The New Sudoku Players' Forum

by **Ocean** » Tue Nov 07, 2006 8:56 pm

Finlip wrote:
Ocean wrote:First diagram: r1c7<>8. Second diagram: r4c9<>8.
Choose any of the two eliminations of an '8', and the puzzle is solved.

I was able to see that both your elimination are perfectly all right but fail to understand the terms.

I hoped somebody else should answer, that knows the terms better than me. Here is ronk's explanation of this pattern (finned bbbb\rrcc).

by **daj95376** » Tue Nov 07, 2006 9:08 pm

[Edited] Removed because it was only for fun, but is now a distraction.

by **ronk** » Tue Nov 07, 2006 10:29 pm

daj95376 wrote:# the elusive Franken 1-Wing [r7] in <8>
# (@) cells forced to <8> by fin cell -- one in every box

daj, in a franken fish an exclusion cell must directly see (all) the fin cells(s) AFAIK.

Is it a Kraken "1-fish?" Maybe ... as Mike Barker definitely uses chains there. But I don't think even Mike's exotic fish have used multiple inferences ... as your example does several times.

Keep on fishing.

by **daj95376** » Wed Nov 08, 2006 1:48 am

[Edited] Removed. Original post deleted.

by **Ruud** » Mon Nov 13, 2006 10:32 pm

Tarek,

this mutant Jellyfish also seems to be pecking at its own tentacles...

Code: Select all: . / . | * X * | X X * . / . | * X * | X X * . / . | * X * | X X * ----------+----------+--------- . / . | . / . | / / . * X * | * *X * | X X * . / . | . / . | / / . ----------+----------+--------- . / . | . / . | / / . * X * | * *X * | X X * . / . | . / . | / / . Fig 4H: cccc\rrbb rrrr\ccbb transpose

Ruud

by **tarek** » Tue Nov 14, 2006 12:46 am

Ruud wrote:Tarek,
this mutant Jellyfish also seems to be pecking at its own tentacles...

Thanx & you're right Ruud......
The Xs in either r5c5 or r8c5 if true would force a hidden single in c2 with a combined effect of leaving an X in b3 to cover 2 columns ..... an impossibility.....

may be something like this is more accurate....as the presence of the Xs or not would NOT affect the rest of the eliminations....

Code: Select all: . / . | * X * | X X * . / . | * X * | X X * . / . | * X * | X X * ----------+----------+--------- . / . | . / . | / / . * X * | * . * | X X * . / . | . / . | / / . ----------+----------+--------- . / . | . / . | / / . * X * | * . * | X X * . / . | . / . | / / . Fig 4H: cccc\rrbb Franken Jellyfish

& I agree that the diagrams should have the fish names -as agreed- on them.... as this is an "(a Cloumn + b Box) * (c Row + d Box)" where a+b=c+d=N && b+d>0 then this is a Franken fish

tarek

by **Ocean** » Tue Nov 14, 2006 1:05 am

Not sure if there should be examples of basic fish in this thread. But here are some appearances of the Swordfish anyway. Every fish comes with a dual. The Swordfish may have dual Swordfish, Jellyfish, Squirmbag or 6-fish. And they may come single or pairwise. Examples shown here: Single SF/SF, Single SF/JF, Single SJ/SQ, Pairs of SF/JF, Pairs of SF/SQ.

000010002003000040050002600000003100100040007004800000005900030090000800200070000 # Single SF/SF
000010002003000040040005600000006300100070008008300000009200050070000900200080000 # Single SF/JF
001000200030000040500060007000102000040030070000809000600000004070050030002000900 # Single SF/SQ
000010002003000010040005600000007800100020009009500000007300050080000300200090000 # Pairwise SF/JF
001000200030000040500030006000107000080040050000902000100000008040060030002000100 # Pairwise SF/SQ

Code: Select all: # Single Swordfish w/ dual Swordfish. 000010002003000040050002600000003100100040007004800000005900030090000800200070000 *-----------------------------------------------------------* | 678-9 4 *689 | 56 1 *79 | 3 57-9 2 | | 67-9 2 3 | 56 8 *79 |*59 4 1 | |A79 5 1 | 4 3 2 | 6 A79 8 | |-------------------+-------------------+-------------------| |A5689 68 2 | 7 A59 3 | 1 A689 4 | | 1 3 *89 | 2 4 6 |*59 58-9 7 | |A569 7 4 | 8 A59 1 | 2 A69 3 | |-------------------+-------------------+-------------------| | 4 1 5 | 9 2 8 | 7 3 6 | | 3 9 7 | 1 6 4 | 8 2 5 | | 2 68 68 | 3 7 5 | 4 1 9 | *-----------------------------------------------------------* # Swordfish pattern * (or alternatively the dual pattern A).

