The New Sudoku Players' Forum

[ArkieTech]

Code: Select all

 *-----------*
 |4..|...|7..|
 |.53|...|...|
 |..7|.4.|..8|
 |---+---+---|
 |...|.5.|.9.|
 |...|..7|46.|
 |8..|..2|..5|
 |---+---+---|
 |.8.|1.5|..3|
 |...|...|6..|
 |.3.|79.|.8.|
 *-----------*


Play/Print this puzzle online

[Leren]

Code: Select all

*--------------------------------------------------------------*
| 4     1269  8      | 2356  1236  1369   | 7     1235  169    |
| 1269  5     3      | 268   7     1689   | 129   124   1469   |
| 1269  1269  7      | 2356  4     1369   | 12359 1235  8      |
|--------------------+--------------------+--------------------|
| 12367 12467 1246   | 346   5     134-6  | 8     9    b127    |
| 12359 129   1259   | 389   138   7      | 4     6    b12     |
| 8     14679 1469   | 469  c16    2      |c13   c137   5      |
|--------------------+--------------------+--------------------|
| 2679  8     2469   | 1     2-6   5      | 29    247   3      |
| 12579 12479 12459  | 2348  238   348    | 6     12457 1479   |
| 1256  3     12456  | 7     9    a46     | 125   8    a14     |
*--------------------------------------------------------------*

ALS XY Wing: (6=1) r9c69 - (1=7) r45c9 - (7=6) r6c578 => - 6 r4c6, r7c5; lclste

Leren

[SteveG48]

Code: Select all

 *--------------------------------------------------------------------*
 | 4      1269   8      | 2356   1236   1369   | 7      1235   169    |
 | 1269   5      3      | 268    7      1689   | 129    124    1469   |
 | 1269   1269   7      | 2356   4      1369   | 12359  1235   8      |
 *----------------------+----------------------+----------------------|
 | 12367  12467  1246   | 346    5      1346   | 8      9      127    |
 | 12359  129    1259   | 389    138    7      | 4      6      12     |
 | 8      14679  1469   | 469    16     2      | 13     137    5      |
 *----------------------+----------------------+----------------------|
 | 2679   8      2469   | 1      26     5      | 29     247    3      |
 | 12579  12479  12459  | 2348   238    348    | 6      12457 A1479   |
 | 1256   3      12456  | 7      9      46     | 125    8      4-1    |
 *--------------------------------------------------------------------*


I'm not sure about the terminology or notation here. I'm not going to try to mark the grid.
I think you would call this a kraken cell, r8c9:

(1)r8c9 - r9c9 => -1 r9c9
(4)r8c9 - (4=238)r8c456 - (2=6)r7c5 - (6=1)r6c5 - (1=37)r6c78 - (7=12)r45c9 => -1 r9c9
(7)r8c9 - (7=12)r45c9 => -1 r9c9
(9)r8c9 - (9=2)r7c7 - (2=6)r7c5 - (6=1)r6c5 - (1=37)r6c78 - (7=12)r45c9 => -1 r9c9

=> -1 r9c9 ; stte

[tlanglet]

Code: Select all

 *--------------------------------------------------------------------*
 | 4      1269   8      | 2356   1236   1369   | 7      1235   169    |
 | 1269   5      3      | 268    7      1689   | 129    124    1469   |
 | 1269   1269   7      | 2356   4      1369   | 12359  1235   8      |
 |----------------------+----------------------+----------------------|
 | 12367  12467  1246   | 346    5      1346   | 8      9     *127    |
 | 12359  129    1259   | 389    138    7      | 4      6     *12     |
 | 8      14679  1469   | 469    16     2      | 13     13-7   5      |
 |----------------------+----------------------+----------------------|
 | 2679   8      2469   | 1      26     5      | 29    *47=2   3      |
 | 12579  12479  12459  | 2348   238    348    | 6      12457  149-7  |
 | 1256   3      12456  | 7      9      46     | 125    8     *14     |
 *--------------------------------------------------------------------*


Almost ALS xy-wing(14-7) with pivot (14)r9c9, pincers als(127)r45c9 and (47=2)r7c8 => r6c8,r8c9<>7
als sy-wing(14-7): (7=4)r7c8-(4=1)r9c9-(1=27)r54c9 => r6c8,r8c9<>7
||
2r7c8(2=6)r7c5-(6=41)r9c69-(1=27)r54c9 => r6c8,r8c9<>7

Ted

[daj95376]

(1)r8c9 - r9c9 => -1 r9c9
(4)r8c9 - (4=238)r8c456 - (2=6)r7c5 - (6=1)r6c5 - (1=37)r6c78 - (7=12)r45c9 => -1 r9c9
(7)r8c9 - (7=12)r45c9 => -1 r9c9
(9)r8c9 - (9=2)r7c7 - (2=6)r7c5 - (6=1)r6c5 - (1=37)r6c78 - (7=12)r45c9 => -1 r9c9

=> -1 r9c9 ; stte

Steve, time to work on grouped strong links.

(1)r8c9 - r9c9 => -1 r9c9
(4)r8c9 - (4=238)r8c456 - (2=6)r7c5 - (6=1)r6c5 - (1)r6c78 = (1)r45c9 => -1 r9c9
(7)r8c9 - (7=12)r45c9 => -1 r9c9
(9)r8c9 - (9=2)r7c7 - (2=6)r7c5 - (6=1)r6c5 - (1)r6c78 = (1)r45c9 => -1 r9c9


FWIW: All of the chains found by my solver needed Basics to complete the solution.

[SteveG48]

Steve, time to work on grouped strong links.

