The New Sudoku Players' Forum

[ArkieTech]

Code: Select all

 *-----------*
 |...|5..|4..|
 |3..|8..|.6.|
 |.7.|..2|..9|
 |---+---+---|
 |4..|.2.|...|
 |...|.3.|8..|
 |.5.|...|..6|
 |---+---+---|
 |.91|..8|..2|
 |.6.|1..|...|
 |..7|..9|..5|
 *-----------*


Play/Print this puzzle online

[SteveG48]

Code: Select all

 *-----------------------------------------------------------------------------*
 | 169     128     2689    | 5       1679   e367     | 4       12378  d1378    |
 | 3       124     2459    | 8       1479    47      | 1257    6       17      |
 | 16      7       4568    | 346     146     2       | 135     1358    9       |
 *-------------------------+-------------------------+-------------------------|
 | 4      b138    a3689    | 679     2       1567    | 13579   13579  c137     |
 | 1679    12      269     | 4679    3       14567   | 8       124579  147     |
 | 179     5      a239     | 479     8       147     | 12379   123479  6       |
 *-------------------------+-------------------------+-------------------------|
 | 5       9       1       | 3467    467     8       | 367     347     2       |
 | 2       6      b4-3     | 1       5      f347     | 379     34789  c3478    |
 | 8       34      7       | 2       46      9       | 136     134     5       |
 *-----------------------------------------------------------------------------*


(3)r46c3 = r4c2,r8c3 - r48c9 = r1c9 - r1c6 = (3)r8c6 => -3 r8c3 ; stte

[pjb]

Code: Select all

 169     128     2689   | 5      1679  *367    | 4      12378 *1378   
 3       124     2459   | 8      1479   47     | 1257   6      17     
 16      7       4568   | 346    146    2      | 135    1358   9     
------------------------+----------------------+---------------------
 4      *138     3689   | 679    2      1567   | 13579  13579 *137   
 1679    12      269    | 4679   3      14567  | 8      124579 147   
 179     5       239    | 479    8      147    | 12379  123479 6     
------------------------+----------------------+---------------------
 5       9       1      | 3467   467    8      | 367    347    2     
 2       6       4-3    | 1      5     *347    | 379    34789 *3478   
 8      *34      7      | 2      46     9      | 136    134    5     


Sashimi swordfish of 3s at r1c69, r4c29, r8c69, r9c2 => -3 r8c3; stte

Phil

[bat999]

(3)r46c3 = r4c2,r8c3 - r48c9 = r1c9 - r1c6 = (3)r8c6 => -3 r8c3 ; stte

Hi
This puzzle I solved OK, but I couldn't crack it in one step.
As a last resort I tried cheating, tried to figure out how to force a candidate that would let HoDoKu solve in one with "Solve up to".
One of them I had in mind was -3 r8c3, but no luck, had to give up.

Steve's solution is tricky for me to read from l-to-r, but from r-to-l it's sweet...

...r1c9 - r1c6 = (3)r8c6
If r8c6<>3 then r1c9<>3...
so
either r8c9=3 -> -3 r8c3
or r4c9=3, which puts the 3 of box 4 in r6c3 -> -3 r8c3

That's unusual (to me), covering both outcomes when r1c9<>3.

[daj95376]

Code: Select all

 *-----------------------------------------------------------------------------*
 | 169     128     2689    | 5       1679   e367     | 4       12378  d1378    |
 | 3       124     2459    | 8       1479    47      | 1257    6       17      |
 | 16      7       4568    | 346     146     2       | 135     1358    9       |
 *-------------------------+-------------------------+-------------------------|
 | 4      b138    a3689    | 679     2       1567    | 13579   13579  c137     |
 | 1679    12      269     | 4679    3       14567   | 8       124579  147     |
 | 179     5      a239     | 479     8       147     | 12379   123479  6       |
 *-------------------------+-------------------------+-------------------------|
 | 5       9       1       | 3467    467     8       | 367     347     2       |
 | 2       6      b4-3     | 1       5      f347     | 379     34789  c3478    |
 | 8       34      7       | 2       46      9       | 136     134     5       |
 *-----------------------------------------------------------------------------*


(3)r46c3 = r4c2,r8c3 - r48c9 = r1c9 - r1c6 = (3)r8c6 => -3 r8c3 ; stte

Hello Steve. Sorry to pick on two of your solutions in such a short time period.