Code: Select all: # Single Swordfish w/ dual Jellyfish 000010002003000040040005600000006300100070008008300000009200050070000900200080000 *-----------------------------------------------------------------------------* | 5679 8 567 | 4679 1 3479 | 57 379 2 | | 567-9 12 3 | 67-9 *269 8 | 157 4 *1579 | | 79 4 12 | 79 23 5 | 6 8 13 | |-------------------------+-------------------------+-------------------------| | 457 *259 257 | 8 *459 6 | 3 17-9 *14579 | | 1 3 56 | 459 7 2 | 45 69 8 | | 4567 *569 8 | 3 *459 1 | 2 67-9 *45679 | |-------------------------+-------------------------+-------------------------| | 3 16 9 | 2 46 47 | 8 5 1467 | | 8 7 1456 | 1456 3456 34 | 9 2 1346 | | 2 156 1456 | 145679 8 3479 | 147 1367 13467 | *-----------------------------------------------------------------------------* # Swordfish pattern * (dual Jellyfish not shown).

Code: Select all: # Single Swordfish w/ dual Squirmbag 001000200030000040500060007000102000040030070000809000600000004070050030002000900 ???? *-----------------------------------------------------------* | 4 6 1 | 3 *89 7 | 2 *89 5 | | 89 3 7 | 259 12-89 158 | 6 4 189 | | 5 2 89 | 4 6 18 | 3 1-89 7 | |-------------------+-------------------+-------------------| | 789 5-89 36 | 1 47 2 | 45 6-89 389 | | 1289 4 89 | 56 3 56 | 18 7 1289 | | 127 15 36 | 8 47 9 | 45 126 123 | |-------------------+-------------------+-------------------| | 6 *189 5 | 29 *1289 3 | 7 *128 4 | | 189 7 4 | 269 5 168 | 18 3 128 | | 3 *18 2 | 7 *18 4 | 9 5 6 | *-----------------------------------------------------------* # Swordfish pattern * (dual Squirmbag not shown).

Code: Select all: # Pairwise Swordfish w/ dual Jellyfish 000010002003000010040005600000007800100020009009500000007300050080000300200090000 *-----------------------------------------------------------* | 5 9 68 | 468 1 3468 | 7 38 2 | | 67-8 2 3 | 68 67-8 9 | 4 1 5 | |*78 4 1 | 2 *378 5 | 6 9 *38 | |-------------------+-------------------+-------------------| | 34 5 2 | 9 346 7 | 8 46 1 | | 1 67 48 | 468 2 3468 | 5 3467 9 | |*348 67 9 | 5 *3468 1 | 2 467 346 | |-------------------+-------------------+-------------------| | 46 1 7 | 3 *468 2 | 9 5 *468 | | 9 8 46 | 1 5 46 | 3 2 7 | | 2 3 5 | 7 9 468 | 1 468 46-8 | *-----------------------------------------------------------* # First Swordfish pattern * (dual Jellyfish not shown). *-----------------------------------------------------------* | 5 9 *68 | 46-8 1 *3468 | 7 *38 2 | | 678 2 3 | 68 678 9 | 4 1 5 | | 78 4 1 | 2 378 5 | 6 9 38 | |-------------------+-------------------+-------------------| | 34 5 2 | 9 346 7 | 8 46 1 | | 1 67 *48 | 46-8 2 *3468 | 5 3467 9 | | 348 67 9 | 5 3468 1 | 2 467 346 | |-------------------+-------------------+-------------------| | 46 1 7 | 3 468 2 | 9 5 468 | | 9 8 46 | 1 5 46 | 3 2 7 | | 2 3 5 | 7 9 *468 | 1 *468 46-8 | *-----------------------------------------------------------* # Second Swordfish pattern * (dual Jellyfish not shown).

