The New Sudoku Players' Forum

by **eleven** » Wed May 20, 2015 11:17 pm

blue wrote:
Code: Select all
2 3 . . . 7 . . 4 . . 4 . . 7 2 3 . . . 7 . . 4 . . 1 . . 1 . . 4 . . 7 . . 4 . . 1 8 9 . 8 9 . . . 1 . . 4 1 2 3 . . . . . . swap stacks . . . . . . 1 2 3 . . . . . . . . . 1 and 3 . . . . . . . . . . . . . . . 7 8 9 <-- --> 7 8 9 . . . . . . . 1 2 9 . . 6 . . 6 . . 9 . . . 1 2 9 . . 6 . . 3 . . 3 . . 6 . . 9 . . 6 . . 3 . . . 7 8 . 7 8 3 . . 6 . .

Nice find, blue.
With all the symmetries (each clue forces 4 or 8) i would not have expected a minimal puzzle at all.

The funny thing is, that we have
- quarter symmetry (4)
- half turn (2)
- diagonal (2)
- antidiagonal (2)
- column sticks (2)
- row sticks (2)

8 automorphisms

but only 2 single symmetries, diagonal and sticks, imply them all.

by **Serg** » Thu May 21, 2015 3:59 pm

Hi, blue!

blue wrote:Here's a puzzle that can have either, but not both at the same time.
Does one "look more symmetric" than the other ?

Code: Select all
2 3 . . . 7 . . 4 . . 4 . . 7 2 3 . . . 7 . . 4 . . 1 . . 1 . . 4 . . 7 . . 4 . . 1 8 9 . 8 9 . . . 1 . . 4 1 2 3 . . . . . . swap stacks . . . . . . 1 2 3 . . . . . . . . . 1 and 3 . . . . . . . . . . . . . . . 7 8 9 <-- --> 7 8 9 . . . . . . . 1 2 9 . . 6 . . 6 . . 9 . . . 1 2 9 . . 6 . . 3 . . 3 . . 6 . . 9 . . 6 . . 3 . . . 7 8 . 7 8 3 . . 6 . .

Note: column and row stick symmetries, are also present.
I think it's the only (truly) minimal puzzle of its kind.
Too bad it's solvable with singles.

Impressive example! I couldn't imagine such possibility

.
I should think better ...

Serg

by **eleven** » Thu May 21, 2015 8:06 pm

I looked for puzzles with DD+JD symmetries (12 automorphisms). There seem to be only 5 grids, which have it, with 12,36,72,108 and 648 (MC grid) automorphisms.
The 169 puzzles i found in 1/2 an hour, all are easy, and only four are minimal.
The pattern i liked most, was this one (not minimal, singles).

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

The "hardest" one was

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

This is one of the minimal puzzles

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

by **blue** » Fri May 22, 2015 3:14 pm

eleven wrote:I looked for puzzles with DD+JD symmetries (12 automorphisms). There seem to be only 5 grids, which have it, with 12,36,72,108 and 648 (MC grid) automorphisms.

Hi eleven,

Nice work !
There are some additional grid and puzzle types, that I'm guessing you wern't looking for (?).
The JD's in your examples, have 3 cycles of 3 digits for the relabeling part the automorphism.
There are also varieties with 0,1 and 2 cycles of 3 digits.
There's a puzzle for each, shown below (with DD symmetry too).

Something similar happens for Gliding Rows (SR), DBS (D+BS), and BxCxSR (BxCx+SR) symmetries.

For SR, it's 0,1,2 or 3 cycles of 3.
For DBS and BxCxSR, the options are:

I like to think that it expands the 26 "basic" symmetries to 38, but I'm sure I'ld get a little feedback on that idea.

---

DD+BS puzzles ...

Two cycles of 3, and 3 fixed digits:

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

One cycle of 3, and 6 fixed digits:

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

9 fixed digits:

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

by **eleven** » Fri May 22, 2015 9:43 pm

Thank you blue,

i should have known that, at least i remember, that we had JD's (and gliding rows) without digit changes, probably in the MC grid. But i forgot to think about it here.

The 26 "basic symmetries" come from Red Ed's classes, and these are only about cell automorphisms, with cell cycles, not digit cycles.
But it's an option to split them into possible digit cycles, of course.

by **champagne** » Sat May 23, 2015 4:43 pm

some results on the DD symmetry.

The minimal count on my side confirms blue's results, 50781 puzzles.

I also collected puzzles having a "minimal symmetry", but not minimal stricto sensu and I found 1 906 799 puzzles.

