The New Sudoku Players' Forum

[Havard]

Here are four 17s I could not find in the lists:
#

Code: Select all

..............12...34..5...........4.6...3...2..1..7......8..3........6.8..72....
..............12...34..5...........5.6...3...2..1..7......8..3........6.8..72....
..............12.3.45..6.............3...4...2..1..7......8..4........3.8..72....
.............1.2.34...56..................56..2.37......1........52.....3..8...4.


#

According to the updated list from Gordon, these are unfortunately not new.

Havard

[ronk]

I have strong hopes that we will see 40000 in the near future!

Does that mean your 2-off 2-on search is ongoing? Or have you deepened the search?

[Havard]

I have strong hopes that we will see 40000 in the near future!

Does that mean your 2-off 2-on search is ongoing? Or have you deepened the search?

It is ongoing. It has been a bit of on and off with a mixture of server-problems and bugs in the program, but now it is going! Should finish in 11 sec * 39200 sudoku = 5 days.

[gsf]

Here are four 17s I could not find in the lists:
#

Code: Select all

..............12...34..5...........4.6...3...2..1..7......8..3........6.8..72....
..............12...34..5...........5.6...3...2..1..7......8..3........6.8..72....
..............12.3.45..6.............3...4...2..1..7......8..4........3.8..72....
.............1.2.34...56..................56..2.37......1........52.....3..8...4.


#

According to the updated list from Gordon, these are unfortunately not new.

Havard

there were a flurry of additions due to a 1-off search in january
the most recent list before havard's additions is sudoku17-2007-01-22.can.gz

[gfroyle]

Greetings again everybody.. it feels as though I have been away for ages!

Just to fill you in, the list of 39227 was actually computed 6 months or so ago, but I was (foolishly) waiting till I got to 40000 before I did an update.

Then I lost them all when my 400Gb drive (unbacked up, nowhere to back up 400Gb of stuff) died, so I sort of forgot all about it.

Then Havard emailed me separately that the hunt for 17s was still on, and sent me some new ones, which inspired me to remember that I had only transferred my Sudoku stuff to the big drive fairly recently, and that previously it had lived on a shared disk drive that WAS backed up, and magically it could be recovered.

So I got them back, incorporated the 63 new ones that Havard sent me, found 14 more and updated the webpage at

http://people.csse.uwa.edu.au/gordon/sudokumin.php

So now there are 39304, but I predict that the next 696 will be tough to find.

I'll post again soon about the way that I've been trying to search for them, but that will have to wait till tomorrow...

Cheers

Gordon

[Havard]

hi there gfroyle. Good to have you back.

This is getting a bit confusing, since I have found quite a few more already. I'll keep posting them, cross refrenced with your 39227 list here: http://forum.enjoysudoku.com/viewtopic.php?p=43947#p43947

can you sepratly post those 14 extras you found?

Havard

[coloin]

Welcome back too.

We may be approaching the end of this saga........or not.

Havard is now doing a 2off/2on search of the whole catalogue.

In particular I think it would be worth noting whether any grids exist which dont have any 17s within a 2off/2on distance.
If there are any then it does really beg the question "how did we [Gordon] find it ! ?"

If we have identified one of these remote beasts there possibly exists the possiblity that these will be difficult to find ! [As Gordon has warned !]

I suppose we can be reassured that it is extremly likely that all 17puzzles that we find will be within 3off/3on of another 17puzzle- or has Gordon got an example of one of those as well ?

It will be interesting to see your search methods - maybe Havard's "blunderbus" approach can advance things much furthur. [? to a conclusion ?]

C

[tarek]

Good work Havard.........

But shouldn't you include your new finds in the search & loop with every batch of new finds until there are 0 new finds......


tarek

[Havard]

Good work Havard.........

But shouldn't you include your new finds in the search & loop with every batch of new finds until there are 0 new finds......

tarek

And what makes you think that I am not doing that? I have a big supercomputer making the main search, and every now and then I collect the output from it, and run that on my little laptop to check the new ones.

[gsf]

This is getting a bit confusing, since I have found quite a few more already.

I just merged the 2007-01-22 results with havard's recent (dated 2007-05-13) and then gordon's (dated 2007-05-14)
gordn's recent covered havard's and the 2007-01-22 updates, but I retained the posted dates and attributions
gordon's recent post missed this one from his original list

Code: Select all

7..2....51...9.....8.....4.3.....6.....4.8......5.........6.3...52.......4.......


(I do believe there are gremlins at play collating these 17's)

the new collation contains these space separated fields fields
< canonical-puzzle canonical-grid original-puzzle original-ordinal contributer date >
where canonical is row-order minlex based on the solution grid
and original-ordinal is 0 for the latest contributions
http://www.research.att.com/~gsf/sudoku/data/sudoku17-2007-05-14.can.gz

[Havard]

(I do believe there are gremlins at play collating these 17's)

Maybe you could help confirm that the new number now is 39382.

Havard

[gfroyle]

gordon's recent post missed this one from his original list

Code: Select all

7..2....51...9.....8.....4.3.....6.....4.8......5.........6.3...52.......4.......


(I do believe there are gremlins at play collating these 17's)

This is #36855 on the list of 39304 that I referred to last post....

[gfroyle]

hi there gfroyle. Good to have you back.

This is getting a bit confusing, since I have found quite a few more already.

Havard

Sorry, my fault... I did not add *ALL* of the new ones that you found to the list...

You are correct that the list should have contained 39382.

However putting your new ones into *my* program has now increased the number a bit more... I am up to 39393 now.

BTW, gsf's file of 2007-05-14 contains 39305 lines of which 1 line is duplicated....

More updates later..

Gordon

[Havard]

However putting your new ones into *my* program has now increased the number a bit more... I am up to 39393 now.

hm, we will hit 40000 quite quickly with this speed... (your online catalog is still on 39304)

[gsf]

gordon's recent post missed this one from his original list

Code: Select all

7..2....51...9.....8.....4.3.....6.....4.8......5.........6.3...52.......4.......


(I do believe there are gremlins at play collating these 17's)

This is #36855 on the list of 39304 that I referred to last post....

thanks
once again it takes two posts for me to get it right
the 2007-01-22 data I posted used a different canonical form than the 2007-05-14
(which uses row-order minlex based on the solution grid)
I used unix join(1) to find new entries / diffs / omissions on files sorted on different canonical forms
its amazing more diffs didn't show up

so
I reposted both catalogs using the new row-order minlex canonical form
http://www.research.att.com/~gsf/sudoku/data/sudoku17-2007-01-22.can.gz
http://www.research.att.com/~gsf/sudoku/data/sudoku17-2007-05-14.can.gz

the 2007-05-14 catalog is 2007-01-22 + havard-2007-05-13 + gfroyle-2007-05-14 = 39304 entries
which confirms that gfroyle covered the 2007-01-22 additions as well as havards
I kept the 2007-01-22 and havard labels and attributed the rest to gfroyle with date 2007-05-14

gfroyle, you had noted on your site that an attributed catlog would be nice
I hope this comes close, but I bet you have better records from the early days than I gleaned off the forums