The New Sudoku Players' Forum

[ronk]

I've been thinking about the implications of smaller unfinned fish in the degeneration of an unfinned fish. ........

Since we don't know whether ultimately the (unfinned) fish is true OR at least one of the fin cells is true ... I have no idea why you would want to pursue all those details for the unfinned fish.

[Pat]

I believe that these 5 are the only ones to survive degeneration

Code: Select all

  mutant Jellyfish r49c8b1\r3c36b6
  +-----------------------------------+
  |  /  /  X  |  .  .  *  |  .  /  .  |
  |  /  /  X  |  .  .  *  |  .  /  .  |
  |  X  X  *  |  *  *  *  |  *  X  *  |
  |-----------+-----------+-----------|
  |  /  /  X  |  /  /  X  |  X  /  X  |
  |  .  .  *  |  .  .  *  |  *  X  *  |
  |  .  .  *  |  .  .  *  |  *  X  *  |
  |-----------+-----------+-----------|
  |  .  .  *  |  .  .  *  |  .  /  .  |
  |  .  .  *  |  .  .  *  |  .  /  .  |
  |  /  /  X  |  /  /  X  |  /  /  /  |
  +-----------------------------------+


Code: Select all

  mutant Whale r68c28b35\r357c369
  +-----------------------------------+
  |  .  /  *  |  .  .  *  |  /  /  X  |
  |  .  /  *  |  .  .  *  |  /  /  X  |
  |  *  X  *  |  *  *  *  |  X  /  *  |
  |-----------+-----------+-----------|
  |  .  /  *  |  /  /  X  |  .  /  *  |
  |  *  X  *  |  X  X  *  |  *  X  *  |
  |  /  /  X  |  /  /  /  |  /  /  X  |
  |-----------+-----------+-----------|
  |  *  X  *  |  *  *  *  |  *  X  *  |
  |  /  /  X  |  /  /  X  |  /  /  X  |
  |  .  /  *  |  .  .  *  |  .  /  *  |
  +-----------------------------------+


Code: Select all

  mutant Whale r68c248b3\r357c39b8
  +-----------------------------------+
  |  .  /  *  |  /  .  .  |  /  /  X  |
  |  .  /  *  |  /  .  .  |  /  /  X  |
  |  *  X  *  |  X  *  *  |  X  /  *  |
  |-----------+-----------+-----------|
  |  .  /  *  |  /  .  .  |  .  /  *  |
  |  *  X  *  |  X  *  *  |  *  X  *  |
  |  /  /  X  |  /  /  /  |  /  /  X  |
  |-----------+-----------+-----------|
  |  *  X  *  |  *  *  *  |  *  X  *  |
  |  /  /  X  |  /  X  X  |  /  /  X  |
  |  .  /  *  |  X  *  *  |  .  /  *  |
  +-----------------------------------+


Code: Select all

  mutant Starfish r367c2b3\r1c159b4
  +-----------------------------------+
  |  *  X  *  |  *  *  *  |  X  X  *  |
  |  *  /  .  |  .  *  .  |  /  /  X  |
  |  X  /  /  |  /  X  /  |  /  /  /  |
  |-----------+-----------+-----------|
  |  *  X  *  |  .  *  .  |  .  .  *  |
  |  *  X  *  |  .  *  .  |  .  .  *  |
  |  *  /  X  |  /  X  /  |  /  /  X  |
  |-----------+-----------+-----------|
  |  X  /  /  |  /  X  /  |  /  /  X  |
  |  *  /  .  |  .  *  .  |  .  .  *  |
  |  *  /  .  |  .  *  .  |  .  .  *  |
  +-----------------------------------+