The highest number of clues for such puzzles was 42 with 12 puzzles

Hidden Text: Show

Only hard puzzles in that group with a very small number of redundant clues are of interest. I'll look for such puzzles.

The process applied does not use the solution grids, but I have still one test to run to see what is the best process in that family.

by **Serg** » Sun May 24, 2015 8:12 am

Hi, champagne!

champagne wrote:some results on the DD symmetry ...

Great work, congratulations!

champagne wrote:I also collected puzzles having a "minimal symmetry", but not minimal stricto sensu and I found 1 906 799 puzzles.
The highest number of clues for such puzzles was 42 with 12 puzzles ...

Very interesting! What about "low clue" end of these "minimal symmetric" puzzles?

Serg

by **champagne** » Sun May 24, 2015 1:20 pm

Serg wrote:What about "low clue" end of these "minimal symmetric" puzzles?
Serg

the lowest number of clues for such a symmetry of given was 22.
EDIT something wrong in my search, I gave page 5 of that thread the list of such puzzles or size 20

Hidden Text: Show

here the 658 puzzles of size 22

a quick check gives (not checked for ED elementary puzzles, but not likely to come at depth -1)

1460 minimal puzzles oz size 21
546 puzzles of size 20
150 puzzles of size 19
24 puzzles of size 18

Hidden Text: Show

champagne

by **Serg** » Mon May 25, 2015 10:13 am

Hi, blue!

blue wrote:There are 23391 canonical grids, that can be made to show diagonal symmetry ...

I encountered problems trying to confirm your number of e-d solution grids which can be morphed to main diagonal symmetric form. It's very likely that I missed something ...

First I searched for all main diagonal symmetric solution grids having following digits in places:

Code: Select all: 1 . . 5 . . . . . . 5 . . . . . . . . . 9 . . . . . . 5 . . 1 2 3 . . . . . . 4 5 6 . . . . . . 7 8 9 . . . . . . . . . 1 . . . . . . . . . 5 . . . . . . . . . 9

I got 173992 solution grids. Then I canonicalized them by GridChecker deleting duplicated grids. And I got 22682 e-d solution grids. It is less than your 23391 number ... Where is my mistake? Could you publish your 23391 canonical grids, that can be made to show diagonal symmetry?

Serg

by **blue** » Mon May 25, 2015 3:16 pm

Serg wrote:First I searched for all main diagonal symmetric solution grids having following digits in places:
Code: Select all
1 . . 5 . . . . . . 5 . . . . . . . . . 9 . . . . . . 5 . . 1 2 3 . . . . . . 4 5 6 . . . . . . 7 8 9 . . . . . . . . . 1 . . . . . . . . . 5 . . . . . . . . . 9

I got 173992 solution grids. Then I canonicalized them by GridChecker deleting duplicated grids. And I got 22682 e-d solution grids. It is less than your 23391 number ... Where is my mistake?

Hi Serg,

You aren't getting results for grids containing this arrangement of 5's:

Code: Select all: 1---------5 . . . . 5 . . . | . . . . . 9 . . | 5 . . . . . 1 . | . . 5 . . . . 5 | . . . 5 . . . . 9 . . . . . 5 . . . 1 . . . . . . . . . 5 . . . . 5 . . . . 9

For each of the off-diagonal 5's, the row and column containing the 5, intersect the main diagonal at cells containing distinct digits (1 & 9).

Best Regards,
Blue.

Added: I think the difference in counts is as low as it is, because the 1's and 9's would need to be in a similar situation. It does happen, though. Each of these, has an extension to a complete grid with diagonal symmetry:

Code: Select all: 1 . . . 9 5 . . . . 5 . . . 1 9 . . . . 9 . . . 5 1 . . . . 1 . . . 9 5 9 . . . 5 . . . 1 5 1 . . . 9 . . . . 9 5 . . . 1 . . . . 1 9 . . . 5 . . . . 5 1 . . . 9

Code: Select all: 1 . . | . . 5 | . 9 . . 5 . | 9 . . | . . 1 . . 9 | . 1 . | 5 . . ------+-------+------ . 9 . | 1 . . | . . 5 . . 1 | . 5 . | 9 . . 5 . . | . . 9 | . 1 . ------+-------+------ . . 5 | . 9 . | 1 . . 9 . . | . . 1 | . 5 . . 1 . | 5 . . | . . 9

by **champagne** » Mon May 25, 2015 3:35 pm

an example of "hard puzzles" not far from a "symmetry of given", and the solution has a symmetry of given.
How to solve them ?

