The New Sudoku Players' Forum

[mith]

Code: Select all

+-------+-------+-------+
| . . . | 3 . . | 1 . . |
| . 4 . | . 1 5 | . . . |
| . . . | . . . | . . 9 |
+-------+-------+-------+
| 2 . . | 6 . . | 5 . 3 |
| . 5 . | . 8 . | . 9 . |
| 7 . 9 | . . 3 | . . 2 |
+-------+-------+-------+
| 3 . . | . . . | . . . |
| . . . | 8 4 . | . 6 . |
| . . 2 | . . 6 | . . . |
+-------+-------+-------+
...3..1...4..15...........92..6..5.3.5..8..9.7.9..3..23...........84..6...2..6...

[SpAce]

Code: Select all

.---------------------.----------------.------------------.
|  589    2     578   | 3     6   4789 | 1     4578  458  |
|  689    4     678   | 79    1   5    | 2     3     68   |
|  1568   3     15678 | 47    2   478  | 4678  4578  9    |
:---------------------+----------------+------------------:
|  2     a18   a48    | 6     79  1479 | 5     1478  3    |
|  6-4    5     3     | 2     8   147  | 467   9     146  |
|  7     a168   9     | 14    5   3    | 468   148   2    |
:---------------------+----------------+------------------:
|  3     b678   68-4  | 1579  79  179  | 489   2     1458 |
|  15     9     15    | 8     4   2    | 3     6     7    |
| b48    b78    2     | 1579  3   6    | 489   1458  1458 |
'---------------------'----------------'------------------'

Grouped W-Wing:

(48)b4p328 = (84)b7p287 => -4 r5c1,r7c3; stte

PS. It took me literally two seconds to spot that pattern. I'm not saying it to boast. I'm just wondering why, because it rarely happens even with regular W-Wings. Perhaps the neat symmetry made it more obvious. (It could be routed via box 1 also, but this is the version I saw first.)

[Ajò Dimonios]

Code: Select all

+----------------+--------------+----------------+
| 589  2   578   | 3    6  4789 | 1    4578 458  |
| 689  4   678   | 79   1  5    | 2    3    68   |
| 1568 3   15678 | 47   2  478  | 4678 4578 9    |
+----------------+--------------+----------------+
| 2    18  48    | 6    79 1479 | 5    1478 3    |
| 46   5   3     | 2    8  147  | 467  9    146  |
| 7    168 9     | 14   5  3    | 468  148  2    |
+----------------+--------------+----------------+
| 3    678 468   | 1579 79 179  | 489  2    1458 |
| 15   9   15    | 8    4  2    | 3    6    7    |
| 48   78  2     | 1579 3  6    | 489  1458 1458 |
+----------------+--------------+----------------+


4r4c3=(4-6)r7c3=6r7c2-r6c2=(6-4)r5c1=4r4c3=>+4r4c3=>stte

Paolo

[Cenoman]

Code: Select all

 +-----------------------+---------------------+-----------------------+
 |  589    2     578     |  3      6    4789   |  1      4578   458    |
 |  689    4     678     |  79     1    5      |  2      3      68     |
 |  1568   3     15678   |  47     2    478    |  4678   4578   9      |
 +-----------------------+---------------------+-----------------------+
 |  2     b18   b48      |  6      79   1479   |  5      1478   3      |
 |  6-4    5     3       |  2      8    147    |  467    9      146    |
 |  7     b168   9       |  14     5    3      |  468    148    2      |
 +-----------------------+---------------------+-----------------------+
 |  3    Aa678y  68-4    |  1579   79   179    |  489    2      1458   |
 |  15     9     15      |  8      4    2      |  3      6      7      |
 | B48z   B78    2       |  1579   3    6      |  489    1458   1458   |
 +-----------------------+---------------------+-----------------------+