Code: Select all: # Pairwise Swordfish w/ dual Squirmbag 001000200030000040500030006000107000080040050000902000100000008040060030002000100 *-----------------------------------------------------------* | 48 6 1 | 458 79 458 | 2 79 3 | | 789 3 -789 | 2 1 6 | 5-789 4 579 | | 5 2 *4789 |*478 3 489 |*789 1 6 | |-------------------+-------------------+-------------------| | 469 59 3459 | 1 8 7 | 36 2 49 | | 2 8 *79 | 6 4 3 |*79 5 1 | | 467 1 34-7 | 9 5 2 | 36 8 47 | |-------------------+-------------------+-------------------| | 1 579 6 | 3 2 59 | 4 79 8 | | 89 4 589 |*57 6 1 |*579 3 2 | | 3 579 2 | 48 79 48 | 1 6 579 | *-----------------------------------------------------------* # First Swordfish pattern * (dual Squirmbag not shown). *-----------------------------------------------------------* | 48 6 1 | 458 *79 458 | 2 *79 3 | | 789 3 789 | 2 1 6 | 5789 4 579 | | 5 2 4789 | 478 3 489 | 789 1 6 | |-------------------+-------------------+-------------------| | 469 59 3459 | 1 8 7 | 36 2 49 | | 2 8 79 | 6 4 3 | 79 5 1 | | 467 1 347 | 9 5 2 | 36 8 47 | |-------------------+-------------------+-------------------| | 1 *579 6 | 3 2 59 | 4 *79 8 | | 89 4 589 | 57 6 1 | 579 3 2 | | 3 *579 2 | 48 *79 48 | 1 6 5-79 | *-----------------------------------------------------------* # Second Swordfish pattern * (dual Squirmbag not shown).

#

by **tarek** » Tue Nov 14, 2006 1:21 am

Great examples Ocean........
I'm not sure how many true example per fish should be put there.... I will probably choose some representative example(s) & leave some line puzzles as excercises (benchmark list seems best for the line puzzle excercises)....

tarek

by **ronk** » Tue Nov 14, 2006 1:38 am

tarek wrote:I agree that the diagrams should have the fish names -as agreed- on them.... as this is an "(a Cloumn + b Box) * (c Row + d Box)" where a+b=c+d=N && b+d>0 then this is a Franken fish

My understanding is that the franken fish ...

neither allows a box as a base unit
nor allows for more than one box as a cover unit

Figures 4H and '4H inverse' are problematic because they are degenerative patterns. I thought about omitting them at first, but decided to leave them for completeness sake. Then when we think the set of exemplars really is complete, we can decide whether or not to discard the "useless" ones.

by **tarek** » Tue Nov 14, 2006 1:46 am

ronk wrote:My understanding is that the franken fish ...
neither allows a box as a base unit
nor allows for more than one box as a cover unit
Figures 4H and '4H inverse' are problematic because they are degenerative patterns.

During the discussions, my understanding was that boxes were NOW allowed wherever in Franken fish but mixing of columns with rows is the factor that makes the fish mutant

ronk wrote:I thought about omitting them at first, but decided to leave them for completeness sake. Then when we think the set of exemplars really is complete, we can decide whether or not to discard the "useless" ones.

Fair enough !

tarek

by **Mike Barker** » Tue Nov 14, 2006 1:52 am

Ruud, interesting catch, but it may be better described as a Franken Swordfish followed by a line/box elimination. I've sometimes found in these cannabalistic fish that there's actually a smaller fish or a Siamese fish (two fish combined into one) which is responsible for the self eliminations.

Code: Select all: . / . | . . . | X X * . / . | . . . | X X * . / . | . . . | X X * ----------+----------+--------- . / . | . . . | / / . * X * | * * * | X X * . / . | . . . | / / . ----------+----------+--------- . / . | . . . | / / . * X * | * * * | X X * . / . | . . . | / / .

Code: Select all: . . . | * X * | . . . . . . | * X * | . . . . . . | * X * | . . . ----------+----------+--------- . . . | . / . | . . . . . . | . / . | . . . . . . | . / . | . . . ----------+----------+--------- . . . | . / . | . . . . . . | . / . | . . . . . . | . / . | . . .