Code: Select all

  Franken Jellyfish c2345\r149b5
  mutant  Jellyfish c23b28\r149c6   ('X' supplied)
  +-----------------------------------+
  |  *  X  X  |  X  X  *  |  *  *  *  |
  |  .  /  /  |  /  /  X  |  .  .  .  |
  |  .  /  /  |  /  /  X  |  .  .  .  |
  |-----------+-----------+-----------|
  |  *  X  X  |  *  *  *  |  *  *  *  |
  |  .  /  /  |  .  .  *  |  .  .  .  |
  |  .  /  /  |  .  .  *  |  .  .  .  |
  |-----------+-----------+-----------|
  |  .  /  /  |  /  /  X  |  .  .  .  |
  |  .  /  /  |  /  /  X  |  .  .  .  |
  |  *  X  X  |  X  X  *  |  *  *  *  |
  +-----------------------------------+


the first Jellyfish ( rrcb\rccb ) is new

but the last ( cccc\rrrb or ccbb\rrrc )
already appears in ronk's catalogue ( 4B )

[ronk]

Code: Select all

5-Fish r378c37\r2c14b49            ...  <> X  [r1c14],[r2c1245689],[r456c124],[r9c1489]
5-Fish r3c37b78\r29c14b4           ...  <> X  [r1c14],[r2c1245689],[r456c124],[r9c1489]

Because one cover unit (b4) intersects only one base unit (c3), it is relatively easy to see that these fish are sashimi.

If only they were all that easy.

[daj95376]

Last five grids withdrawn because of bookeeping errors. I'm sorry!!!

[daj95376]

Well, I reprocessed all of the fish patterns from scratch and hope that these four qualify as non-degenerative and new. As they say, the 99th try is the charm!

Code: Select all

  mutant Jellyfish r49c8b1\r3c36b6
  +-----------------------------------+
  |  /  /  X  |  .  .  *  |  .  /  .  |
  |  /  /  X  |  .  .  *  |  .  /  .  |
  |  X  X  *  |  *  *  *  |  *  X  *  |
  |-----------+-----------+-----------|
  |  /  /  X  |  /  /  X  |  X  /  X  |
  |  .  .  *  |  .  .  *  |  *  X  *  |
  |  .  .  *  |  .  .  *  |  *  X  *  |
  |-----------+-----------+-----------|
  |  .  .  *  |  .  .  *  |  .  /  .  |
  |  .  .  *  |  .  .  *  |  .  /  .  |
  |  /  /  X  |  /  /  X  |  /  /  /  |
  +-----------------------------------+ # sdaj_03_27a aj 4


Code: Select all

  mutant Whale r68c28b35\r357c369
  +-----------------------------------+
  |  .  /  *  |  .  .  *  |  /  /  X  |
  |  .  /  *  |  .  .  *  |  /  /  X  |
  |  *  X  *  |  *  *  *  |  X  /  *  |
  |-----------+-----------+-----------|
  |  .  /  *  |  /  /  X  |  .  /  *  |
  |  *  X  *  |  X  X  *  |  *  X  *  |
  |  /  /  X  |  /  /  /  |  /  /  X  |
  |-----------+-----------+-----------|
  |  *  X  *  |  *  *  *  |  *  X  *  |
  |  /  /  X  |  /  /  X  |  /  /  X  |
  |  .  /  *  |  .  .  *  |  .  /  *  |
  +-----------------------------------+ # sdaj_05_14b ab 6


Code: Select all

  mutant Whale r68c248b3\r357c39b8
  +-----------------------------------+
  |  .  /  *  |  /  .  .  |  /  /  X  |
  |  .  /  *  |  /  .  .  |  /  /  X  |
  |  *  X  *  |  X  *  *  |  X  /  *  |
  |-----------+-----------+-----------|
  |  .  /  *  |  /  .  .  |  .  /  *  |
  |  *  X  *  |  X  *  *  |  *  X  *  |
  |  /  /  X  |  /  /  /  |  /  /  X  |
  |-----------+-----------+-----------|
  |  *  X  *  |  *  *  *  |  *  X  *  |
  |  /  /  X  |  /  X  X  |  /  /  X  |
  |  .  /  *  |  X  *  *  |  .  /  *  |
  +-----------------------------------+ # sdaj_05_14b ad 6


