To solve hardest Sudoku

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

Postby Papy » Tue Jan 23, 2007 7:38 pm

Here is the way to renumber reniumber to grid

9 by 1
5 by 2
2 by 3
7 by 4
7 by 5
8 by 6
2 by 7
9 by 8

It's well the same gid
To be sure: solve the 1 grid and when done remap the grid with the row1 at 1213456789. You will see that is well the same puzzle.
Perhaps my method is good. It depends the way you use to renumber.

Papy
Papy
 
Posts: 131
Joined: 15 August 2006

Postby RW » Tue Jan 23, 2007 9:06 pm

Papy wrote:Here is the way to renumber reniumber to grid

9 by 1
5 by 2
2 by 3
7 by 4
7 by 5
8 by 6
2 by 7
9 by 8

Code: Select all
 *-----------*
 |3..|.7.|..8|
 |.2.|...|.9.|
 |..1|...|4..|
 |---+---+---|
 |...|1.6|...|
 |8..|.5.|..6|
 |...|2.8|...|
 |---+---+---|
 |..4|...|5..|
 |.9.|...|.1.|
 |6..|.8.|..7|
 *-----------*

 *-----------*
 |1..|.5.|..9|
 |.8.|...|.4.|
 |..4|...|6..|
 |---+---+---|
 |...|2.3|...|
 |9..|.6.|..1|
 |...|8.7|...|
 |---+---+---|
 |..6|...|8..|
 |.4.|...|.2.|
 |3..|.7.|..5|
 *-----------*

Above are the two puzzles you have posted. They show absolutely no connection to your table of relabeling. According to your table you have also relabeled digits 2, 7 and 9 twice...

Even if you had relabeled the grid in a correct way, finding the digit mappings to sort the first row is a result of a lot of guessing. It is a lot more effective to just guess the digits of the first row without relabeling. Or why not just guess r9c2=3 and the puzzle is solved immediately.

RW
RW
2010 Supporter
 
Posts: 1010
Joined: 16 March 2006

Postby Papy » Wed Jan 24, 2007 6:22 pm

Sorry RW I make a mistake
The originaml grid is

900040006
010000070
007000880

000050200
600080009
000103000

008000100
070000050
200030004

Solution

952 748 316
816 359 472
347 216 895

734 592 681
621 487 539
589 163 247

468 975 123
173 624 958
295 893 764

And If You renbumber You find

123 456 789
689 721 543
754 389 612

475 213 968
938 564 271
261 897 354

596 142 837
847 935 126
312 678 495

And nox All is good!
The puzzle are isomorphs with the digits
and you cxan well bein with 123456789 in the first rox like hypopthése.

Papy
Really soory.
Papy
 
Posts: 131
Joined: 15 August 2006

Postby ronk » Wed Jan 24, 2007 7:10 pm

Papy wrote:Solution

952 748 316
816 359 472
347 216 895

734 592 681
621 487 539
589 163 247

468 975 123
173 624 958
295 893 764

And If You renbumber You find

123 456 789
689 721 543
754 389 612

475 213 968
938 564 271
261 897 354

596 142 837
847 935 126
312 678 495

Papy, to renumber in a manner that "normalizes" row 1 is trivial when you know the solution. Without the solution, however, you would have to guess to arrive at the same renumbering.

And if someone else renumbered the puzzle the same way, as a solver you would need to correctly guess that the puzzle is row-normalized to begin by placing digits in the first row.

IOW this is not a solving technique, so what is the purpose of this exercise?
ronk
2012 Supporter
 
Posts: 4764
Joined: 02 November 2005
Location: Southeastern USA

Postby Papy » Thu Jan 25, 2007 8:34 am

The last mail was to repoare my miustake and show how to cntrol the result
But I have explain ONE way use to renumber digits in my first message with the Box 5
The goal is clear: complet arbitrary one set with the best probability.
If this method doesn't correspond to your mind don't use it: I don't use many elaborate methods because I don't assume them!

With this method I solve the Flower and before I set no digitd!

I really think that is you organize all the clue from one set you have more
probalnility

In a second time but it's more complexe You can alse swap row and columns to arrive to a 'canon' grid

Papy
Papy
 
Posts: 131
Joined: 15 August 2006

Previous

Return to General