The New Sudoku Players' Forum

[koushanejad74]

Benoku is a logic-based number-placement puzzle. It is distinct from but shares some properties and rules with Str8ts and Sudoku.
Rules:
• All sudoku rules apply.
• Some rows and columns are divided into two compartments by a gray cell
• Each compartment, vertically or horizontally, must contain a straight – a set of consecutive numbers, but in any order. For example: 7, 6, 4, 5 is valid, but 1, 3, 8, 7 is not.

Benoku_Easy_001.png (16.16 KiB) Viewed 1218 times

Benoku Definition
Benoku Sample (Easy)

[SpAce]

Your rules as stated are ambiguous and confusing, or maybe I'm too stupid to understand them. Furthermore, even if I make some assumptions about them (like that a gray cell must be either 1 or 9 and actually part of one or the other compartment), I can't solve even this "easy" example. Can you show a walk-through of the logic you're expecting to be used here?

Here's as far as I get with basic techniques (or anything I'm willing to do anyway):

Code: Select all

.-----------------.------------------------.---------------------.
|  2     78   9   | 1       345    345     | 345     6       78  |
|  3     678  567 | 24579   4578   24589   | 1245    1245    789 |
| [1]    4    57  | 2579    6      23589   | 235     235     789 |
:-----------------+------------------------+---------------------:
|  467   567  3   | 45679   2      45689   | 45679   4578   [1]  |
|  467   567  12  | 45679   34578  1345689 | 245679  24578   245 |
|  8    [9]   12  | 4567    457    1456    | 24567   2457    3   |
:-----------------+------------------------+---------------------:
|  9     1    8   | 2456    45     2456    | 37      37      245 |
|  5     2    4   | 3      [1]     7       | 8       9       6   |
|  67    3    67  | 8       9      245     | 1245    1245    245 |
'-----------------'------------------------'---------------------'


Have I made mistakes? If not, how do you proceed from there? It seems to me that all the compartment rules have already been exhausted, so it's just vanilla sudoku rules left -- and with just those this is not possible to solve. (Or more accurately, it has over 500 solutions according to SudokuWiki, so it can't be solved logically. The same thing with your other example in the Definition page. I'm thinking the published solution must be just one of many. Did you ever check that?)

What am I missing? It must be something obvious since you call this easy... And please don't give me a similar non-answer as you did here. I'm still awaiting a further explanation for that puzzle as well.

PS. I'm starting to suspect that you don't even check that your puzzles have a single solution. If so, that's simply unforgivable, because nothing is more frustrating for a manual solver (except maybe that a puzzle has no solution at all). It's already been clear that you don't have a proper way to assess the difficulty of your puzzles because you don't have a solver that uses logical techniques. That's annoying too but less severe. At the very least you should have a brute force solver that checks for multiple solutions before you publish anything. Until you address this suspicion, I won't be trying more of your puzzles.

[eleven]

SpAce,

you completeley misunderstood that.
There is no claim, that the puzzles would have a unique solution. They are easy, because almost every guess leaves a valid solution

[SpAce]

SpAce,

you completeley misunderstood that.
There is no claim, that the puzzles would have a unique solution. They are easy, because almost every guess leaves a valid solution

You're right. My bad

[koushanejad74]

SpAce,

you completeley misunderstood that.
There is no claim, that the puzzles would have a unique solution. They are easy, because almost every guess leaves a valid solution

Eleven,

Although I didn't mention it, the puzzle MUST have a unique solution, have you found more than one solution?

Thanks,

-Kousha

[Wecoc]

The puzzle provided in the example (Benoku Definition) has multiple solutions, or at least I don't see why any of those would be invalid.
Here are two more solutions, and there may be more.