Death Blossom, stem (678)r8c2
(6)r8c2 - (6=184)b4p238
(7)r8c2 - (7=84)r9c12
(8)r8c2 - (8=4)r9c1
=> -4 r5c1, r7c3; ste

[pjb]

Code: Select all

 589     2       578    | 3      6      4789   | 1      4578   458   
 689     4       678    | 79     1      5      | 2      3      68     
 1568    3       15678  | 47     2      478    | 4678   4578   9     
------------------------+----------------------+---------------------
 2       18     b48     | 6      79     1479   | 5      1478   3     
a46      5       3      | 2      8      147    | 467    9      146   
 7       18-6    9      | 14     5      3      | 468    148    2     
------------------------+----------------------+---------------------
 3      d678    c468    | 1579   79     179    | 489    2      1458   
 15      9       15     | 8      4      2      | 3      6      7     
 48      78      2      | 1579   3      6      | 489    1458   1458   


(6=4)r5c1 - (4)r4c3 = (4-6)r7c3 = (6)r7c2 => -6 r6c2; stte

Phil

[Mauriès Robert]

Hello to all.
After 6 months without doing sudoku, I come back to participate a little in the life of the forum.
Here is my resolution, with TDP of course!
P'(6r7c2) : (-6r7c2) => {6r6c2->6r7c3->4r9c1}->1r5c1->8r4c2 => 8B9 empty => r7c2=6, stte

[denis_berthier]

Code: Select all

(solve "...3..1...4..15...........92..6..5.3.5..8..9.7.9..3..23...........84..6...2..6...")
***********************************************************************************************
***  SudoRules 20.1.s based on CSP-Rules 2.1.s, config = W+SFin
***  Using CLIPS 6.32-r770
***********************************************************************************************
singles and whips[1]
finned-x-wing-in-columns: n6{c9 c1}{r5 r2} ==> r2c3 ≠ 6
biv-chain[3]: r5c1{n6 n4} - b7n4{r9c1 r7c3} - c3n6{r7 r3} ==> r2c1 ≠ 6, r3c1 ≠ 6
stte

[rjamil]

Code: Select all

 +--------------------+----------------+------------------+
 | 589   2      578   | 3     6   4789 | 1     4578  458  |
 | 689   4      678   | 79    1   5    | 2     3     68   |
 | 1568  3      15678 | 47    2   478  | 4678  4578  9    |
 +--------------------+----------------+------------------+
 | 2     (18)   48    | 6     79  1479 | 5     1478  3    |
 | (46)  5      3     | 2     8   147  | 467   9     146  |
 | 7     (168)  9     | 14    5   3    | 468   148   2    |
 +--------------------+----------------+------------------+
 | 3     67-8   468   | 1579  79  179  | 489   2     1458 |
 | 15    9      15    | 8     4   2    | 3     6     7    |
 | (48)  7-8    2     | 1579  3   6    | 489   1458  1458 |
 +--------------------+----------------+------------------+

WXYZ-Wing: 1468 @ r59c1 r46c2 => -8 @ r79c2; stte

::: OR :::

Code: Select all

 +--------------------+----------------+------------------+
 | 589   2      578   | 3     6   4789 | 1     4578  458  |
 | 689   4      678   | 79    1   5    | 2     3     68   |
 | 1568  3      15678 | 47    2   478  | 4678  4578  9    |
 +--------------------+----------------+------------------+
 | 2     18     48    | 6     79  1479 | 5     1478  3    |
 | (46)  5      3     | 2     8   147  | 467   9     146  |
 | 7     18-6   9     | 14    5   3    | 468   148   2    |
 +--------------------+----------------+------------------+
 | 3     (678)  468   | 1579  79  179  | 489   2     1458 |
 | 15    9      15    | 8     4   2    | 3     6     7    |
 | (48)  (78)   2     | 1579  3   6    | 489   1458  1458 |
 +--------------------+----------------+------------------+

