The New Sudoku Players' Forum

[yzfwsf]

98.7..6..7...9.5....4..3...4....9.72.9.2......721.....3.......5.4...2.38.........

[Cenoman]

With help of yzfwsf's solver (presented the way I understand the pattern)

Code: Select all

 +--------------------------+------------------------------+--------------------------+
 |  9       8     B135      |  7        1245      145      |  6       124     134     |
 |  7       1236  B136      |  468      9         1468     |  5       1248    134     |
 |  1256    1256   4        | <568     <12568     3        | T12789  T1289   T179     | CL1
 +--------------------------+------------------------------+--------------------------+
 |  4      T1356   13568    | <3568    <3568      9        | <138     7       2       |
 |  1568    9      13568    |  2        345678    45678    |  1348    14568   146     |
 |  568     7      2        |  1        34568     4568     |  3489    45689   469     |
 +--------------------------+------------------------------+--------------------------+
 |  3      <126    16789    |  4689     14678     14678    |  12479   12469   5       |
 | <156     4      15679    | <569     <1567      2        | <179     3       8       | CL2
 |  12568  <1256   156789   |  345689   1345678   145678   |  12479   12469   14679   |
 +--------------------------+------------------------------+--------------------------+
            CL3                CH 56    CH 156                CH 1
   b7 CH 156

Senior exocet (1356)r12c3, r4c2, r3c789 (three cells with two locked digits as object cells)
"Cross-lines" are rows 3, 8 and column 2 (unsolved S-cells tagged with '<')
+3r12c3 => +3r4c2 (no count of cover houses for digit 3)
Cover-houses for digit 1: c57, b7; for digits 5, 6: c45, b7.

Eliminations from the pattern: -28 r3c78 (not locked, non base digits in target cells) => basics to the end.

Note 2: another choice for S-cells and CL/CH: r56c1 instead of r79c2, b4 as CL, c1 as CH.
In such case, I agree with SpAce, "cross-line" is not the right name...

[SpAce]

Code: Select all

          *56^1   \156               \1        *56      \1
.-------------------------.----------------------------.---------------------.
|  9       8       13-5   |  7        1245     *145    | 6      124    134   | \5
|  7      *1236    13-6   |  468      9        *1468   | 5      1248   134   | \6
|  1256^  *1256^   4      |  568      12568^    3      | 12789  1289   179   | ^1
:-------------------------+----------------------------+---------------------:
|  4      #13-56  *13568^ | *3568    *3568      9      | 138^   7      2     | *56^1
|  1568    9       13568  |  2        345678   *45678  | 1348   14568  146   |
|  568     7       2      |  1        34568    *4568   | 3489   45689  469   |
:-------------------------+----------------------------+---------------------:
|  3      *126^    16789  |  4689     14678    *14678  | 12479  12469  5     |
| *156^    4      *15679^ | *569     *1567^     2      | 179^   3      8     | *56^1
|  12568  *1256^   156789 |  345689   1345678  *145678 | 12479  12469  14679 |
'-------------------------'----------------------------'---------------------'
    b17:156                       \b58:56

Dual Almost-Finned-Mutant-Jellyfish

(56)r48c26\r12c3b1578 = (5|6-1|3|56)r4c2 =[basics]= ((3)r2c2 & (1)r348c2\c357b117 & (5|6)r48c26\r1|2c3b1578) - (3|1|56=!)r12c3 => -56 r4c2,r12c3; btte

note: Show

The first double-fish notation is not really valid, but I hope it communicates the intention.

as a kraken: Show

((5)r48c26\r1c3b1578 & (6)r48c26\r2c3b1578)
||
(5-1|3|6)r4c2 =[basics]= ((3)r2c2 & (1)r348c2\c357b117 & (6)r48c26\r2c3b1578) - (3|1|6=!)r2c3
||
(6-1|3|5)r4c2 =[basics]= ((3)r2c2 & (1)r348c2\c357b117 & (5)r48c26\r1c3b1578) - (3|1|5=!)r1c3

=> -56 r4c2,r12c3; btte

8x8 TM: Show

Code: Select all

 56fish  5|6r4c2
 . . . . 3r4c2 . 3r2c2
 . . . . . . . . 2r2c2 2r2c8
 . . . . . . . . 3r2c9 2r1c8 14b3p62
 . . . . 1r4c2 . . . . . . . 1r3c789 1fish
 . . . . . . . . 3r1c3 . . . . . . . 1r1c3 5r1c3
 . . . . . . . . 3r2c3 . . . . . . . 1r2c3 . . . 6r2c3
 . . . . 56r4c2  . . . . . . . . . . . . . 5fish 6fish
======================================================
+56fish -5|6r4c2

=> -56 r12c12,r4c2; btte

Code: Select all

 1fish : Finned Mutant Jellyfish (1)r348c2\c357b117  => -1 r1c3,r2c3
 5fish : Finned Mutant Jellyfish (5)r48c26\r1c3b1578 => -5 r1c3
 6fish : Finned Mutant Jellyfish (6)r48c26\r2c3b1578 => -6 r2c3
56fish : 5fish & 6fish                               => -5 r1c3, -6r2c3


[SpAce]

Hi Cenoman,

Senior exocet (1356)r12c3, r4c2, r3c789 (three cells with two locked digits as object cells)

Very nice! Once again I failed to see the desired pattern, but at least I found the corresponding logic in other ways.

Eliminations from the pattern: -28 r3c78 (not locked, non base digits in target cells) => basics to the end.

Not that it makes any difference here, but you could also conclude +1r12c3 (true base candidate because it's the only one that can go to the r3c789 target) and -1r4c2 (digit 1 must be true in the other target, thus false in this). In fact, those two conclusions alone would also yield btte. I'm not sure but I think that in some other situations one might need to use both kinds of conclusions explicitly, since they don't necessarily imply each other trivially.