Grids containing a 21 but no 20

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

Re: Grids containing a 21 but no 20

Postby Mathimagics » Tue Jul 23, 2019 2:32 pm

Thanks blue!

The time I gave was correct, but I must have been running an LCT catalog update at the time - when this goes to dump an updated band to the slow disk, some peculiarity of the IO system means that other processes get locked out, even if they aren't doing IO, and this can be for up to (and clearly sometimes over) a minute ...

The automorphic grid table is interesting ...
blue wrote:We still don't have a "no-668" grid, that only has the non-trivial automorphism.

BlueMagic will hopefully find one, it's not on active duty just yet ...
User avatar
Mathimagics
2017 Supporter
 
Posts: 1926
Joined: 27 May 2015
Location: Canberra

Re: Grids containing a 21 but no 20

Postby Mathimagics » Wed Jul 24, 2019 7:00 pm

Here's a new entry for blue's table:
Code: Select all
123456789456789123789123465215937648634218957897564312361872594542391876978645231


Code: Select all
> BlueMagic Test20C.txt
no puzzle: 123456789456789123789123465215937648634218957897564312361872594542391876978645231
1 grids, 0 puzzles, 78.125 ms/g

> NotMagic Test20C.txt 999
05:58:16: [Find 20C] Processing Test20C.txt (bailout = 999s)
1....6....5...9..3....2.4......3...8..4...9....75......6.8.......23...7.9.....2.. #     43.415s


The grid has automorph count of 2, sadly. It did take a very long time to find the 20C, so perhaps it has a low total 20C puzzle count?

Also of possible interest is that this grid is the only grid with MCN = 16 (of 222,278 known cases) that failed the 668 check.
Last edited by Mathimagics on Thu Jul 25, 2019 2:33 pm, edited 1 time in total.
User avatar
Mathimagics
2017 Supporter
 
Posts: 1926
Joined: 27 May 2015
Location: Canberra

Re: Grids containing a 21 but no 20

Postby Mathimagics » Wed Jul 24, 2019 8:06 pm

I have fixed the post above which had the wrong grid quoted ...

But now I see it's NOT new at all ... 4th from the end in the table ... oh, good grief, I give up! :oops:

The observation about MCN=16 grids is, on the other hand, 100% kosher ...
User avatar
Mathimagics
2017 Supporter
 
Posts: 1926
Joined: 27 May 2015
Location: Canberra

Re: Grids containing a 21 but no 20

Postby Mathimagics » Tue Sep 01, 2020 9:13 am

It is just come to my attention that the 4 known grids which have no 20C puzzle are all automorphic. The automorphism class is shown for each grid below.

Code: Select all
123456789456789123789123456214897365365214897897365214541632978632978541978541632 # 18
123456789456789123789132465218967534564213978937548216391875642645321897872694351 #  6
123456789457189326689327154216534897745891632938672541361245978574918263892763415 #  6
123456789457189326689327154216534897745891632938672541392765418574918263861243975 #  2


Also, for the 268,296 grids with a 20C but no 19C puzzle, the automorphism rate is still quite high. There are 4815 automorphic grids, which is ~1.8%.

The background level is around 1 automorphic grid in 11,000 ...
User avatar
Mathimagics
2017 Supporter
 
Posts: 1926
Joined: 27 May 2015
Location: Canberra

Previous

Return to General