(1)r8c9 - r9c9 => -1 r9c9
(4)r8c9 - (4=238)r8c456 - (2=6)r7c5 - (6=1)r6c5 - (1)r6c78 = (1)r45c9 => -1 r9c9
(7)r8c9 - (7=12)r45c9 => -1 r9c9
(9)r8c9 - (9=2)r7c7 - (2=6)r7c5 - (6=1)r6c5 - (1)r6c78 = (1)r45c9 => -1 r9c9

Thanks, Danny. I should have seen that. I guess I got hung up on the 1/2 pair at r45c9 in handling the 7 chain and couldn't let it go.

Is there a better way to write this sort of thing?

[daj95376]

Is there a better way to write this sort of thing?

Not that I recall. However, I often drop the "=>" and treat everything as chain segments with a common ending. (it saves repetition like in your first chain segment.) I will also use a CODE segment so that I can get details to align, but that's just my OCD showing.

Code: Select all

(1)r8c9                                                               - (1)r9c9
(7)r8c9 - (7=12 )r45c9                                                - (1)r9c9
(4)r8c9 - (4=238)r8c456 - (2=6)r7c5 - (6=1)r6c5 - (1)r6c78 = (1)r45c9 - (1)r9c9
(9)r8c9 - (9=2  )r7c7  /


[Marty R.]

Code: Select all

*--------------------------------------------------------------*
| 4     1269  8      | 2356  1236  1369   | 7     1235  169    |
| 1269  5     3      | 268   7     1689   | 129   124   1469   |
| 1269  1269  7      | 2356  4     1369   | 12359 1235  8      |
|--------------------+--------------------+--------------------|
| 12367 12467 1246   | 346   5     134-6  | 8     9    b127    |
| 12359 129   1259   | 389   138   7      | 4     6    b12     |
| 8     14679 1469   | 469  c16    2      |c13   c137   5      |
|--------------------+--------------------+--------------------|
| 2679  8     2469   | 1     2-6   5      | 29    247   3      |
| 12579 12479 12459  | 2348  238   348    | 6     12457 1479   |
| 1256  3     12456  | 7     9    a46     | 125   8    a14     |
*--------------------------------------------------------------*

ALS XY Wing: (6=1) r9c69 - (1=7) r45c9 - (7=6) r6c578 => - 6 r4c6, r7c5; lclste

Essentially the same, but my starting point was (6=1)r6c5.

Leren

[Luke]

Is there a better way to write this sort of thing?

I can't say it's better, but King Of All Networks ttt and others use vertical notation as Ted did above. Slight difference: this variation only states the conclusion once, in the header. The vertical pipes (||) of course represent the strong symbol (=) turned on its side.


kraken cell (1479)r8c9 =>r9c9<>1

(1)r8c9
||
(7)r8c9-(7=21)r45c9
||
(9)r8c9-(9=2)r7c7-(2=6)r7c5-(6=1)r6c5-(1)r6c78=(1)r45c9
||
(4)r8c9-(4=238)r8c456-(2=6)r7c5-(6=1)r6c5-(1)r6c78=(1)r45c9

[SteveG48]

There are elements that I like from both Danny's and Luke's notation. I'm not sure the that || really serves much purpose.

[daj95376]

I can't say it's better, but King Of All Networks ttt and others use vertical notation as Ted did above. Slight difference: this variation only states the conclusion once, in the header. The vertical pipes (||) of course represent the strong symbol (=) turned on its side.

Hmmm!!! In "C", a pair of vertical pipes (||) mean logical OR. In my notation, I use them to mean exclusive logical OR. In this case, it would translate into:

Code: Select all

(1)r8c9 XLOR (4)r8c9 XLOR (7)r8c9 XLOR (9)r8c9  ...  exactly one is true


Another example is where multiple choices are split off in a network:

Code: Select all

... - (2)r1c4 = (3)r1c4 ...
                    ||
              = (5)r1c4 ...
                    ||
              = (6)r1c4 ...


However, for a Kraken Cell, I tend to only use vertical pipes when it's needed in a (networked) chain segment.

Bottom Line: I'm only talking personal preference ... and accept Luke's notation as also acceptable.

[Luke]

For perspective, amusement, and a peek into how the other half lives, check out this incredible solution from the mad genius ttt.

He applies the notation I referenced to kraken cell, dual kraken, aals, aaals, and the kitchen sink.

[DonM]

There are elements that I like from both Danny's and Luke's notation. I'm not sure the that || really serves much purpose.

Unfortunately, this is another example of where because of the destruction of all of the Eureka forum and some of the old Player's forum, the impression is given that there is some question as to how the notation should or can be because the history has been almost completely erased.

The notation that Luke indicates above is the standard way of vertically notating Almost AICs (aka Krakens) (and, IMO, AALSs and BUGs) and that has been true since early 2007. I know this to be true for Almost AICs because I was the 2nd person on the Eureka forum (after Steve K) to start notating them for manual solutions starting in early 2007 (this was before ttt arrived on the scene).

At first, I tried a single-line notation and others had their own suggestions, but as time went on, it became evident that the vertical method (which was already being used for AALSs and BUGs) was an overall superior format and that became the standard method used by Steve K, ttt, myself and others. As Luke mentions, ttt took network solutions to a whole new level which included Almost-Almost AICs and a complexity that would have been almost impossible to decipher if he had used single-line network notation rather than vertical.

The vertical pipes indicating the strong inferences are not discretionary; they are mandatory, otherwise you have what are a series of unrelated statements.

Btw Steve, while these days I avoid using net-based solutions for simpler puzzles, I also learned how to find them by using them in puzzles that could be solved with simpler methods. Your AAIC/cell move using a 4-digit cell is not the easiest thing to find (ie. read clever) and if you keep it up, you will have some good tools to solve far more difficult puzzles.