Code: Select all

+-------+----------+----------+
| 4 3 5 |  9 67 67 | 18  2 18 |
| 8 9 7 |  1  2  3 |  6  4  5 |
| 2 1 6 |  8  4  5 |  9  3  7 |
+-------+----------+----------+
| 6 4 9 |  7  1  8 |  3  5  2 |
| 5 7 8 | 36 36  2 |  4  1  9 |
| 3 2 1 |  5  9  4 | 78 78  6 |
+-------+----------+----------+
| 7 5 3 | 46  8 16 |  2  9 14 |
| 9 8 4 |  2  5 17 | 17  6  3 |
| 1 6 2 | 34 37  9 |  5 78 48 |
+-------+----------+----------+

Image: Show

[Mathimagics]

Agreed! The rules seem clear and not ambiguous, but nevertheless SpAce is correct with his puzzle state as listed above. There are many ways to complete it.

Wecoc has found a similar problem with the linked example.

So either the rules are missing some vital extra constraint, or the puzzle examples as presented are ambiguous. If the poster is using software to generate and/or check them, then I suggest that software is dodgy ...

[Wecoc]

I liked the Blocku variant better because of 2 reasons:
- In this one all blocks must be given, which spoils the chance for harder puzzles like in the other one
- While trying to solve the example, I noticed the candidates on the cells with blocks are very limited

They can only be 19, except for 5 coordinates in the entire grid:

Code: Select all

+-------+---------------+-------+
| . . . |   .   .   .   | . . . |
| . . . |   .   .   .   | . . . |
| . . . |   .   .   .   | . . . |
+-------+---------------+-------+
| . . . | 1469  .  1469 | . . . |
| . . . |   .  159  .   | . . . |
| . . . | 1469  .  1469 | . . . |
+-------+---------------+-------+
| . . . |   .   .   .   | . . . |
| . . . |   .   .   .   | . . . |
| . . . |   .   .   .   | . . . |
+-------+---------------+-------+

I did a fast check with my solver on the other idea I gave about adapting the Blocku rules on a Sudoku grid using 9s as blocks. As far as I tried it looks like it's impossible.
As I said, the constrain with consecutive digits on both sides (and both axis) is too restrictive, and adding the 3x3 box rule it always results on a crash.

[koushanejad74]

My apologies, specially to space
I'm afraid I've posted the wrong one, the algorithm I have ensures the solution is unique, I even tried it on paper before posting it, but then I uploaded another one. will correct it soon,

[Leren]

In the meantime, if I understand Benoku correctly you can make the solution unique by adding two extra clues : 4 in r5c5 & 6 in r5c7. The puzzle is easy to solve and the unique solution is

Code: Select all

289135467
376482159
045769328
463528970
751943682
802671543
918256734
524307896
637894215

The 57's in r67c57 are OK, because he says that all Sudoku rules apply, so Box 8 can't have two 7's' and Box 9 can't have two 5's.

Leren

[Mathimagics]

Confirmed. Nice job, Leren!!

[Leren]

Well, seeing as I appear to be on a roll here, you can also add r3c5 & r6c5 = 5 for a unique solution

Code: Select all

289135467
365974218
047268359
473629580
651783924
802451673
918546732
524307896
736892145


This one appears to be harder to solve, requiring some chaining, or some UR moves, if I'm not mistaken.

Leren

[koushanejad74]

As promised, a corrected sample has been posted here
{Benoku (Part II) -- URL corrected by Moderator}

[Leren]

For some reason I can't get koushanejad74's here link to work. In the meantime here is my own Benoku puzzle.

Code: Select all

  7    0    0    0    1    0    0    0    0 
  0    3    0    0    0    0    0   [0]   0 
  0    0    0    0    0    4    0    0    0 
  0    0    0   [0]   0    0    0    0    0 
  1    0    0    0    0    9    0    0    2 
  0   [0]   0    2    0    0    0    0    0 
  2    0    0    0    6    7    0    0    3 
  0    0    0    0    0    0    0    2    0 
  0    0    0    0    0    0   [0]   0    0

Leren

<Edit> Improved drawing style of puzzle and changed to be Benoku compliant.

[SpAce]

Code: Select all

........9
.3.....X.
.....4...
...X7....
7....8...
.X.....5.
......5..
..6..2...
..3...X..


What have I missed if I don't understand what can go in r2c8?