hardest 6x6?

Everything about Sudoku that doesn't fit in one of the other sections

Re: hardest 6x6?

Postby Mathimagics » Tue Jan 01, 2019 11:29 am

1to9only wrote:A quick hack of SE, and run against Mathimagics's AllTW list of 14154 6x6 grids …

Nice work! 8-)

But it leaves tarek with nothing to do! :(

PS: can you attach my file with puzzles tagged with the SE rating? That would be useful ...
User avatar
Mathimagics
2017 Supporter
 
Posts: 882
Joined: 27 May 2015

Re: hardest 6x6?

Postby Mathimagics » Tue Jan 01, 2019 11:40 am

.
What tarek could do is to road-test that 8.3 example - how hard is it for P&P solving? Is it somewhat harder than the OP's 2 x 7.3 puzzles?
User avatar
Mathimagics
2017 Supporter
 
Posts: 882
Joined: 27 May 2015

Re: hardest 6x6?

Postby 1to9only » Tue Jan 01, 2019 11:43 am

Mathimagics wrote:can you attach my file with puzzles tagged with the SE rating? That would be useful ...
Attachments
AllTW.rated.zip
AllTW grids SE6 rated
(119.91 KiB) Downloaded 6 times
1to9only
 
Posts: 458
Joined: 04 April 2018

Re: hardest 6x6?

Postby 1to9only » Tue Jan 01, 2019 11:49 am

For the 8.3/8.3/6.6 grid, the SE solution path is:
Code: Select all
..3...4.....2..3.5.....1.6.5.45...1. ED=8.3/8.3/6.6
..3...4.....2..3.5.....1.6.5.45...1. 6.6, Turbot Fish: R1C5.6 off
..3...4.....2..3.5.....1.6.5.45...1. 7.1, Forcing Chain: R2C3.2 off
..3...4.....2..3.5.....1.6.5.45...1. 2.6, Pointing: Cells R1C2,R2C2: 2 in block and column
..3...4.....2..3.5.....1.6.5.45...1. 7.1, Forcing Chain: R2C2.1 off
..3...4.....2..3.5.....1.6.5.45...1. 7.1, Forcing Chain: R2C4.2 off
..3...4.....2..3.5.....1.6.5.45...1. 7.1, Forcing Chain: R4C3.4 off
..3...4.....2..3.5.....1.6.5.45...1. 7.2, Forcing Chain: R1C4.6 off
..3...4.....2..3.5.....1.6.5.45...1. 7.2, Forcing Chain: R1C5.2 off
..3...4.....2..3.5.....1.6.5.45...1. 7.2, Forcing Chain: R2C5.2 off
..3...4.....2..3.5.....1.6.5.45...1. 7.1, Forcing Chain: R4C5.6 off
..3...4.....2..3.5.....1.6.5.45...1. 7.2, Forcing Chain: R2C6.6 off
..3...4.....2..3.5.....1.6.5.45...1. 7.2, Bidirectional Cycle: R1C1,R5C1,R6C2,R6C6,R2C6,R1C6
..3...4.....2..3.5.....1.6.5.45...1. 2.6, Pointing: Cells R1C6,R2C6: 2 in block and column
..3...4.....2..3.5.....1.6.5.45...1. 8.3, Cell Forcing Chains: R2C3 ==> R2C5.6 off
..3...4.....2..365.....1.6.5.45...1. 1.5, Hidden Single: R3C5: 6 in column
..3...4.....2..365.....1.6.5.45...1. 2.6, Pointing: Cells R4C4,R4C5: 4 in block and row
..3...4.....2..365.....1.6.5.45...1. 3.4, Hidden Pair: Cells R2C3,R4C3: 5,6 in column
..3...4..1..2..365.....1.6.5.45...1. 1.5, Hidden Single: R2C4: 1 in row
..3..64..1..2..365.....1.6.5.45...1. 1.2, Hidden Single: R1C6: 6 in block
..3..64.61..2..365.....1.6.5.45...1. 1.2, Hidden Single: R2C3: 6 in block
..3..64.61.22..365.....1.6.5.45...1. 1.2, Hidden Single: R2C6: 2 in block
..3..64.61.22..365.....1.6.5.45...13 1.0, Hidden Single: R6C6: 3 in column
.23..64.61.22..365.....1.6.5.45...13 1.2, Hidden Single: R1C2: 2 in block
123..64.61.22..365.....1.6.5.45...13 1.2, Hidden Single: R1C1: 1 in block
123..64561.22..365.....1.6.5.45...13 1.0, Hidden Single: R2C2: 5 in block
123..64561322..365.....1.6.5.45...13 1.0, Hidden Single: R2C5: 3 in row
123.564561322..365.....1.6.5.45...13 1.2, Hidden Single: R1C5: 5 in block
1234564561322..365.....1.6.5.45...13 1.0, Hidden Single: R1C4: 4 in block
1234564561322..365..5..1.6.5.45...13 1.2, Hidden Single: R4C3: 5 in block
1234564561322..3656.5..1.6.5.45...13 1.2, Hidden Single: R4C1: 6 in block
1234564561322..3656.5..136.5.45...13 1.0, Hidden Single: R5C1: 3 in column
1234564561322..365635..136.5.45...13 1.2, Hidden Single: R4C2: 3 in block
1234564561322..365635.4136.5.45...13 1.2, Hidden Single: R4C5: 4 in block
1234564561322..36563524136.5.45...13 1.0, Hidden Single: R4C4: 2 in block
1234564561322..36563524136.5.45..613 1.0, Hidden Single: R6C4: 6 in column
1234564561322..36563524136.5245..613 1.0, Hidden Single: R5C5: 2 in block
1234564561322..3656352413615245..613 1.0, Hidden Single: R5C3: 1 in row
12345645613221.3656352413615245..613 1.2, Hidden Single: R3C2: 1 in block
1234564561322143656352413615245..613 1.0, Hidden Single: R3C3: 4 in block
12345645613221436563524136152454.613 1.0, Hidden Single: R6C2: 4 in column
123456456132214365635241361524542613 1.0, Hidden Single: R6C3: 2 in block
ED=8.3/8.3/6.6