One reason that I took so long to accept network steps is that people often use "selective" logic when presenting it. Your solution is such an example because you ignore an important elimination associated with your logic.

(3)r46c3 = r4c2,r8c3 - r8c6,r48c9 ; contradiction [ r1c6=3 & r1c9=3 ] => =3 r46c3

_

[bat999]

Code: Select all

.-------------------.--------------------.----------------------.
| 169    128   2689 | 5     1679  *367   | 4      12378   *1378 |
| 3      124   2459 | 8     1479   47    | 1257   6        17   |
| 16     7     4568 | 346   146    2     | 135    1358     9    |
:-------------------+--------------------+----------------------:
| 4     *138  *3689 | 679   2      1567  | 13579  13579   *137  |
| 1679   12    269  | 4679  3      14567 | 8      124579   147  |
| 179    5    *239  | 479   8      147   | 12379  123479   6    |
:-------------------+--------------------+----------------------:
| 5      9     1    | 3467  467    8     | 367    347      2    |
| 2      6     4-3  | 1     5     *347   | 379    34789   *3478 |
| 8      34    7    | 2     46     9     | 136    134      5    |
'-------------------'--------------------'----------------------'

(3)r1c9 - r1c6 = r8c6 - (3)r8c3
(3)r4c9 - r4c23 - r6c3 - (3)r8c3
(3)r8c9 - (3)r8c3
=> -3 r8c3; stte


Inspired by Steve's solution.

[SteveG48]

Code: Select all

 *-----------------------------------------------------------------------------*
 | 169     128     2689    | 5       1679   e367     | 4       12378  d1378    |
 | 3       124     2459    | 8       1479    47      | 1257    6       17      |
 | 16      7       4568    | 346     146     2       | 135     1358    9       |
 *-------------------------+-------------------------+-------------------------|
 | 4      b138    a3689    | 679     2       1567    | 13579   13579  c137     |
 | 1679    12      269     | 4679    3       14567   | 8       124579  147     |
 | 179     5      a239     | 479     8       147     | 12379   123479  6       |
 *-------------------------+-------------------------+-------------------------|
 | 5       9       1       | 3467    467     8       | 367     347     2       |
 | 2       6      b4-3     | 1       5      f347     | 379     34789  c3478    |
 | 8       34      7       | 2       46      9       | 136     134     5       |
 *-----------------------------------------------------------------------------*


(3)r46c3 = r4c2,r8c3 - r48c9 = r1c9 - r1c6 = (3)r8c6 => -3 r8c3 ; stte

Hello Steve. Sorry to pick on two of your solutions in such a short time period.

One reason that I took so long to accept network steps is that people often use "selective" logic when presenting it. Your solution is such an example because you ignore an important elimination associated with your logic.

(3)r46c3 = r4c2,r8c3 - r8c6,r48c9 ; contradiction [ r1c6=3 & r1c9=3 ] => =3 r46c3

_

No problem, Danny. This is much better than being ignored.

I certainly see that your way of writing it gives an equivalent solution, but I don't know why the "selective" logic bothers you. In many non-networked chains we ignore some links (especially weak links) that aren't necessary to reaching our conclusion. This seems similar to me.

[daj95376]

I certainly see that your way of writing it gives an equivalent solution, but I don't know why the "selective" logic bothers you. In many non-networked chains we ignore some links (especially weak links) that aren't necessary to reaching our conclusion. This seems similar to me.

It's my problem. I keep trying to treat networks and chains differently, but there's an overlap of chains into networks that trips me up all too often. Sorry!!!

_