WXYZ-Wing: 4678 @ r5c1 r7c2 r9c12 => -6 @ r6c2; stte

::: OR :::

Code: Select all

 +----------------------+----------------+------------------+
 | 5(8)9   2    57(8)   | 3     6   4789 | 1     4578  458  |
 | 6(8)9   4    67(8)   | 79    1   5    | 2     3     68   |
 | 156(8)  3    1567(8) | 47    2   478  | 4678  4578  9    |
 +----------------------+----------------+------------------+
 | 2       18   (48)    | 6     79  1479 | 5     1478  3    |
 | 6-4     5    3       | 2     8   147  | 467   9     146  |
 | 7       186  9       | 14    5   3    | 468   148   2    |
 +----------------------+----------------+------------------+
 | 3       678  68-4    | 1579  79  179  | 489   2     1458 |
 | 15      9    15      | 8     4   2    | 3     6     7    |
 | (48)    78   2       | 1579  3   6    | 489   1458  1458 |
 +----------------------+----------------+------------------+

Grouped W-Wing: 48 @ r4c3 r9c1 SL 8 @ r123c13 => -4 @ r5c1 r7c3; stte

R. Jamil

[SteveG48]

Code: Select all

 *--------------------------------------------------------------------*
 | 589    2      578    | 3      6      4789   | 1      4578   458    |
 | 689    4      678    | 79     1      5      | 2      3      68     |
 | 1568   3      15678  | 47     2      478    | 4678   4578   9      |
 *----------------------+----------------------+----------------------|
 | 2      18     48     | 6      79     1479   | 5      1478   3      |
 |a46     5      3      | 2      8      147    | 467    9      146    |
 | 7      18-6   9      | 14     5      3      | 468    148    2      |
 *----------------------+----------------------+----------------------|
 | 3     b678    468    | 1579   79     179    | 489    2      1458   |
 | 15     9      15     | 8      4      2      | 3      6      7      |
 |a48    b78     2      | 1579   3      6      | 489    1458   1458   |
 *--------------------------------------------------------------------*


(6=48)r59c1 - (8=76)r79c2 => -6 r6c2 ; stte

[ghfick]

A grouped W-Wing would be more like the below :

attachment=0]grouped w-wing.png[/attachment]

I am unsure I would call the previous example a grouped W-Wing. Maybe it is more like a Death Blossom?

[Leren]

Code: Select all

*--------------------------------------------------*
| 589   2    578   | 3    6  4789 | 1    4578 458  |
| 689   4    678   | 79   1  5    | 2    3    68   |
| 1568  3    15678 | 47   2  478  | 4678 4578 9    |
|------------------+--------------+----------------|
| 2    b18  a48    | 6    79 1479 | 5    1478 3    |
| 6-4   5    3     | 2    8  147  | 467  9    146  |
| 7    b168  9     | 14   5  3    | 468  148  2    |
|------------------+--------------+----------------|
| 3    c678  68-4  | 1579 79 179  | 489  2    1458 |
| 15    9    15    | 8    4  2    | 3    6    7    |
|d48   c78   2     | 1579 3  6    | 489  1458 1458 |
*--------------------------------------------------*

Another way to express the grouped W Wing : (4=8) r4c3 - r46c2 = r79c2 - (8=4) r9c9 => - 4 r5c1, r7c3; stte

More importantly the "mysterious" name of this puzzle may have been found. Is it a co-incidence that the 23rd letter of the English alphabet is... W ?

Leren

[mith]

More importantly the "mysterious" name of this puzzle may have been found. Is it a co-incidence that the 23rd letter of the English alphabet is... W ?

[SpAce]

A grouped W-Wing would be more like the below

More than what? Your example is the same as rjamil's and what I originally mentioned as an alternative to mine:

It could be routed via box 1 also

Leren's is the same as mine but written differently.

I am unsure I would call the previous example a grouped W-Wing. Maybe it is more like a Death Blossom?

Which W-Wing do you think is wrong? Exactly two Grouped W-Wings have been presented, and both are valid.

[ghfick]

Of course both steps are valid. I was attempting to address the namings.
YZF_Sudoku calls the one with the box 1 connection a 'Grouped Almost Locked Set W-Wing' and it calls the one with the column 2 connection a 'Death Blossom'.
HoDoKu calls it the one with the column 2 connection an 'Almost Locked Set Chain'.

[SpAce]

Of course both steps are valid. I was attempting to address the namings.

Which is fine, but you got it wrong.

YZF_Sudoku calls the one with the box 1 connection a 'Grouped Almost Locked Set W-Wing' and it calls the one with the column 2 connection a 'Death Blossom'.

I don't have YZF_Sudoku so I can't know what it uses, but I seriously doubt that yzfwsf would make elementary mistakes like that. I haven't seen him make mistakes at all. Most likely you're looking at totally different moves using different candidates and logic, even if they're using the same cells. (That said, I don't see any Grouped ALS W-Wing in those cells.)

Isn't the image you posted taken from YZF_Sudoku? Did it not call it Grouped W-Wing? I assume it did because you presented it as such. If it did, then it got it right, as I would expect. Are you claiming that it calls the exact same pattern "Grouped ALS W-Wing" too? I seriously doubt it, because it's not. If you disagree, please post the image of YZF's move. [Added. That's no longer necessary. It appears that my assumption was wrong, and YZF did in fact use a wrong name.]

As for the column 2 variant, look at Cenoman's solution above. That is indeed a Death Blossom (as it was correctly called) which uses the same cells as my Grouped W-Wing. The difference is that it uses the candidates 6r67c2 which are not part of the W-Wing logic at all. It could also be expressed as an ALS-XZ: (481=6)b4p328 - (6=784)b7p281 => -4 r5c1,r7c3. The W-Wing can't be.

HoDoKu calls it the one with the column 2 connection an 'Almost Locked Set Chain'.

Hodoku doesn't even have a concept of Grouped W-Wing, so of course you can't find that move there with that name. You most certainly can't find it as an "ALS Chain" because it's not. Again you're looking at a totally different move without realizing it. What you should find instead is this:

Code: Select all

.---------------------.----------------.------------------.
|  589    2     578   | 3     6   4789 | 1     4578  458  |
|  689    4     678   | 79    1   5    | 2     3     68   |
|  1568   3     15678 | 47    2   478  | 4678  4578  9    |
:---------------------+----------------+------------------:
|  2     *18   *48    | 6     79  1479 | 5     1478  3    |
| -46     5     3     | 2     8   147  | 467   9     146  |
|  7     *168   9     | 14    5   3    | 468   148   2    |
:---------------------+----------------+------------------:
|  3     *678  -468   | 1579  79  179  | 489   2     1458 |
|  15     9     15    | 8     4   2    | 3     6     7    |
| *48    *78    2     | 1579  3   6    | 489   1458  1458 |
'---------------------'----------------'------------------'
Grouped AIC: 4 4- r4c3 -8- r46c2 =8= r79c2 -8- r9c1 -4 => r5c1,r7c3<>4

That is equivalent to my (and Leren's) Grouped W-Wing. If that chain is written in Eureka, it's exactly what Leren wrote.

For some reason I can't find the box 1 version under any name in Hodoku, but if it were there, it would be another Grouped AIC:

Code: Select all

(4=8)r9c1 - r123c1 = r123c3 - (8=4)r4c3 => -4 r5c1,r7c3

(Hodoku would of course write it in its horrible Nice Loop notation.)

The same written in my abbreviated style:

Code: Select all

(48)r9123c1 = (84)r1234c3 => -4 r5c1,r7c3

I suggest you study some fundamentals before claiming our namings are wrong. You can't rely on what some software tells you if you don't understand what it's actually telling you.

--
Edit. Added a correction based on new information.