The New Sudoku Players' Forum

[Yogi]

....421362....1...63...7.......1.3....9..4..8.4..3...7.......2.712......4...7.6..

Wanna try this with Pen & Paper?

P&P.png (20.04 KiB) Viewed 763 times

[Leren]

Well, maybe not pen and paper but how about a fishing rod with some shark bait.

Code: Select all

*----------------------------------------------*
|*589f *5789 578  | 89  4    2    | 1    3   6 |
| 2     8-9  4    | 3   6    1    |*89f  7   5 |
| 6     3    1    | 589 589  7    | 4   *89  2 |
|-----------------+---------------+------------|
| 58    2    67   | 67  1    58   | 3    4   9 |
| 3     67   9    | 67  2    4    | 5    1   8 |
| 1     4    58   |*589 3   *589  | 2    6   7 |
|-----------------+---------------+------------|
| 589   5689 3568 | 1   589  3589 | 7    2   4 |
| 7     1    2    | 4   589  6    | 89   589 3 |
| 4    *589  358  | 2   7   *3589 | 6   *589 1 |
*----------------------------------------------*

Double Finned Franken Jellyfish in 9's r169b3 c2468 with fin Cells r1c1 & r2c7 => - 9 r2c2; stte

PS all the single candidate stte opportunities 9 r1c4, 9 r2c2, 8 r2c7, 9 r3c8, 9 r6c6, 9 r7c1 & 9 r8c7 were all 9's, except r2c7, which is a bi-value cell with a 9.

Does this suggest that templating on 9 may work ? Yes it does. Single digit templating on 9 => r2c7 = 9; stte

Code: Select all

*------------------------------------------*
| 589 5789 578  | 89  4   2    | 1   3   6 |
| 2   89   4    | 3   6   1    |*9-8 7   5 |
| 6   3    1    | 589 589 7    | 4   89  2 |
|---------------+--------------+-----------|
| 58  2    67   | 67  1   58   | 3   4   9 |
| 3   67   9    | 67  2   4    | 5   1   8 |
| 1   4    58   | 589 3   589  | 2   6   7 |
|---------------+--------------+-----------|
| 589 5689 3568 | 1   589 3589 | 7   2   4 |
| 7   1    2    | 4   589 6    | 89  589 3 |
| 4   589  358  | 2   7   3589 | 6   589 1 |
*------------------------------------------*

Leren

[SpAce]

Pen(cil) and paper (almost) as requested.

Step 1: Replace candidates: {589} -> {abc}; Assign: r3c5=a, r7c5=b, r8c5=c ->

Code: Select all

.-------------------.-----------------.---------------.
| abc  7abc   7abc  | abc   4    2    | 1    3     6  |
| 2    abc    4     | 3     6    1    | abc  7    (5) |
| 6    3      1     | abc  (a)   7    | 4    abc   2  |
:-------------------+-----------------+---------------:
| abc  2      67    | 67    1    abc  | 3    4    (9) |
| 3    67    (9)    | 67    2    4    | 5    1    (8) |
| 1    4      abc   | abc   3    abc  | 2    6     7  |
:-------------------+-----------------+---------------:
| abc  6abc   36abc | 1    (b)   3abc | 7    2     4  |
| 7    1      2     | 4    (c)   6    | abc  abc   3  |
| 4    abc    3abc  | 2     7    3abc | 6    abc   1  |
'-------------------'-----------------'---------------'

Step 2: Apply basics ->

Code: Select all

.--------------------.-----------------.----------------.
|  ac   7ac    7ac   | *b    4    2    |  1    3     6  |
|  2    abc    4     |  3    6    1    |  ac   7    (5) |
|  6    3      1     | *c   (a)   7    |  4   *b     2  |
:--------------------+-----------------+----------------:
| *b    2      67    |  67   1   *c    |  3    4    (9) |
|  3    67    (9)    |  67   2    4    |  5    1    (8) |
|  1    4     *c     | *a    3   *b    |  2    6     7  |
:--------------------+-----------------+----------------:
|  ac   6ac    36ac  |  1   (b)   3a   |  7    2     4  |
|  7    1      2     |  4   (c)   6    | *b   *a     3  |
|  4    ab     3ab   |  2    7    3a   |  6   *c     1  |
'--------------------'-----------------'----------------'

Step 3: Read results -> b|c <> 9 & b <> 5 => a=9, b=8, c=5; stte

[Ngisa]

Code: Select all

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


Play this puzzle online at the Daily Sudoku site

[Ngisa]

Ignore the above submission, it was too late to delete it.

