Any engine can solve this mate in 29 by now ?

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

Moderators: hgm, Dann Corbit, Harvey Williamson

Forum rules
This textbox is used to restore diagrams posted with the [d] tag before the upgrade.
BBauer
Posts: 658
Joined: Wed Mar 08, 2006 7:58 pm

Re: The mate in 29 by Herbert Grasemann

Post by BBauer » Mon Dec 28, 2015 3:26 pm

Try this one first
[D]8/3p1p2/3K1P2/4P3/1p6/1Pp5/PbP2ppp/1R1N1bqk w - - bm Kc7; "#10, nach H. Grasemann 1964";
Kind regards
Bernhard

mar
Posts: 2320
Joined: Fri Nov 26, 2010 1:00 pm
Location: Czech Republic
Full name: Martin Sedlak

Re: The mate in 29 by Herbert Grasemann

Post by mar » Mon Dec 28, 2015 3:38 pm

What might help if you have the mainline is to analyze backwards.
There are some repetitions so it may as well be a mate in less moves.
Impressive that Rainbow Serpent can solve this.
While it's "only" +2, it's failing high so I bet it would get a much larger score sooner or later.

User avatar
Eelco de Groot
Posts: 4279
Joined: Sun Mar 12, 2006 1:40 am
Location: Groningen

Re: The mate in 29 by Herbert Grasemann

Post by Eelco de Groot » Mon Dec 28, 2015 3:53 pm

Hi,

Well, the belly of the serpent is not unlike that of the fire breathing dragon, like an oven. It has to contain fire. I think it can confine even the explosions, but the the serpent may need some entosorbine or Aka-Seltzer pills after eating this position.

It got as far as this now:


59/74 48:17 +3.77++ 1.Kc7 (12.206.895.639) 4212

59/74 48:45 +5.18++ 1.Kc7 (12.318.382.595) 4210

59/74 48:59 +6.95++ 1.Kc7 (12.370.528.037) 4208

59/74 49:11 +9.19++ 1.Kc7 (12.418.331.409) 4207

59/74 49:32 +12.00++ 1.Kc7 (12.501.357.002) 4205

59/74 50:08 +15.54++ 1.Kc7 (12.650.351.043) 4204

59/74 50:34 +20.75++ 1.Kc7 (12.758.476.964) 4203

59/74 53:19 +28.64++ 1.Kc7 (13.421.396.780) 4194

59/74 53:30 +M39++ 1.Kc7 (13.462.405.386) 4193

59/74 84:45 +M45 1.Kc7 Ba3 2.Kb8 Bb2 3.Ka7 Ba3 4.Ka8 Bb2
5.Kb8 Ba3 6.Kc7 Bb2 7.Kd6 Ba3 8.Kc5 h5
9.Kb6 Bb2 10.Kc7 Ba3 11.Kb8 Bb2
12.Ka8 Ba3 13.Ka7 Bb2 14.Kb6 (20.520.042.334) 4034

60/74 84:49 +M41++ 1.Kc7 (20.530.951.633) 4034
Debugging is twice as hard as writing the code in the first
place. Therefore, if you write the code as cleverly as possible, you
are, by definition, not smart enough to debug it.
-- Brian W. Kernighan

Ignacio
Posts: 150
Joined: Wed Mar 08, 2006 7:15 pm

Re: The mate in 29 by Herbert Grasemann

Post by Ignacio » Mon Dec 28, 2015 4:02 pm

Gustav 3.3b (f0, z12) needs 23 min. 7 sec. for mate in 29.

User avatar
Michael Diosi
Posts: 672
Joined: Mon Jun 22, 2009 11:37 am

Re: The mate in 29 by Herbert Grasemann

Post by Michael Diosi » Mon Dec 28, 2015 4:06 pm

Hi,


as I said my machine is too slow and my Gustav too old...

:( :shock:

Michael Diosi
http://www.playwitharena.com

User avatar
F.Huber
Posts: 810
Joined: Thu Mar 09, 2006 3:50 pm
Location: Austria
Contact:

Re: The mate in 29 by Herbert Grasemann

Post by F.Huber » Mon Dec 28, 2015 4:32 pm

Ignacio wrote:Gustav 3.3b (f0, z12) needs 23 min. 7 sec. for mate in 29.
Yep, C20/R0/K1/P4/X12 are the ChestUCI parameters I get from RetroAnalysis, but I could only solve the last 17 or 18 moves, everything before just takes too long.

BTW, Gustav 3.3b? Wow, you are to be envied, I also only have the first version 3.0. :(

Franz

peter
Posts: 2199
Joined: Sat Feb 16, 2008 6:38 am
Full name: Peter Martan

Re: The mate in 29 by Herbert Grasemann

Post by peter » Mon Dec 28, 2015 5:55 pm

Hi Eelco!

