The New Sudoku Players' Forum

[FLOCKMAN]

One of my hobbies is creating difficult Sudokus.
A recent one is ..36....1.4..1..3.7....59.......36......8..2....4....7..1.....9.7..2..8.3....95..
If you load it you will see that there are no isolated clues and a pattern is formed. I call it Snail Trails.
Can anyone tell me if it is an original and if so, can it be listed?
I would like to canonicalise it but don’t have the program to do it. Can anyone tell me where I can get it, preferably by sending me a link to it? It grades quite high in Sudoku Explainer.
Regards,
David Filmer MA (Cantab)

[StrmCkr]

..36....1.4..1..3.7....59.......36......8..2....4....7..1.....9.7..2..8.3....95..

Er 1.21 rated it as a 10.6, pretty difficult if its an isomorph of a known puzzle or known already i couldn't tell you, i don't have a data base to cross check.

as for canonical form: that depends on how you are defining what is canonical? box (1..9) or Row (1-9) or Col (1-9) as the starting area for digits 1-9


gsf program has built command line features to do that and a lot more.

[Pat]

..36....1.4..1..3.7....59.......36......8..2....4....7..1.....9.7..2..8.3....95..

Can anyone tell me if it is an original

isomorph of

1....7.9..3..2...8..96..5....53..9...1..8...26....4...3......1..4......7..7...3..

by ArtoI (2006.Nov.3)

[tarek]

I can't believe the choice of title for this thread knowing now what the original puzzle is

Tarek

[eleven]

Seems to be a new trend to repost Arto's old puzzles.
If someone finds a 10.6+ puzzle by himself, the chance is far less than 1:10000 to get an equivalent of Arto's puzzles.
Too stupid.

[tarek]

This guy posts puzzles to websites claiming that they are his. I remember that he posted an isomorph of one my puzzles to the sudokuwiki.org (Andrew Stuart site)

The problem is that if left unchallenged, people will start crediting him with these puzzles

[FLOCKMAN]

I genuinely did create the Snail Trial myself. I try to create Sudokus that have a visually pleasing look to them and this one had two intersecting tracks.
Eleven commented "If someone finds a 10.6+ puzzle by himself, the chance is far less than 1:10000 to get an equivalent of Arto's puzzles."
I had no idea of puzzles created by Arto, so the 1:10,000 chance really did come off. It seems that whatever puzzles I create, someone has already created a morph of it. I looked at Arto's puzzle and the shape of it is nowhere near the shape of the Snail Trail, So what would be the chance of permuting his puzzle to randomly produce the Snail Trail?
So I would like to ask the experts, what is the chance that anyone can create a new Sudoku puzzle that is not a morph of one that someone else has already created before?

[tarek]

Hi Flockman,

If you start from a known puzzle and then re-shuffle it to have a more pleasing look then that is fine but IMHO you should be crediting the original poster or at least say that it is taken from somewhere else. I have a feeling that you had good intentions when you started this years ago. You have posted elsewhere puzzles which are isomorphs of puzzles not originally posted by you. Too many coincidences if you ask me

I'm not sure if you are aware of the time and hard work some of the original posters took to post those puzzles.

Tarek

[tarek]

If you are doing vicinity searches next to these known puzzles then your chances are high in stumbling across an already known puzzle. To build a puzzle from scratch then getting the Escargot or Golden nugget is near impossible IMO

A simple way of posting unique puzzles is to have a database of puzzles in canonical form (many lists can be found on this forum but the still need canonicalisation) which you cross reference your candidate (still in canonical form) with. If it is unique then shape it as you please and post it as yours.

Tarek

[StrmCkr]

So I would like to ask the experts, what is the chance that anyone can create a new Sudoku puzzle...

that is antori

from scratch with no reference?

1: 6,670,903,752,021,072,936,960 {if u want the exact same grid}
1: 5,472,730,538 { if you consider issomorph }

starting with the same puzzle, simply re-scrambling it
1: 3,359,232

but only if you use all the 3!^8*2 Sudoku preserving symmetries

so its extremely low so low that's its anything but hard pressed to say you did not start with this puzzle and re-morph it,
given the other threads you are posting in playing around with re-morphing other known puzzles; I'm more keen on the fact you did the same to this one.

to create a unique puzzle that is not posted some where period. I doubt any single 1 person has every grid published to date stored in a single location to cross check.