The New Sudoku Players' Forum

[Quino]

...or is it possible for a Sudoku puzzle from Sudoku.com to have two solutions?
If you choose Bangkok Post, type in the numbers 93857 as the first five numbers in this specific puzzle, you come to my problem. If you check up boxes 4, 5 and 8, it seems like the numbers 4 and 5 in these three specific boxes could replace each other, without anything happen to the rest of the puzzle. (Of course, if you change the one for the other, all the other affected 4's and 5's in these specific boxes will change accordingly.) This is a bit strange, since the puzzles from Sudoku.com is supposed to have only ONE solution, no?
Or am i completely wrong, having done some fatal error?
Please enlighten me.
/Quino

[daj95376]

I loaded Bangkok Post puzzle 93857 into Simple Sudoku and it accepted it as having one solution.

Code: Select all

 *-----------*
 |...|...|...|
 |.93|..8|.5.|
 |.78|5..|34.|
 |---+---+---|
 |1.9|.2.|...|
 |..6|...|1..|
 |...|.5.|8.6|
 |---+---+---|
 |.34|..6|51.|
 |.2.|3..|97.|
 |...|...|...|
 *-----------*


Code: Select all

 *-----------------------------------------------------------------------------*
 | 2456    1456    125     | 124679  134679  123479  | 267     2689    12789   |
 | 246     9       3       | 12467   1467    8       | 267     5       127     |
 | 26      7       8       | 5       169     129     | 3       4       129     |
 |-------------------------+-------------------------+-------------------------|
 | 1       458     9       | 4678    2       347     | 47      3       3457    |
 | 234578  458     6       | 4789    34789   3479    | 1       239     234579  |
 | 2347    4       27      | 1479    5       13479   | 8       239     6       |
 |-------------------------+-------------------------+-------------------------|
 | 789     3       4       | 2789    789     6       | 5       1       28      |
 | 568     2       15      | 3       148     145     | 9       7       48      |
 | 56789   1568    157     | 124789  14789   124579  | 246     2368    2348    |
 *-----------------------------------------------------------------------------*


[ArkieTech]

Quino,

Here is the only answer to Bangkok Post puzzle 93857

Code: Select all

 *-----------*
 |561|432|789|
 |493|718|652|
 |278|569|341|
 |---+---+---|
 |189|627|435|
 |356|894|127|
 |742|153|896|
 |---+---+---|
 |934|276|518|
 |625|381|974|
 |817|945|263|
 *-----------*


Can you see the problem?

dan

[Steve48]

The 5 in box 5 is one of the clues. It can't be changed.

[Quino]

Hello again,
and thanks for your interest in what I see as something I really can't understand. I know that you are right, but i need to understand how.
For example, the nr. 5 in box 5 can (in my opinion) be changed to a 4, and the 4s and 5s in box 4, 5 and 8 will consequently change too, of course. They don't affect any other 4s and 5s in the puzle, and they "fit" with each other in order to be exchanged for the respective other digit. I mean, what makes nr. 5 in box 5 unchangeable (for a 4)? Let me explain how I see it:
5 in box 5 becomes a 4,
4 in box 5 becomes a 5,
5 in box 4 becomes a 4,
4 in box 4 becomes a 5,
4 in box 8 becomes a 5 and
5 in box 8 becomes a 4.
So number 5 in box 5 is not locked. This makes perfect sense to me, but I am of course wrong here. But I don't see how...
Is my logic above wrong? Can't the numbers be exchanged in the way I describe it above?

Cheers,
/Q.

[ravel]

These six cells in the solution grid form an "unavoidable set". If none of the numbers of an unavoidable set is given, then the puzzle has multiple solutions. But here the 5 in r6c5 is given, so you cannot swap the numbers in these cells to get a second solution.
Each solution grid has a lot of them, of size 4 upwards, e.g. this one has another one in cells r58c46 (13), but 3 is a given in r8c4.

[Quino]

Ravel, do you mean that the specific 5 you mention was already printed in the paper? Not in the Bangkok Post where i found it. Of course, if that would've been the case, I would understand the logic. But since it wasn't in the paper...

Regards,
/Q.

[ravel]

I dont have the paper or program, but it is listed in daj95376's post above.

[Pat]

do you mean that the specific 5 you mention was already printed in the paper? Not in the Bangkok Post where i found it. Of course, if that would've been the case, I would understand the logic. But since it wasn't in the paper...

mis-prints have been known to happen

daj95376 posted the puzzle the way it should've appeared in the newspaper -- see http://www.sudoku.com/solutions.html

[Quino]

Thank you all for your help.

And what a shame that a mis-print takes up all this time...

Cheers!
/Q.

[Pat]

i should've mentioned that Pappocom guarantees symmetrical arrangement of the givens.

mis-prints are rare,
and the broken symmetry will usually expose them.

If only papers would print the puzzles from the PDF files I supply them with there would never be a problem. It's when they try to transcribe from my PDF file, using it as a data source, that problems arise.

~ Pat