Oooops! nothing for tarek to do again!
1to9only
 
Posts: 458
Joined: 04 April 2018

Re: hardest 6x6?

Postby Mathimagics » Tue Jan 01, 2019 12:19 pm

.
Sudoku 6x6 is "Sudoku for kids"?

Apparently not - unless the kid is pretty smart … :?
User avatar
Mathimagics
2017 Supporter
 
Posts: 882
Joined: 27 May 2015

Re: hardest 6x6?

Postby tarek » Tue Jan 01, 2019 1:28 pm

Well done all for keeping me with nothing to do. You’ve been busy ;)

I can understand the appeal for these puzzles because if indeed there no singles backdoor size bigger than 1 then these difficult ones are ideal for a one trick pony type puzzles

Excellent

HNY sudoku players/programmers

Tarek
User avatar
tarek
 
Posts: 2791
Joined: 05 January 2006

Re: hardest 6x6?

Postby 1to9only » Tue Jan 01, 2019 7:46 pm

1to9only wrote:I'll have a Sudoku6(2Rx3C)Explainer.jar out soon...
The GUI is untouched (still 9x9), so will NOT work well with the 6x6 solver!!

I am trying to get the GUI working for 6x6 grids, to save having to patch 789 into 9x9 SE...
1to9only
 
Posts: 458
Joined: 04 April 2018

Re: hardest 6x6?

Postby eleven » Tue Jan 01, 2019 10:56 pm

1to9only wrote:For the 8.3/8.3/6.6 grid, the SE solution path is: ...

..3 | ...
4.. | ...
-----------
2.. | 3.5
... | ..1
-----------
.6. | 5.4
5.. | .1.