12.3.....4....5.7.....4.3..7..1...5...2...8...5...9..3....6.....3.5....6.....7.89 _1
5...2.4...1.....5....3....8..714....4..2.8..6....693..2....7..........9...6.8...5 _2
12.3.....4....2.3...5.8.7..7..1...6...6...4...4...9..3..3.2.5.....8....6.....7.89 _3

Code: Select all: 12.3..... 4....5.7. ....4.3.. 7..1...5. ..2...8.. .5...9..3 ....6.... .3.5....6 .....7.89 5...2.4.. .1.....5. ...3....8 ..714.... 4..2.8..6 ....693.. 2....7... .......9. ..6.8...5 12.3..... 4....2.3. ..5.8.7.. 7..1...6. ..6...4.. .4...9..3 ..3.2.5.. ...8....6 .....7.89

by **blue** » Mon May 25, 2015 3:36 pm

champagne wrote:some results on the DD symmetry.

The minimal count on my side confirms blue's results, 50781 puzzles.

I also collected puzzles having a "minimal symmetry", but not minimal stricto sensu and I found 1 906 799 puzzles.

The highest number of clues for such puzzles was 42 with 12 puzzles

Nice Work !

champagne wrote:Only hard puzzles in that group with a very small number of redundant clues are of interest.

Are you sure ? eleven's 27-clue "hardest" non-minimal puzzle with DD+JD symmetry (here), has 27 redundant clues.

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

Regards,
Blue.

by **champagne** » Mon May 25, 2015 5:23 pm

blue wrote:
champagne wrote:Only hard puzzles in that group with a very small number of redundant clues are of interest.

Are you sure ? eleven's 27-clue "hardest" non-minimal puzzle with DD+JD symmetry (here), has 27 redundant clues.

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

Regards,
Blue.

Hi blue,

This was a "general sentence".

having no clear definition of what is "a hard puzzle", the door is open.

I selected puzzles with a rating over 9.0 but up to 10.x in the small lot posted above.

another point is the number of redundant clues.

In my understanding, it is the number of clues you can simultaneously kill and still have a valid puzzle. You are clearly using another definition. In the lot I posted, I did not pass 2 redundant clues in depth.

One missing information about that work, the final code piling all information we shared could find all puzzles in the range 18_37 clues (all your valid puzzles) in less than 4 hours. I did not test the final code in the range 38_42 clues.

by **Leren** » Mon May 25, 2015 8:56 pm

Champagne wrote :
an example of "hard puzzles" not far from a "symmetry of given", and the solution has a symmetry of given.
How to solve them ?

12.3.....4....5.7.....4.3..7..1...5...2...8...5...9..3....6.....3.5....6.....7.89 _1
5...2.4...1.....5....3....8..714....4..2.8..6....693..2....7..........9...6.8...5 _2
12.3.....4....2.3...5.8.7..7..1...6...6...4...4...9..3..3.2.5.....8....6.....7.89 _3

Not sure what you mean here. All 3 puzzles have anti-diagonal symmetry and only the first one is still hard after the symmetry move.

Leren

by **Serg** » Mon May 25, 2015 11:02 pm

Hi, blue!
Thank you for clarification. One should consider 2 variants of "outer" digits "5" placement:

Code: Select all: 1 . . 5 . . . . . Variant 1 . 5 . . . . . . . . . 9 . . . 5 . . 5 . . 1 2 3 . . . . . . 4 5 6 . . . . . . 7 8 9 . . 5 . . 5 . . . 1 . . . . . . . . . 5 . . . . . . 5 . . 9 1 . . . . 5 . . . Variant 2 . 5 . . . . . . . . . 9 . . . 5 . . . . . 1 2 3 . . 5 . . . 4 5 6 . . . 5 . . 7 8 9 . . . . . 5 . . . 1 . . . . . . . . . 5 . . . . 5 . . . . 9

blue wrote:There are 23391 canonical grids, that can be made to show diagonal symmetry ...

Now I am able to confirm your number of e-d solution grids which can be morphed to main diagonal symmetric form.

Serg

The New Sudoku Players' Forum

Symmetrical Givens

Re: Symmetrical Givens

Re: Symmetrical Givens

Re: Symmetrical Givens

Re: Symmetrical Givens

Re: Symmetrical Givens

Re: Symmetrical Givens

Re: Symmetrical Givens

Re: Symmetrical Givens

Re: Symetrical Givens

Re: Symetrical Givens

Re: Symmetrical Givens

Re: Symmetrical Givens

Re: Symmetrical Givens

Re: Symmetrical Givens

Re: Symetrical Givens