8/3p1p1p/3K1P2/4P3/1p6/1Pp5/PbP2ppp/1R1N1bqk w - - 0 1

Analysis by SF MateFinder 64 POPCNT: (study mode on)

...
1.Kc5 La3 2.Kd4 Lb2 3.Kc5
= (0.00) Tiefe: 38/25 00:02:46 3255MN
...
1.Kc5
+- (#44 ++) Tiefe: 39/43 00:03:01 3573MN
...
1.Kc5 La3 2.Kb6 Lb2 3.Ka7 La3 4.Ka8 Lb2 5.Kb8 La3 6.Kc7 Lb2 7.Kb6 La3 8.Kc5 h6 9.Kb6 Lb2 10.Ka7 La3 11.Ka8 Lb2 12.Kb8 La3 13.Kc7 Lb2 14.Kb6 La3 15.Ka5 h5 16.Kb6 Lb2 17.Kc7 La3 18.Kb8 Lb2 19.Ka8 La3 20.Ka7 Lb2
+- (#44) Tiefe: 40/43 00:04:07 4991MN
1.Kc7 La3 2.Kb6 Lb2 3.Ka7 La3 4.Ka8 Lb2 5.Kb8 La3 6.Kc7 Lb2 7.Kb6 La3 8.Kc5 h6 9.Kb6 Lb2 10.Ka7 La3 11.Ka8 Lb2 12.Kb8 La3 13.Kc7 Lb2 14.Kb6 La3 15.Kc5 h5 16.Kb6 Lb2 17.Kc7 La3 18.Kb8 Lb2 19.Ka8 La3 20.Ka7 Lb2 21.Kb6
+- (#46) Tiefe: 41/46 00:05:38 6949MN
...
1.Kc7 La3 2.Kb8 Lb2 3.Ka8 La3 4.Ka7 Lb2 5.Kb6 La3 6.Kc5 h6 7.Kb6 Lb2 8.Ka7 La3 9.Ka8 Lb2 10.Kb8 La3 11.Kc7 Lb2 12.Kb6 La3 13.Kc5 h5 14.Kb6 Lb2 15.Kc7 La3 16.Kb8 Lb2 17.Ka8 La3 18.Ka7 Lb2 19.Kb6 La3 20.Ka5 h4 21.Kb6 Lb2 22.Kc7 La3
+- (#42) Tiefe: 44/46 00:09:44 12175MN
...
1.Kc7 La3 2.Kb8 Lb2 3.Ka8 La3 4.Ka7 Lb2 5.Kb6 La3 6.Kc5 h6 7.Kd6 Lb2 8.Ke7 La3 9.Kf8 Lb2 10.Kg7 h5 11.Kf8 La3 12.Ke7 Lb2 13.Kd6 La3 14.Kc5 h4 15.Kb6 Lb2 16.Kc7 La3 17.Kb8 Lb2 18.Ka8 La3 19.Ka7 Lb2 20.Kb6 La3 21.Ka5 h3 22.Kb6 Lb2 23.Kc7 La3
+- (#35) Tiefe: 46/46 00:16:04 20047MN

If Kc5 would mate too the study would be somewhat cooked.
:(

The variant of the MateFinder's last 1.Kc5- output transposes to the solution- variant up to 6.Kc5
Peter.

User avatar
Eelco de Groot
Posts: 4279
Joined: Sun Mar 12, 2006 1:40 am
Location: Groningen

Re: The mate in 29 by Herbert Grasemann

Post by Eelco de Groot » Mon Dec 28, 2015 6:37 pm

Hi Peter!

Quick find by Stockfish Matefinder! Does it find the Mate in 29 for Kc7? Rainbow serpent also switches to Kc5, even reporting Mate in 31 but then loses some of the moves maybe, because then it is only Mate in 35 :( That is not very desirable behaviour of course. But if information is lost in the transposition table, I think it can not always be avoided. Then it switches to Kc7 again. If it really would be Mate in 31 for Kc5 then the solution is still valid I think? Kc7 leads to a shorter mate?

Regards, Eelco

Rainbow Serpent now, fifteen minutes ago, arrived at Mate in 30 8-)


59/57 41:11 +2.37++ 1.Kc7 (10.451.585.528) 4229

59/74 48:17 +3.77++ 1.Kc7 (12.206.895.639) 4212

59/74 48:45 +5.18++ 1.Kc7 (12.318.382.595) 4210

59/74 48:59 +6.95++ 1.Kc7 (12.370.528.037) 4208

59/74 49:11 +9.19++ 1.Kc7 (12.418.331.409) 4207

59/74 49:32 +12.00++ 1.Kc7 (12.501.357.002) 4205

59/74 50:08 +15.54++ 1.Kc7 (12.650.351.043) 4204

59/74 50:34 +20.75++ 1.Kc7 (12.758.476.964) 4203

59/74 53:19 +28.64++ 1.Kc7 (13.421.396.780) 4194

59/74 53:30 +M39++ 1.Kc7 (13.462.405.386) 4193

59/74 84:45 +M45 1.Kc7 Ba3 2.Kb8 Bb2 3.Ka7 Ba3 4.Ka8 Bb2
5.Kb8 Ba3 6.Kc7 Bb2 7.Kd6 Ba3 8.Kc5 h5
9.Kb6 Bb2 10.Kc7 Ba3 11.Kb8 Bb2
12.Ka8 Ba3 13.Ka7 Bb2 14.Kb6 (20.520.042.334) 4034

60/74 84:49 +M41++ 1.Kc7 (20.530.951.633) 4034

60/74 96:25 +M45-- 1.Kc7 Ba3 (23.274.284.337) 4022

60/74 96:28 +M41++ 1.Kc7 (23.284.172.691) 4022

60/74 100:37 +M31 1.Kc5 Ba3 2.Kb6 Bb2 3.Ka7 Ba3 4.Ka8 Bb2
5.Kb8 Ba3 6.Kc7 Bb2 7.Kd6 Ba3 8.Kc5 h6
9.Kd6 Bb2 10.Ke7 Ba3 11.Kf8 Bb2
12.Kg7 h5 13.Kf8 Ba3 14.Ke7 (24.217.148.229) 4011

61/74 101:24 +M31 1.Kc5 Ba3 2.Kb6 Bb2 3.Ka7 Ba3 4.Ka8 Bb2
5.Kb8 Ba3 6.Kc7 Bb2 7.Kd6 Ba3 8.Kc5 h6
9.Kd6 Bb2 10.Ke7 Ba3 11.Kf8 Bb2
12.Kg7 h5 13.Kf8 Ba3 14.Ke7 (24.399.704.011) 4010

62/74 116:48 +M35-- 1.Kc5 Ba3 (28.557.968.600) 4074

62/74 122:20 +M37 1.Kc5 Ba3 2.Kb6 Bb2 3.Ka7 Ba3 4.Ka8 Bb2
5.Kb8 Ba3 6.Kc7 Bb2 7.Kd6 Ba3 8.Kc5 h6
9.Kd6 Bb2 10.Ke7 Ba3 11.Kf8 Bb2
12.Kg7 h5 13.Kf8 Ba3 14.Ke7 (29.917.839.883) 4075

63/74 125:31 +M33++ 1.Kc7 (30.644.009.238) 4068

63/74 130:34 +M33 1.Kc7 Ba3 2.Kb6 Bb2 3.Ka7 Ba3 4.Ka8 Bb2
5.Kb8 Ba3 6.Kc7 Bb2 7.Kd6 Ba3 8.Kc5 h6
9.Kd6 Bb2 10.Ke7 Ba3 11.Kf8 Bb2
12.Kg7 h5 13.Kf8 Ba3 14.Ke7 (31.818.953.455) 4061

64/74 146:32 +M37-- 1.Kc7 Ba3 (35.488.534.755) 4036

64/74 146:35 +M33++ 1.Kc7 (35.501.637.508) 4036

64/74 168:48 +M37-- 1.Kc7 Ba3 (40.610.336.606) 4009

64/74 174:06 +M32 1.Kc7 Ba3 2.Kb6 Bb2 3.Ka7 Ba3 4.Ka8 Bb2
5.Kb8 Ba3 6.Kc7 Bb2 7.Kd6 Ba3 8.Kc5 h6
9.Kd6 Bb2 10.Ke7 Ba3 11.Kf8 Bb2
12.Kg7 h5 13.Kf8 Ba3 14.Ke7 (41.771.575.556) 3998

65/74 175:26 +M32 1.Kc7 Ba3 2.Kb6 Bb2 3.Ka7 Ba3 4.Ka8 Bb2
5.Kb8 Ba3 6.Kc7 Bb2 7.Kd6 Ba3 8.Kc5 h6
9.Kd6 Bb2 10.Ke7 Ba3 11.Kf8 Bb2
12.Kg7 h5 13.Kf8 Ba3 14.Ke7 (42.063.542.361) 3995

66/74 179:21 +M32 1.Kc7 Ba3 2.Kb6 Bb2 3.Ka7 Ba3 4.Ka8 Bb2
5.Kb8 Ba3 6.Kc7 Bb2 7.Kd6 Ba3 8.Kc5 h6
9.Kd6 Bb2 10.Ke7 Ba3 11.Kf8 Bb2
12.Kg7 h5 13.Kf8 Ba3 14.Ke7 (42.938.353.801) 3990

67/74 202:50 +M33 1.Kc7 Ba3 2.Kb8 Bb2 3.Ka7 Ba3 4.Ka8 Bb2
5.Kb8 Ba3 6.Kc7 Bb2 7.Kd6 Ba3 8.Kc5 h6
9.Kd6 Bb2 10.Ke7 Ba3 11.Kf8 Bb2
12.Kg7 h5 13.Kf8 Ba3 14.Ke7 (48.317.246.942) 3970

68/74 225:42 +M31 1.Kc7 Ba3 2.Kb8 Bb2 3.Ka8 Ba3 4.Ka7 Bb2
5.Kb6 Ba3 6.Kc5 h6 7.Kd6 Bb2 8.Ke7 Ba3
9.Kf8 Bb2 10.Kg7 h5 11.Kf8 Ba3
12.Ke7 Bb2 13.Kd6 h4 14.Kc7 (53.576.201.632) 3956

69/74 235:09 +M30 1.Kc7 Ba3 2.Kb8 Bb2 3.Ka8 Ba3 4.Ka7 Bb2
5.Kb6 Ba3 6.Kc5 h6 7.Kd6 Bb2 8.Ke7 Ba3
9.Kf8 Bb2 10.Kg7 h5 11.Kf8 Ba3
12.Ke7 Bb2 13.Kd6 h4 14.Ke7 (55.746.802.890) 3951
Debugging is twice as hard as writing the code in the first
place. Therefore, if you write the code as cleverly as possible, you
are, by definition, not smart enough to debug it.
-- Brian W. Kernighan

peter
Posts: 2199
Joined: Sat Feb 16, 2008 6:38 am
Full name: Peter Martan

Re: The mate in 29 by Herbert Grasemann

Post by peter » Mon Dec 28, 2015 7:02 pm

Eelco de Groot wrote: Quick find by Stockfish Matefinder!
Cannot be sure, if there couldn't have been leavings of solution- hash from other trials, didn'tn restart the GUI, but the engine had been changed in between.
Eelco de Groot wrote: Does it find the Mate in 29 for Kc7?
I stopped the search at an #35- output after 1.Kc7 as shown in the posting.
Eelco de Groot wrote: If it really would be Mate in 31 for Kc5 then the solution is still valid I think? Kc7 leads to a shorter mate?
That depends on the pov, Eelco, hardliners would see it yet as a major dual, I guess, for my personal view it wouldn't matter, but I'm not a judge of studies, just a consumer.
:)

I'm trying to keep the correct number of moves to mate in hash at Backward, if it works, I'll come back with it
Peter.

peter
Posts: 2199
Joined: Sat Feb 16, 2008 6:38 am
Full name: Peter Martan

Re: The mate in 29 by Herbert Grasemann

Post by peter » Mon Dec 28, 2015 8:30 pm

peter wrote: I'm trying to keep the correct number of moves to mate in hash at Backward, if it works, I'll come back with it
Took some time:

8/3p1p1p/3K1P2/4P3/1p6/1Pp5/PbP2ppp/1R1N1bqk w - - 0 1

Analysis by SF MateFinder 64 POPCNT:

1.Kc7 Ba3 2.Kb8 Bb2 3.Ka8 Ba3 4.Ka7 h6 5.Kb8 Bb2 6.Kc7 Ba3 7.Kd8 Bb2 8.Ke7 Ba3 9.Kf8 Bb2 10.Kg7 h5 11.Kf8 Ba3 12.Ke7 Bb2 13.Kd6 Ba3 14.Kc5 h4 15.Kd4 Bb2 16.Ke3 Ba3 17.Kf4 Bb2 18.Kg5 Ba3 19.Kxh4 d6 20.exd6 Bb2 21.Kh3 Ba3
+- (#29) Depth: 42/42 00:26:30 53884MN

And here's what I got after playing 1.Kc5 with still full hash after Backward:

8/3p1p1p/5P2/2K1P3/1p6/1Pp5/PbP2ppp/1R1N1bqk b - - 0 1

Analysis by SF MateFinder 64 POPCNT:

1...Ba3 2.Kb6 Bb2 3.Ka7 Ba3 4.Ka8 Bb2 5.Kb8 Ba3 6.Kc7 Bb2 7.Kd6 Ba3 8.Kc5 h6 9.Kd6 Bb2 10.Ke7 Ba3 11.Kf8 Bb2 12.Kg7 h5 13.Kf8 Ba3 14.Ke7 Bb2 15.Kd6 Ba3 16.Kc5 h4 17.Kd4 Bb2 18.Ke3 Ba3 19.Kf4 Bb2 20.Kg5 Ba3
+- (#30) Depth: 39/39 00:06:03 12539MN

Would be funny if it really was just one move or two moves longer
Peter.

Post Reply