The New Sudoku Players' Forum

[ixsetf]

Hope this isn't too tedious.

Code: Select all

+-------+-------+-------+
| 9 2 . | . . . | . 7 . |
| 4 . . | 9 . . | 3 . 2 |
| . . 6 | . . . | . 4 . |
+-------+-------+-------+
| . 9 . | 1 2 . | . . . |
| . . . | 4 . 6 | . . . |
| . . . | . 8 . | 5 . . |
+-------+-------+-------+
| . 7 . | . . 5 | 4 . . |
| 3 . 2 | . . . | . . 6 |
| . 4 . | . . . | . 8 3 |
+-------+-------+-------+


Play online

[Leren]

Code: Select all

*--------------------------*
| 9 2 3  | 5 4 1  | 6 7 8  |
| 4 5 7  | 9 6 8  | 3 1 2  |
| 1 8 6  | 2 3 7  | 9 4 5  |
|--------+--------+--------|
| 7 9 5  | 1 2 3  | 8 6 4  |
| 8 3 1  | 4 5 6  | 2 9 7  |
| 2 6 4  | 7 8 9  | 5 3 1  |
|--------+--------+--------|
| 6 7 8  | 3 1 5  | 4 2 9  |
| 3 1 2  | 8 9 4  | 7 5 6  |
| 5 4 9  | 6 7 2  | 1 8 3  |
*--------------------------*

Well, there's your solution. How did I get there - well let's just say brute force forcing chains

Leren

[Ngisa]

Code: Select all

I have the following grid
+---------------+------------------+----------------+
| 9   2   138   | 3568  1346 1348  | 168  7    158  |
| 4   5   178   | 9     167  178   | 3    16   2    |
| 178 138 6     | 23578 137  12378 | 189  4    1589 |
+---------------+------------------+----------------+
| 578 9   34578 | 1     2    37    | 68   36   478  |
| 128 138 1378  | 4     5    6     | 1289 1239 1789 |
| 127 6   1347  | 37    8    9     | 5    123  147  |
+---------------+------------------+----------------+
| 168 7   189   | 2368  1369 5     | 4    129  19   |
| 3   18  2     | 78    1479 1478  | 179  5    6    |
| 156 4   159   | 267   1679 127   | 1279 8    3    |
+---------------+------------------+----------------+


I used the SL 4 in r4 which lead to a conflict in BOX 8, so, it should be in r4c9 and tried 3 in r4c6; stte.
Is this a brute force chain? Or is it to say that not all Sudoku puzzles have a logical solution? I would like to see your brute force forcing chain Leren.

[SteveG48]

Or is it to say that not all Sudoku puzzles have a logical solution?

That would depend on what you accept as a logical solution wouldn't it?

[eleven]

I used the SL 4 in r4 which lead to a conflict in BOX 8, so, it should be in r4c9 and tried 3 in r4c6; stte.
Is this a brute force chain? Or is it to say that not all Sudoku puzzles have a logical solution? I would like to see your brute force forcing chain Leren.

Your first move is (could be written as) a contradiction net and is widely accepted as logical elimination.
Trying a candidate, which leads to a solution, is called guessing, and widely not accepted as logical solution (though it can be argumented as such, if you know, that the puzzle only has one solution - as needed for uniqueness techniques too).
It also is widely accepted, that the solutions for the hardest known puzzles, which are calculated by solvers using human-like solution techniques, and mostly contain nested (2 level) contradiction chains/nets, are logical solutions. But normally it is just boring to follow them.
The fast brute force solvers, which mostly work backtracking "guessing" still possible candidates, of course also give logically correct solutions, but these are widely not accepted, because they are practically unreadable for humans.

[champagne]

Well, there's your solution. How did I get there - well let's just say brute force forcing chains

Leren

My solver has nothing exiting in that puzzle quoted 8.5 by skfr.

May be ixself has a better path

[ixsetf]

The solution I have is vaguely:
r3c63=> r8c5 both 4 and 9, remove 3 from r3c6,
r2c81=>r8c7 everything off. This chain requires noticing a hidden pair midway through to show r9c39.

I know this is leaving out a lot of detail, but the second chain is pretty long and it will take me some time to write it out with the correct notation. (Mostly since I am not super familiar with it)

[JC Van Hay]

Analysis of ixsetf's solution : r2c8=6 -> contradiction :=> r2c8=1; stte

Note : 6r2c8 is at the vertex of a Y cluster of 6s and 1 is only given once at the start.
Therefore, r2c8=16 is a good starting point to analyse the puzzle even though it is not the only one.
For example : the SL of the strongly coupled hub cells : (2-5)r3c4=(5-9)r3c9, (3-2)r7c4=(2-9)r7c8, (4-9)r8c5=(9-7)r8c7=(7-2)r9c7, (6-5)r9c1=(5-9)r9c3.

Code: Select all

+---------------------+--------------------------+---------------------+
| 9      2      1(38) | 3568     136(4)  (1348)  | (168)  7     158    |
| 4      5      17(8) | 9        (167)   (178)   | 3      1-6   2      |
| 17(8)  1(38)  6     | 378(25)  (137)   78(123) | 18(9)  4     18(59) |
+---------------------+--------------------------+---------------------+
| (578)  9      34578 | 1        2       (37)    | 68     (36)  478    |
| 1278   (138)  1378  | 4        5       6       | 1289   1239  1789   |
| 127    6      1347  | 37       8       9       | 5      123   147    |
+---------------------+--------------------------+---------------------+
| 168    7      189   | 2368     1369    5       | 4      129   19     |
| 3      8(1)   2     | 78       (1479)  78(14)  | 7(19)  5     6      |
| (156)  4      159   | 267      (1679)  27(1)   | 1279   8     3      |
+---------------------+--------------------------+---------------------+


r2c8=6 ->

r4c8=3, r4c6=7;
3r1c6=(3-2)r3c6=(2-5)r3c4=(5-9)r3c9=9r3c7-9r8c7=(9-4)r8c5=4r1c5-(4=3)r1c6;
r23c5=17, r2c6=8;
r89c6=41, r89c5=96, r94c1=58;
r3c2=3;
r1c3=8 & r5c2=1=r8c7; r1c7={}