Code: Select all

+------------------------+-----------------------+------------------+
|h589      5789     578  | 8-9      4       2    | 1      3       6 |
| 2        89       4    | 3        6       1    | 89     7       5 |
| 6        3        1    |a589     a5*8*9   7    | 4     b89      2 |
+------------------------+-----------------------+------------------+
| 58       2        67   | 67       1       58   | 3      4       9 |
| 3        67       9    | 67       2       4    | 5      1       8 |
| 1        4        58   | 589      3       589  | 2      6       7 |
+------------------------+-----------------------+------------------+
|g589      5689     3568 | 1       f589     3589 | 7      2       4 |
| 7        1        2    | 4       e589     6    |d89    c589     3 |
| 4        589      358  | 2        7       3589 | 6     c589     1 |
+------------------------+-----------------------+------------------+


(9=5*8*)r3c45 - (8=9)r3c8 - r89c8 = (9)r8c7 - (9=58)r8c5 - (5*8*|58=9)r7c5 - r7c1 = (9)r1c1 => - 9r1c4; stte

Clement

[Cenoman]

Code: Select all

 +----------------------+---------------------+------------------+
 |  589   5789   578    |  8-9   4     2      |  1    3     6    |
 |  2     89*    4      |  3     6     1      |  89*  7     5    |
 |  6     3      1      |  589*  589#  7      |  4    89*   2    |
 +----------------------+---------------------+------------------+
 |  58    2      67     |  67    1     58     |  3    4     9    |
 |  3     67     9      |  67    2     4      |  5    1     8    |
 |  1     4      58     |  589*  3     589*   |  2    6     7    |
 +----------------------+---------------------+------------------+
 |  589   5689   3568   |  1     589   3589   |  7    2     4    |
 |  7     1      2      |  4     589   6      |  89   589   3    |
 |  4     589*   358    |  2     7     3589*  |  6    589*  1    |
 +----------------------+---------------------+------------------+

Another fish (*) with one fin r3c5 (#)
Another single-digit 9 logic:

Code: Select all

9r3c45 9r3c8
9r6c4        9r6c6
       9r2c7       9r2c2
       9r9c8 9r9c6 9r9c2
=> -9 r1c4; ste

[SpAce]

Another fish (*) with one fin r3c5 (#)

I like your simple net and its expression, but how would you actually describe it as a fish (especially with just one fin)? I've tried but I can't. Hodoku can't find any fish with those cells either (but several others with slightly different cells that would produce the same elimination -- but they all have two fins). Now, I'm pretty sure there's a way to do it with your cells (since the net works, and everything should be explainable with set logic), but it's probably a pretty complex fish with maybe endo-fins and cannibalism, or something like that. Or maybe I'm just missing something. What was the fish you saw?

Code: Select all

9r3c45 9r3c8
9r6c4        9r6c6
       9r2c7       9r2c2
       9r9c8 9r9c6 9r9c2
=> -9 r1c4; ste

This is nice. The matrix format is a very clear way to notate nets. Unfortunately I usually get lost when I try to do that with anything but trivial cases, but I've seen you do it with quite complex ones too. Since you're the expert, could you give us a lesson sometime? Here I'm wondering why you chose that row order instead of the typical triangle format (where everything flows from top to bottom):

Code: Select all

9r3c45 9r3c8
       9r2c7 9r2c2
       9r9c8 9r9c2 9r9c6
9r6c4              9r6c6

[Cenoman]

... how would you actually describe it as a fish (especially with just one fin)?

I am not at all a keen angler. I just hoped that a single-digit logic written as a matrix 4*4 had a 4-fish name.
In the doubt, I chose the matrix format.
It can be seen as a finned swordfish [(9)r369\c468 + fin r3c5] with an extra remote fin r9c2. Very poor interest to me.
Provisionally I cancel the name. If you, careful reader of the UFG, tell me that there is no 4-fish name for this logic, I trust you.

[Leren]

Here is another fish example :

Code: Select all

*---------------------------------------------------*
|*589f1 5789 578  | 89   4   2    | 1       3     6 |
| 2     89   4    | 3    6   1    | 89f12t  7     5 |
| 6     3    1    | 589 *589 7    | 4      *89    2 |
|-----------------+---------------+-----------------|
| 58    2    67   | 67   1   58   | 3       4     9 |
| 3     67   9    | 67   2   4    | 5       1     8 |
| 1     4    58   | 589  3   589  | 2       6     7 |
|-----------------+---------------+-----------------|
|*589   5689 3568 | 1   *589 3589 | 7       2     4 |
| 7     1    2    | 4   *589 6    | 8-9    *589   3 |
| 4     589  358  | 2    7   3589 | 6      *589f2 1 |
*---------------------------------------------------*

Double Finned Swordfish in 9's c158 r378 with fin Cells r1c1 & r9c8 & transported fin Cell r2c7 => - 9 r8c7; stte

Leren

[SpAce]

Funny. I meant to quote my own reply to Cenoman's post but I guess I clicked the edit-button and and ended up losing everything I wrote before. Let's try again.

I am not at all a keen angler. I just hoped that a single-digit logic written as a matrix 4*4 had a 4-fish name.