Code: Select all

  mutant Starfish r367c2b3\r1c159b4
  +-----------------------------------+
  |  *  X  *  |  *  *  *  |  X  X  *  |
  |  *  /  .  |  .  *  .  |  /  /  X  |
  |  X  /  /  |  /  X  /  |  /  /  /  |
  |-----------+-----------+-----------|
  |  *  X  *  |  .  *  .  |  .  .  *  |
  |  *  X  *  |  .  *  .  |  .  .  *  |
  |  *  /  X  |  /  X  /  |  /  /  X  |
  |-----------+-----------+-----------|
  |  X  /  /  |  /  X  /  |  /  /  X  |
  |  *  /  .  |  .  *  .  |  .  .  *  |
  |  *  /  .  |  .  *  .  |  .  .  *  |
  +-----------------------------------+ # sdaj_06_98c ae 5


[ronk]

Well, I reprocessed all of the fish patterns from scratch and hope that these four qualify as non-degenerative and new. As they say, the 99th try is the charm!

Code: Select all

  mutant Jellyfish r49c8b1\r3c36b6
  +-----------------------------------+
  |  /  /  X  |  .  .  *  |  .  /  .  |
  |  /  /  X  |  .  .  *  |  .  /  .  |
  |  X  X  *  |  *  *  *  |  *  X  *  |
  |-----------+-----------+-----------|
  |  /  /  X  |  /  /  X  |  X  /  X  |
  |  .  .  *  |  .  .  *  |  *  X  *  |
  |  .  .  *  |  .  .  *  |  *  X  *  |
  |-----------+-----------+-----------|
  |  .  .  *  |  .  .  *  |  .  /  .  |
  |  .  .  *  |  .  .  *  |  .  /  .  |
  |  /  /  X  |  /  /  X  |  /  /  /  |
  +-----------------------------------+ # sdaj_03_27a aj 4


I wish I knew, I wish I knew. It's still bothersome to me, e.g., that the above contains the smaller fish ... sashimi mutant 3-fish r49c8\c36b6 + fin r3c8 for r3c3<>X. (The remora 2-fish r9b1\r3c3 + fin r9c6 for r3c6 was applied before finding the 4-fish.)

But I'm inclined to conclude r3c3<>X is a "remora" elimination too ... simply because it's in the intersection of two cover units of the 4-fish (as is r3c6).

[tarek]

Why don't we agree on a "hierarchy" of template checking... so that all of this hopefully can be more fruitful.

As we haven't yet decided on which method is more difficult than the other, maybe at the moment we should go for:

1. clearing all smaller-sized fish regardless of shape & fin
2. clear less sophisticated shapes regardless of fin
2. clear finless (if searching for pure finned)
3. clear sashimi (if searching for pure non-sashimi)

I know it is tempting to go for the finned non-sashimi mutant whale, but it seems a more exhaustive process should be in place prior to that.

tarek

[daj95376]

Code: Select all

  mutant Jellyfish r49c8b1\r3c36b6
  +-----------------------------------+
  |  /  /  X  |  .  .  *  |  .  /  .  |
  |  /  /  X  |  .  .  *  |  .  /  .  |
  |  X  X  *  |  *  *  *  |  *  X  *  |
  |-----------+-----------+-----------|
  |  /  /  X  |  /  /  X  |  X  /  X  |
  |  .  .  *  |  .  .  *  |  *  X  *  |
  |  .  .  *  |  .  .  *  |  *  X  *  |
  |-----------+-----------+-----------|
  |  .  .  *  |  .  .  *  |  .  /  .  |
  |  .  .  *  |  .  .  *  |  .  /  .  |
  |  /  /  X  |  /  /  X  |  /  /  /  |
  +-----------------------------------+ # sdaj_03_27a aj 4


I wish I knew, I wish I knew. It's still bothersome to me, e.g., that the above contains the smaller fish ... sashimi mutant 3-fish r49c8\c36b6 + fin r3c8 for r3c3<>X. (The remora 2-fish r9b1\r3c3 + fin r9c6 for r3c6 was applied before finding the 4-fish.)