I tried that puzzle on paper, and it was surprisingly tough. It took me 6 or 7 chains, but none with a difficulty of an ER 8.3.
Probably my first step (which SE cannot do) broke it down to a lower rating:
(2=6)r1c6 - (6=3)r145c1 - 3r5c5 = 2r5c5&r4c4 => -2r12c45,r6c6
eleven
 
Posts: 1945
Joined: 10 February 2008

Re: hardest 6x6?

Postby tarek » Tue Jan 01, 2019 11:50 pm

Just to confirm that all puzzles in the ALLTW list have a singles backdoor size of 1

tarek
User avatar
tarek
 
Posts: 2791
Joined: 05 January 2006

Re: hardest 6x6?

Postby eleven » Wed Jan 02, 2019 12:55 am

1to9only wrote:I am trying to get the GUI working for 6x6 grids, to save having to patch 789 into 9x9 SE...

Ah, that's a nice trick with the 789.
So i could use a solver to check my solution. And it turned out, that i made a lucky mistake on paper. This is my repaired move, which is the hardest:
Code: Select all
+----------------------+----------------------+
|a16     125    3      | 14     45     26     |
| 4      25     256-1  |e16     356    236    |
+----------------------+----------------------+
| 2      14    d146    | 3      46     5      |
|b36     345   c456    |c246    24     1      |
+----------------------+----------------------+
| 13     6     d12     | 5      23     4      |
| 5      234   d24     | 26     1      36     |
+----------------------+----------------------+

(1=6)r1c1 - 6r4c1 = 6r4c34 - [(6=1)r356c3 | (6=1)r2c4] => -1r2c3
eleven
 
Posts: 1945
Joined: 10 February 2008

Re: hardest 6x6?

Postby 1to9only » Wed Jan 02, 2019 11:05 am

To keep SE stuff in one thread, to download the SE6 solver/rating program see this post:
http://forum.enjoysudoku.com/help-with-sudoku-explainer-t6677-75.html#p273912

Enjoy!
1to9only
 
Posts: 458
Joined: 04 April 2018

Re: hardest 6x6?

Postby tarek » Wed Jan 02, 2019 7:46 pm

I tried venturing into the land of the Partially weak & Totally weak

Some results:
Code: Select all
.43.1.......5.1.4............26..3.. ED=7.2/2.0/2.0
.4.1..5....2123.............3....6.5 ED=7.2/1.2/1.2
6...1...12.....3....4.2......3.5.1.. ED=7.2/1.2/1.2
5.......1..........3..14.....2.2.64. ED=7.1/7.1/2.6
..6.2..2.5.....4.......11...3...3... ED=7.1/3.0/3.0
User avatar
tarek
 
Posts: 2791
Joined: 05 January 2006

Re: hardest 6x6?

Postby Mathimagics » Thu Jan 03, 2019 8:12 am

.
Very adventurous! 8-)

BTW: If you are looking for something to do, you might have a crack at my SudokuPW 16x16 ???
User avatar
Mathimagics
2017 Supporter
 
Posts: 882
Joined: 27 May 2015

Re: hardest 6x6?

Postby tarek » Thu Jan 03, 2019 8:57 am

Mathimagics wrote:.
Very adventurous! 8-)
going back to TW and PW. Did you the number of singles placed before the 1st guess to differentiate them? And therefore would PW0 = TW?

Mathimagics wrote:BTW: If you are looking for something to do, you might have a crack at my SudokuPW 16x16 ???

On the to-do list

Tarek
User avatar
tarek
 
Posts: 2791
Joined: 05 January 2006

Re: hardest 6x6?

Postby Mathimagics » Thu Jan 03, 2019 1:42 pm

tarek wrote:And therefore would PW0 = TW?

Indeed it would! 8-)
User avatar
Mathimagics
2017 Supporter
 
Posts: 882
Joined: 27 May 2015

PreviousNext

Return to General