=Q#;
8/5pk1/8/4P3/1P3p1p/3b4/6r1/4K3 b - - acn 1386; acs 0; bm h3; ce 32762; dm 3; pv h3 e6 h2 e7 h1=Q#;
8/5pk1/8/8/1P2Pp1p/3b4/6r1/4K3 w - - acn 427; acs 0; ce -32761; pv e5 h3 e6 h2 e7 h1=Q#;
8/5pk1/8/4p3/1P2PP1p/3b4/6r1/4K3 b - - acn 14659; acs 0; bm Kh6 Kh7 Kh8 exf4 h3; ce 32760; dm 4; pv h3 f5 h2 f6+ Kh7 b5 h1=Q#;
8/5pk1/8/4p3/1P2P2p/3b1P2/6r1/4K3 w - - acn 443; acs 0; ce -32759;
8/5pk1/8/1b2p3/1P2P2p/3R1P2/6r1/4K3 b - - acn 26350; acs 0; bm Bxd3; ce 32758; dm 5; pv Bxd3 f4 h3 f5 h2 f6+ Kh7 b5 h1=Q#;
8/5pk1/3R4/1b2p3/1P2P2p/3n1P2/6r1/4K3 w - - acn 598472; acs 1; ce -32757; pv Rxd3 Bxd3 f4 h3 f5 h2 f6+ Kh7 b5 h1=Q#;
8/5pk1/3R4/1b2p3/1P2Pn1p/5P2/6r1/4K3 b - - acn 1001721; acs 1; bm Bd3 Nd3+; ce 32756; dm 6; pv Nd3+ Rxd3 Bxd3 f4 h3 f5 h2 f6+ Kh7 b5 h1=Q#;
8/5pk1/3R4/1b2p3/4Pn1p/1P3P2/6r1/4K3 w - - acn 5594169; acs 9; ce -32755;
8/5pk1/3R4/1b2p2p/4Pn2/1P3P2/6r1/4K3 b - - acn 7780479; acs 5; bm h4; ce 32754; dm 7; pv h4 b4 Nd3+ Rxd3 Bxd3 f4 h3 f5 h2 f6+ Kh7 b5 h1=Q#;
8/5pk1/3R4/1b2p2p/4Pn2/1P3P2/6r1/5K2 w - - acn 20848; acs 0; ce -32753; pv Ke1 h4 b4 Nd3+ Rxd3 Bxd3 f4 h3 f5 h2 f6+ Kh7 b5 h1=Q#;
8/3b1pk1/3R4/4p2p/4Pn2/1P3P2/6r1/5K2 b - - acn 30586275; acs 24; bm Bb5+ h4; ce 32752; dm 8; pv Bb5+ Ke1 h4 b4 Nd3+ Rxd3 Bxd3 f4 h3 f5 h2 f6+ Kh7 b5 h1=Q#;
8/3b1pk1/R7/4p2p/4Pn2/1P3P2/6r1/5K2 w - - acn 14324754; acs 14; ce -32751;
8/3b1p2/R4k2/4p2p/4Pn2/1P3P2/6r1/5K2 b - - acn 82032108; acs 64; bm Kg7; ce 32750; dm 9; pv Kg7 Rd6 Bb5+ Ke1 h4 b4 Nd3+ Rxd3 Bxd3 f4 h3 f5 h2 f6+ Kh7 b5 h1=Q#;
8/3b1p2/5k2/R3p2p/4Pn2/1P3P2/6r1/5K2 w - - acn 169757133; acs 145; ce -32749; pv Ra6+ Kg7 Rd6 Bb5+ Ke1 h4 b4 Nd3+ Rxd3 Bxd3 f4 h3 f5 h2 f6+ Kh7 b5 h1=Q#;
The previous move is your quiet move and the moves before that were common to all the other solutions.
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.
Top

Return to “Computer Chess Club: General Topics”