But I'm inclined to conclude r3c3<>X is a "remora" elimination too ... simply because it's in the intersection of two cover units of the 4-fish (as is r3c6).

I started with a finned fish. Then I removed the fin cells so the underlying unfinned fish could be tested for degeneration. Now you're uncomfortable that a smaller finned fish exists in my unfinned fish. Boy, degeneration is getting extremely complex

I'm not sure where to go from here except to say that neither of your elimination cells are the ones eliminated in my original finned fish.

This brings to light something that's been bothering me for awhile. The unfinned fish doesn't convey enough information to justify it's existence. Only when you factor in the fin and elimination cells, does a complete picture exist. Here's my full GFF output. Do you find anything smaller to explain my fin and elimination cells

Code: Select all

4-Fish r49c8b1\r3c36b6          f  211\121  <> 3  [r5c3],[r6c3]
  +-----------------------------------+
  |  /  /  X  |  .  .  *  |  .  /  .  |
  |  /  /  X  |  .  .  *  |  .  /  .  |
  |  X  X  *  |  *  *  *  |  *  X  *  |
  |-----------+-----------+-----------|
  |  /  #  X  |  /  /  X  |  X  /  X  |
  |  .  . **  |  .  .  *  |  *  X  *  |
  |  .  . **  |  .  .  *  |  *  X  *  |
  |-----------+-----------+-----------|
  |  .  .  *  |  .  .  *  |  .  /  .  |
  |  .  .  *  |  .  .  *  |  .  /  .  |
  |  /  /  X  |  /  /  X  |  /  /  /  |
  +-----------------------------------+
//...*././/...*./...*****.*/#.//../...$..**.*..$..**.*..*..*./...*..*././/.//./// # sdaj_03_27a aj 4


[daj95376]

Why don't we agree on a "hierarchy" of template checking... so that all of this hopefully can be more fruitful.

As we haven't yet decided on which method is more difficult than the other, maybe at the moment we should go for:

1. clearing all smaller-sized fish regardless of shape & fin
2. clear less sophisticated shapes regardless of fin
2. clear finless (if searching for pure finned)
3. clear sashimi (if searching for pure non-sashimi)

I know it is tempting to go for the finned non-sashimi mutant whale, but it seems a more exhaustive process should be in place prior to that.

While I study your suggestions closer. Please consider/review my current processing method.

I have two versions of my GFF program. One solves for the smallest size fish present in a PM. This allows for a PM to be decomposed in steps based on the size of the fish present. The second version solves for all fish present in a PM. This lets me analyze an unfinned fish for all-size fish during degeneration.

Below is a sample (decomposed) output listing that I produced by jumping back-n-forth between my solver and my GFF program. Please note that my solver processes many fish before a PM is generated. I'm only presenting results from a PM to this thread. To my knowledge, there aren't any smaller fish present than the ones I list.

Code: Select all

Puzzle #sdaj_03_27:
 89...4..65679.241...4.5.......2...7.7...1...4.4...9.......2.3...184.75922..5...41

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

   c15   X-Wing                          <> 6    [r6c3478]
         Multiple Colors                 <> 6    [r4c6]
 r4      Naked  Triple                   <> 358  [r4c37]
 r19     X-Wing F/S                      <> 3    [r3c6]
   c59   X-Wing F/S                      <> 8    [r6c78]

 +-----------------------------------------------------------------------+
 |  8      9      123    |  137    37     4      |  27     5      6      |
 |  5      6      7      |  9      38     2      |  4      1      38     |
 |  13     23     4      |  13678  5      168    |  278    238    9      |
 |-----------------------+-----------------------+-----------------------|
 |  9      38     16     |  2      4      358    |  16     7      358    |
 |  7      238    2356   |  368    1      3568   |  9      2368   4      |
 |  136    4      1235   |  378    3678   9      |  12     23     358    |
 |-----------------------+-----------------------+-----------------------|
 |  4      5      9      |  168    2      168    |  3      68     7      |
 |  36     1      8      |  4      36     7      |  5      9      2      |
 |  2      7      36     |  5      9      368    |  68     4      1      |
 +-----------------------------------------------------------------------+
4-Fish r49c28\r35c6b6           B  220\211  <> 3  [r5c3]
4-Fish r49c2b3\r35c69           B  211\220  <> 3  [r5c3]
4-Fish r9c28b5\r36c36           f  121\220  <> 3  [r5c3]
4-Fish r9c28b5\r36c6b4          f  121\211  <> 3  [r5c3]
4-Fish r9c8b15\r36c36           f  112\220  <> 3  [r5c3]
4-Fish r1249\c369b2             fF 400\031  <> 3  [r5c3],[r6c3]
4-Fish r1249\c69b24             fF 400\022  <> 3  [r5c3],[r6c3]
4-Fish r49b13\r3c369            f  202\130  <> 3  [r5c3],[r6c3]
4-Fish r49b13\r3c69b4           f  202\121  <> 3  [r5c3],[r6c3]
4-Fish r49c28\r3c36b6           B  220\121  <> 3  [r5c3],[r6c3]
4-Fish r49c28\r3c6b46           A  220\112  <> 3  [r5c3],[r6c3]
4-Fish r49c2b3\r3c369           B  211\130  <> 3  [r5c3],[r6c3]
4-Fish r49c2b3\r3c69b4          A  211\121  <> 3  [r5c3],[r6c3]
4-Fish r49c8b1\r3c36b6          f  211\121  <> 3  [r5c3],[r6c3] *** aj 4
4-Fish r49c8b1\r3c6b46          f  211\112  <> 3  [r5c3],[r6c3]
4-Fish r9c28b5\r356c6           f  121\310  <> 3  [r5c3],[r6c3]
4-Fish r9c8b15\r356c6           f  112\310  <> 3  [r5c3],[r6c3]
4-Fish r49c28\r36c6b6           B  220\211  <> 3  [r6c3]
4-Fish r49c2b3\r36c69           B  211\220  <> 3  [r6c3]
4-Fish r9c28b5\r35c36           f  121\220  <> 3  [r6c3]
4-Fish r9c28b5\r35c6b4          f  121\211  <> 3  [r6c3]
4-Fish r9c8b15\r35c36           f  112\220  <> 3  [r6c3]

         Colors                          <> 3    [r1c5]
 r26     X-Wing                          <> 8    [r4c9]
         XY-Wing  [r4c6]/[r4c9]+[r6c5]   <> 3    [r6c89]
         XY-Wing  [r6c8]/[r4c9]+[r6c3]   <> 5    [r6c9]

 893174256567982413124356789931248675782615934645739128459821367318467592276593841


[ronk]

I started with a finned fish. Then I removed the fin cells so the underlying unfinned fish could be tested for degeneration. Now you're uncomfortable that a smaller finned fish exists in my unfinned fish. Boy, degeneration is getting extremely complex

Until recent conjectures, degeneration has meant "smaller fish." However, it's not clear to me if that includes, e.g., a smaller 2-finned fish when the larger is 1-finned ... or a smaller 1-finned fish when the larger is unfinned.

And "yes", determining whether a mutant fish is sashimi is complicated.

I'm not sure where to go from here except to say that neither of your elimination cells are the ones eliminated in my original finned fish.

Except for recent conjectures, that's never been a requirement ... and I believe it would be a mistake to make it a requirement.

This brings to light something that's been bothering me for awhile. The unfinned fish doesn't convey enough information to justify it's existence. Only when you factor in the fin and elimination cells, does a complete picture exist.

I have not been illustrating finned and sashimi fish without the fin cells ... except for exemplars believed to be non-sashimi and when quoting others, primarily you IIRC.

[daj95376]

ronk: If you examine the output listing above that I posted for tarek, you'll see that [r3c3]=4 is a given and [r3c3]<>3 is performed that way. The listing also shows that [r3c6]<>3 was performed prior to my 4-Fish.

At this point. I no longer give a tinker's toy about degenerate/Sashimi fish.

[tarek]

I have two versions of my GFF program. One solves for the smallest size fish present in a PM. This allows for a PM to be decomposed in steps based on the size of the fish present. The second version solves for all fish present in a PM. This lets me analyze an unfinned fish for all-size fish during degeneration.

I think the second version (The batch mode solver) should be employed.
What I'm trying to propose is like this (from your example):
1. You are looking for 4-fish (regardless of shape & fin) present in a PM
2. Batch solver finds (singles,box-line,subsets) & commits all moves
3. Batch solver solves for (1-fish then 2-fish then 3-fish):regardless of shape or fin & commits all moves as it progresses
4. Batch solver finds 4-fish (that allow elimination) without committing the moves
5. display step 4



From what you are saying this is what you have done, that also means that your fish have no smaller fish equivalent.

I haven't checked your results because I haven't come to terms with your representation of base/cover sets (I getting there)

tarek

[daj95376]

tarek, your description is essentially correct. Here's what I do in a little more detail.

All of my fish puzzles are generated so they can be solved with the following techniques. The ordering is the default hierarchy used in my solver.

Naked Single
Hidden Single

Naked Pair
Naked Triple

Locked Candidate 1
Locked Candidate 2

Naked Quad

Hidden Pair
Hidden Triple
Hidden Quad

X-Wing
Swordfish
Jellyfish

XY-Wing
XYZ-Wing

Colors and Multiple Colors

finned/Sashimi X-Wing
finned/Sashimi Swordfish
finned/Sashimi Jellyfish

(print PM)

Templates

Because I've gone through different approaches for a best way to process a PM with multiple fish present, I haven't merged my GFF program into my solver yet. This leaves the following manual processing done by me ... and why I had bookeeping problems with earlier submissions.

When my solver lists a PM, then I manually take it and run it through my GFF program looking for the smallest size fish present. In my example above, that happened to be for 4-Fish and they only occurred for value <3>. I then manually update the PM and run it back through my solver. I've had to repeat this process as many as four times before my solver was able to complete a puzzle on its own. I then merge all of the above into one listing.

Note: My solver only lists the base sectors for fish. I may update the format before long. Also, I apply the same technique as many times as possible once I enter a routine. This grouping often results in multiple entries for Locked Candidate (1), as an example. It also results in conjugate fish being listed together -- even if they eliminate the same cells.

[tarek]

if your GFF finds ALL SMALLEST FISH then I would suggest that you would integrate it in the solver (then you could remove other fish tests already implemented)

I assume that your GFF would go for 2-fish first & would display everything (basic, franken & mutant /finned & non finned) as batch search.
commit everything then proceed to the next level.
The GFF 4-fish output should essentially not change if what you claim is true.

what complicates things slightly is that you started checking for fish before colouring/multi colouring.

I would re-order your solver 1st & put all fish last in the order (assuming that you already modified your solver to catch & eliminate in BATCH mode)........ to compare results.

tarek

[ronk]

ronk: If you examine the output listing above that I posted for tarek, you'll see that [r3c3]=4 is a given and [r3c3]<>3 is performed that way. The listing also shows that [r3c6]<>3 was performed prior to my 4-Fish.

Had you posted the original puzzle, the pencilmarks, or a solution path prior to that If not, how would I have known those details

But I don't think it makes any difference ... because IMO a "generalized fish finder" should also be able to work with a hidden pattern as its only input. The only stumbling block to this capability AFAIK is the absence of an acceptable -- and implementable -- definition of sashimi.

Ironically, when this thread was started, the plan was to drop the sashimi distinction.