It does. With the exact same cells it can be expressed at least as a UFG Kraken fish, an Obi-Wahn fish, or an Alien Fish. With slightly different cells it can be expressed as a UFG (doubly) finned Franken Fish.

It can be seen as a finned swordfish [(9)r369\c468 + fin r3c5] with an extra remote fin r9c2.

Yes, that is probably the simplest Kraken fish for this elimination. I never meant that it couldn't be expressed as a Kraken fish. Your exact cells work for that too, of course:

Kraken Finned Sashimi Jellyfish 9:r2369\c2468 f:r3c5 rf:r2c7-r3c8=r3c45 => -9 r1c4

I just thought you had something else in mind. Personally I rather see Kraken fishes as almost-(finned)-fishes than fishes in their own right, because they require chains.

Provisionally I cancel the name. If you, careful reader of the UFG, tell me that there is no 4-fish name for this logic, I trust you.

Let's not give up yet! UFG is a bit limited in its expressive power, so I'm not sure if that can offer a direct translation with your exact cells without using chains. Obi-Wahn's ideas might help, however:

Franken Squirmbag (Rank 2): 9:r23369\c2468b223 => -9 r1c4

As far as I understand, that's a direct translation with your exact cells. Of course it doesn't look very nice with those duplicated sectors, but Obi-Wahn offers a remedy for that too with his fish arithmetic:

Code: Select all

r23369\c2468b223 -> add r1 to both sides ->

r123369\r1c2468b223 -> replace r123 with b123 ->

r369b123\r1c2468b223 -> remove b23 from both sides ->

r369b1\r1c2468b2

Thus with that simplification, Obi-Wahn's method ends up with this:

Mutant Jellyfish (Rank 2): 9:r369b1\r1c2468b2 => -9 r1c4

...which in UFG is this:

Finned Franken Jellyfish 9:r369b1\c2468 f:r1c1 f:r3c5

...or this:

Finned Franken Jellyfish 9:r369b1\c268b2 f:r1c1 f:r6c4

So, I guess we can conclude that your pattern is indeed a standard 4-fish (but with two fins instead of one).

Lastly, there's also the Alien Fish view, i.e. my adaptation of Allan Barker's generic set logic. That can also provide a direct translation of your net, but it's a bit complicated:

Alien 4-Fish (Mixed Rank 2/1): 9:r2369\c2468b3/2 => -9 r1c4

My reasoning is that r3c8 is a link triplet (one base: r3, two links: c8,b3) lowering the rank in its base branch (r3) leading to b2. So, b2 has Rank 1 despite the global Rank 2, and therefore only two overlapping links (c4,b2) are required to kill 9r1c4. Could some XSudo user confirm or deny that logic?

[SpAce]

Double Finned Swordfish in 9's c158 r378 with fin Cells r1c1 & r9c8 & transported fin Cell r2c7 => - 9 r8c7; stte

Sure. Rewritten in slightly adapted UFG notation (less verbose, more information):

Kraken Finned Swordfish 9:c158\r378 f:r9c8 rf:r1c1-r2c2=r2c7 => -9 r8c7; stte

However, why not simply:

Finned Mutant Jellyfish 9:r2c158\r378b1 f:r2c7 f:r9c8 => -9 r8c7

Or even more simply with (slightly adapted) Obi-Wahn notation:

Mutant Jellyfish (Rank 2) 9:r2c158\r378c7b19 => -9 r8c7

Not hard to imagine which one I prefer.

[Yogi]

I notice Ngisa loaded this puzzle into the Daily Sudoku solver. Were you able to load it as a code or did you have to enter each digit manually?
I haven't been able to get anyone in that forum to tell me how to load it as a code.

Yogi

[Wecoc]

I notice Ngisa loaded this puzzle into the Daily Sudoku solver. Were you able to load it as a code or did you have to enter each digit manually?
I haven't been able to get anyone in that forum to tell me how to load it as a code.

Yogi

I made a little Javascript you can use anywhere to generate a valid link to the puzzle so you don't have to type each digit.

Code: Select all

function getDailySudokuLink(str){
  var result = 'http://www.dailysudoku.com/sudoku/play.shtml?p=';
  for (i in str){
    var n = str[i];
    if (n == '.') n = '0';
    n = n.concat(':');
    result = result.concat(n);
  }
  return result;
};

Example:

Code: Select all

getDailySudokuLink('....421362....1...63...7.......1.3....9..4..8.4..3...7.......2.712......4...7.6..')

http://www.dailysudoku.com/sudoku/play.shtml?p=0:0:0:0:4:2:1:3:6:2:0:0:0:0:1:0:0:0:6:3:0:0:0:7:0:0:0:0:0:0:0:1:0:3:0:0:0:0:9:0:0:4:0:0:8:0:4:0:0:3:0:0:0:7:0:0:0:0:0:0:0:2:0:7:1:2:0:0:0:0:0:0:4:0:0:0:7:0:6:0:0: