The New Sudoku Players' Forum

[gsf]

ronk's GFF (the Z method in my solver) labels these finned

Code: Select all

fish 7 3 1/-  1X  finned franken sword       c16b6\r2345 ==> r2c3<>7
fish 7 3 1/-  1X  finned mutant sword        r6c16\r2b145 ==> r2c3<>7


from the output using these options

Code: Select all

-v2 -q-XYKO+Z

As those fish are both sashimi (as my current GFF reports), the code I supplied to gsf obviously has a bug. I will correct the bug ASAP.

there may not be a bug in GFF
the fish are slightly different

Code: Select all

daj   c16b6\r245   r6c16\r2b45
GFF   c16b6\r2345  r6c16\r2b145


do those differences explain the sashiminess?

[daj95376]

To address my half of the confusion. My post was inappropriate because I was mixing a fish pattern being applied to a candidate grid as Sashimi ... and an exemplar being Sashimi. (ronk's reply brought that to my attention.) On top of the above mixup, I made the mistake of including the fin cell and elimination cell in the non-Sashimi exemplar. I've corrected that! Finally, because the definition of Sashimi seems to keep changing, I've decided to drop the shape qualifier and simply use N-fish with fin cell(s) from here on.

Note: because equivalent fish patterns may exist for the same '/' and '*' pattern, I no longer include 'X'.

[ronk]

Finally, because the definition of Sashimi seems to keep changing, I've decided to drop the shape qualifier and simply use N-fish with fin cell(s) from here on.

I assume you are referring to the "difference" between ...

take the fin(s) out:
1. if you have a viable basic fish of the original size -----------> Finned Fish
2. if you have a non-viable basic fish of the original size -------> Sashimi Fish

... and ...

If, after all the fin cells of a fish are removed,

either 1) all the candidates of one base unit (aka sector, house) are entirely within one cover unit,

or 2) all the candidates of one cover unit are entirely within one base unit,

then the finned fish is sashimi.

I really don't see a practical difference between the two definitions. The second is unquestionably easier to implement in a programmed solver. Additionally, there is a certain appeal in having a fish classification dependent only upon properties of the fish itself.

I suppose there is the possibility of some borderline cases. For example, a sashimi jellyfish might degenerate into two x-wings ... or a sashimi starfish might degenerate into an x-wing and a swordfish. For these cases, the second definition might fail while the first would not. If you see such cases, be sure to post them.

[daj95376]

I really don't see a practical difference between the two definitions. The second is unquestionably easier to implement in a programmed solver. Additionally, there is a certain appeal in having a fish classification dependent only upon properties of the fish itself.

I suppose there is the possibility of some borderline cases. For example, a sashimi jellyfish might degenerate into two x-wings ... or a sashimi starfish might degenerate into an x-wing and a swordfish. For these cases, the second definition might fail while the first would not. If you see such cases, be sure to post them.

I have 38 examples that degeneration to smaller fish as their first step. I doubt if I should post all of these 'borderline cases'.

Code: Select all

22   X-Wing
 7   Swordfish
 9   Jellyfish


[ronk]

I have 38 examples that degeneration to smaller fish as their first step. I doubt if I should post all of these 'borderline cases'.

Code: Select all

22   X-Wing
 7   Swordfish
 9   Jellyfish


Not all 38, of course, but pencilmarks for one or two would be nice.

Also, approximately how many [edit: degenerate] fish were screened to find those 38 examples?

[edit: seen the term "degenerative" too often it seems]

[daj95376]

Not all 38, of course, but pencilmarks for one or two would be nice.

Also, approximately how many degenerative fish were screened to find those 38 examples?

Here is the breakdown on my HPc14n table.

Code: Select all

  N   description
---   ---------------------------------------------------------------------
 61   ronk Exemplars Oct 24, 2007 (44 non-degenerate)
 36   non-degenerate
473   degenerate: Hidden Single, Locked Candidate (1), Locked Candidate (2)
 38   degenerate: X-Wing, Swordfish, Jellyfish
---
608

38/(38+473) = 7.4% fish among degenerate


Note: Original PMs were skewed towards Franken and mutant fish because unfinned/finned/Sashimi basic fish were eliminated prior to each PM being scanned. Also, Colors and Multiple Colors were applied before each PM, so this reduced other possible fish as well.

[ronk]

38/(38+473) = 7.4% fish among degenerative

Thanks. I was thinking of the percentage in terms of [edit: the for now too tough to get] appearance in random puzzles, but that's still a much higher than I would have guessed.

However, I'm still more interested in seeing the PMs and original puzzles for at least 2 of those 38.

[daj95376]

sdaj_01_27a aq 5: first of two similar fish patterns that degenerate through a smaller fish

Code: Select all

5-Fish r17c6b49\r489c24
+-----------------------------------+
|  /  A  /  |  A  /  /  |  /  /  /  |
|  .  *  .  |  *  .  /  |  .  .  .  |
|  .  *  .  |  *  .  /  |  .  .  .  |
|-----------+-----------+-----------|
|  .  *  .  |  *  *  .  |  *  *  *  |
|  /  .  /  |  *  .  /  |  .  .  .  |
|  /  .  /  |  *  .  /  |  .  .  .  |
|-----------+-----------+-----------|
|  /  A  /  |  A  /  /  |  /  /  /  |
|  *  *  *  |  *  *  .  |  .  .  .  |
|  *  *  *  |  *  *  .  |  .  .  .  |
+-----------------------------------+
r17     -  A     X-Wing
    b4  -  A     Locked Candidate (1)
    b5  -  A     Locked Candidate (1)
r1c4    =  A     Hidden Single
r7c2    =  A     Hidden Single


sdaj_01_27a as 5: second of two similar fish patterns that degenerate through a smaller fish

Code: Select all

5-Fish r167c6b9\r89c24b5
+-----------------------------------+
|  /  A  /  |  A  /  /  |  /  /  /  |
|  .  *  .  |  *  .  /  |  .  .  .  |
|  .  *  .  |  *  .  /  |  .  .  .  |
|-----------+-----------+-----------|
|  .  *  .  |  *  *  .  |  .  .  .  |
|  .  *  .  |  *  *  .  |  .  .  .  |
|  /  .  /  |  *  .  /  |  /  /  /  |
|-----------+-----------+-----------|
|  /  A  /  |  A  /  /  |  /  /  /  |
|  *  *  *  |  *  *  .  |  .  .  .  |
|  *  *  *  |  *  *  .  |  .  .  .  |
+-----------------------------------+
r17     -  A     X-Wing
r6c5    =  A     Hidden Single
r1c4    =  A     Hidden Single
r7c2    =  A     Hidden Single


[daj95376]

sdaj_01_41a ae 4: another fish pattern that degenerates through a smaller fish

Code: Select all

4-Fish c268b4\r149c3
+-----------------------------------+
|  *  A  *  |  *  *  A  |  *  A  *  |
|  .  /  *  |  .  .  /  |  .  /  .  |
|  .  /  *  |  .  .  /  |  .  /  .  |
|-----------+-----------+-----------|
|  .  /  *  |  *  *  A  |  *  A  *  |
|  /  /  .  |  .  .  /  |  .  /  .  |
|  /  /  .  |  .  .  /  |  .  /  .  |
|-----------+-----------+-----------|
|  .  /  *  |  .  .  /  |  .  /  .  |
|  .  /  *  |  .  .  /  |  .  /  .  |
|  *  A  *  |  *  *  A  |  *  A  *  |
+-----------------------------------+
  c268  -  A     233 Swordfish
    b4  -  A     Locked Candidate (1)


[ronk]

sdaj_01_41a ae 4:

Code: Select all

4-Fish c268b4\r149c3
+-----------------------------------+
|  *  A  *  |  *  *  A  |  *  A  *  |
|  .  /  *  |  .  .  /  |  .  /  .  |
|  .  /  *  |  .  .  /  |  .  /  .  |
|-----------+-----------+-----------|
|  .  /  *  |  *  *  A  |  *  A  *  |
|  /  /  .  |  .  .  /  |  .  /  .  |
|  /  /  .  |  .  .  /  |  .  /  .  |
|-----------+-----------+-----------|
|  .  /  *  |  .  .  /  |  .  /  .  |
|  .  /  *  |  .  .  /  |  .  /  .  |
|  *  A  *  |  *  *  A  |  *  A  *  |
+-----------------------------------+
  c268  -  A     233 Swordfish
    b4  -  A     Locked Candidate (1)


At that stage of the puzzle, your sashimi is apparently a mutant with one elimination.

Code: Select all

 *  X  * |  *  *  X |  *  X  *
 .  /  * |  .  .  / |  .  /  .
 .  /  * |  .  .  / |  .  /  .
---------+----------+----------
 X  / *X |  *  *  X |  *  X  *
 /  /  X |  .  .  / |  .  /  .
 /  /  X |  .  .  / |  .  /  .
---------+----------+----------
 .  # ** |  .  .  # |  .  #  .
 .  /  * |  .  .  / |  .  /  .
 *  X  * |  *  *  X |  *  X  *

sashimi mutant jellyfish c268b4\r149c3 + fin r7c268   ==> r7c3<>X

A better choice is a sashimi franken with three eliminations.

Code: Select all

 *  X ** |  *  *  X |  *  X  *
 .  /  . |  .  .  / |  .  /  .
 .  /  . |  .  .  / |  .  /  .
---------+----------+----------
 X  /  X |  *  *  X |  *  X  *
 /  /  # |  .  .  / |  .  /  .
 /  /  # |  .  .  / |  .  /  .
---------+----------+----------
 *  X ** |  *  *  X |  *  X  *
 .  /  . |  .  .  / |  .  /  .
 *  X ** |  *  *  X |  *  X  *

sashimi franken jellyfish c268b4\r1479 + fin r56c3    ==> r179c3<>X

With the latter, when the fin r56c3 is removed, all the candidates in b4 are covered by r4.

When the fin cells are removed, if an N-fish degenerates to an (N-1)-fish, I tend to think there is an implied 1-fish somewhere. No sound basis for that at this point, just a gut feel.

[edit: r6c3 in c268b4\r1479 was shown as an empty cell]

[daj95376]

A fish is or is not sashimi based on the result of a fish pattern applied to an actual candidate grid. In this case, although the exemplar is non-sashimi, the actual fish is sashimi.

I knew I shouldn't have posted the puzzles and PMs. I'm talking about exemplars being Sashimi or not.

[ronk]

I knew I shouldn't have posted the puzzles and PMs. I'm talking about exemplars being Sashimi or not.

There is only one N\(N+1) fish -- with four base sectors (units) and five cover sectors -- for your hidden pattern (after the omitted candidates are added back in).

Code: Select all

 *  X ** |  *  *  X |  *  X  *
 .  /  * |  .  .  / |  .  /  .
 .  /  * |  .  .  / |  .  /  .
---------+----------+----------
 X  / *X |  *  *  X |  *  X  *
 /  /  X |  .  .  / |  .  /  .
 /  /  X |  .  .  / |  .  /  .
---------+----------+----------
 *  X ** |  *  *  X |  *  X  *
 .  /  * |  .  .  / |  .  /  .
 *  X ** |  *  *  X |  *  X  *  # adapted from sdaj_01_41a ae 4

 c268b4\r1479c3  ==> r179c3<>X

Since each of the five cover sectors may be considered to be the fin sector, there are five possible finned N\N interpretations for this one N\(N+1) fish. These five interpretations are:

c268b4\r479c3+r1 (fin r1c268) ==> r1c3<>X
c268b4\r179c3+r4 (fin r4c168) ==> no elims
c268b4\r149c3+r7 (fin r7c268) ==> r7c3<>X
c268b4\r147c3+r9 (fin r9c268) ==> r9c3<>X
c268b4\r1479+c3 (fin r56c3) ==> r179c3<>X

IMO all finned N\N fish are not created equal, and a generalized fish finder (GFF) should not throw everything it catches into the creel. It's quite obvious that the last best fits the illustration.

After obtaining a best fit is the most appropriate time to determine if a finned N\N fish is sashimi or not.

[daj95376]

For starters, my exemplar is unfinned and has [r7c268]='/'. Almost none of what you said above applies.

You are still trying to solve the PM and argue the best choice of fish. I have no argument with your choice. However, that doesn't preclude all possible exemplars for the PM being reviewed independently for degeneration. I found the degeneration of the unfinned exemplar c268b4\r149c3 to be interesting because it entails a Swordfish.

So far, I've listed three of 38 unfinned exemplars that I know to degenerate through smaller fish. That's what I'm trying to demonstrate and discuss as Sashimi or not.

[ronk]

However, that doesn't preclude all possible exemplars for the PM being reviewed independently for degeneration. I found the degeneration of the unfinned exemplar c268b4\r149c3 to be interesting because it entails a Swordfish.

So far, I've listed three of 38 unfinned exemplars that I know to degenerate through smaller fish. That's what I'm trying to demonstrate and discuss as Sashimi or not.

According to one online dictionary, an exemplar is "an ideal that serves as a pattern." I don't see how "all possible" something-or-others for PMs can be "ideals" ... except by accident.

[daj95376]

So, what's the significance of whether or not an unfinned fish pattern (UFP) degenerates on its own.

Code: Select all

Assertion #1: If a non-degenerate UFP is applied to a candidate grid, then the results will need to be checked for unfinned, finned, or Sashimi.

Assertion #2: If a degenerate UFP is applied to a candidate grid, then the results will always be Sashimi.

Assertion #3: There are lots of UFPs that degenerate.


Code: Select all

Observation #1: Most UFPs degenerate through an initial Hidden Single, Locked Candidate (1), or Locked Candidate (2).

Observation #2: Some UFPs degenerate through an initial smaller fish.


Code: Select all

An Example: ronk's 'Sashimi Franken Jellyfish c268b4\r1479' (above). Since the fish pattern c268b4\r1479 degenerates through an initial Locked Candidate (1), it will always produce Sashimi results when applied to a candidate grid.