[Edit: ever write a post and then find the conversation has gone on without you? Anyway my vote would be to not include degenerate patterns. Ron it was your recommendation that we include row/box or column/box basis as Frankenfish ( http://forum.enjoysudoku.com/viewtopic.php?p=36992#p36992 ), but if you want to change I'll happily go along!)

by **ronk** » Tue Nov 14, 2006 2:33 am

tarek wrote:During the discussions, my understanding was that boxes were NOW allowed wherever in Franken fish but mixing of columns with rows is the factor that makes the fish mutant

Mike Barker wrote:Ron it was your recommendation that we include row/box or column/box basis as Frankenfish ( http://forum.enjoysudoku.com/viewtopic.php?p=36992#p36992 ), but if you want to change I'll happily go along!)

You're both correct ... my mistake. But I still think the franken fish has only one box ... as either a base unit or a cover unit.

Mike Barker wrote:Anyway my vote would be to not include degenerate patterns.

Ultimately that will be my vote too, but would like to keep the degenerative patterns until we're sure we've explored all the possibilities. So far the list has been made in a haphazard manner ... rather than with an exhaustive search.

by **Ruud** » Tue Nov 14, 2006 2:55 am

Many of these fish patterns allow different interpretations. For the eliminations, only the empty cells are relevant.

Here are 2 completely different views on the same 8 empty cells:

Code: Select all: * X X | . . . | X X * * X X | . . . | X X * * X X | . . . | X X * ----------+----------+--------- . / / | . . . | / / . * X X | * * * | X X * . / / | . . . | / / . ----------+----------+--------- . / / | . . . | / / . * X X | * * * | X X * . / / | . . . | / / . Franken-Jelly cccc * rrbb

Code: Select all: * . . | . . . | . . * * . . | . . . | . . * * . . | . . . | . . * ----------+----------+--------- X / / | X X X | / / X * . . | * * * | . . * X / / | X X X | / / X ----------+----------+--------- X / / | X X X | / / X * . . | * * * | . . * X / / | X X X | / / X Franken-Jelly rrrr-ccbb

Both are 2-box Franken-Fish, I think...

combined with this:

Code: Select all: . . . | * X * | . . . . . . | * X * | . . . . . . | * X * | . . . ----------+----------+--------- . . . | . / . | . . . . . . | . / . | . . . . . . | . / . | . . . ----------+----------+--------- . . . | . / . | . . . . . . | . / . | . . . . . . | . / . | . . . locked candidates 2

gives us this monster:

Code: Select all: * X X | * X * | X X * * X X | * X * | X X * * X X | * X * | X X * ----------+----------+--------- . / / | . / . | / / . * X X | * *X * | X X * . / / | . / . | / / . ----------+----------+--------- . / / | . / . | / / . * X X | * *X * | X X * . / / | . / . | / / . cannibalistic Franken-Squirmbag

Ruud

by **Mike Barker** » Tue Nov 14, 2006 5:58 am

Degenerate fish can lead to all sorts of fun patterns

. Here's one I'd either call the Fantastic Four or X-Fish:

Code: Select all: X / X | * * * | X / X / / / | . . . | / / / X / X | * * * | X / X ----------+----------+--------- * . * | . . . | * . * * . * | . . . | * . * * . * | . . . | * . * ----------+----------+--------- X / X | * * * | X / X / / / | . . . | / / / X / X | * * * | X / X bbbb/rrrr and bbbb/cccc

Which gets cannibalistic when it becomes four corners

Code: Select all: *X / X | * * * | X / X* / / / | . . . | / / / X / / | * * * | / / X ----------+----------+--------- * . * | . . . | * . * * . * | . . . | * . * * . * | . . . | * . * ----------+----------+--------- X / / | * * * | / / X / / / | . . . | / / / *X / X | * * * | X / X* bbbb/rrrr, bbbb/cccc, and bbbb/rrcc

You could have a positive fish which actually forces a placement (how do we show that?):

Code: Select all: . . . | X / X | . . . . . . | X / X | . . . . . . | X / X | . . . ----------+----------+--------- X X X | * * * | X X X / / / | * . * | / / / X X X | * * * | X X X ----------+----------+--------- . . . | X / X | . . . . . . | X / X | . . . . . . | X / X | . . . bbbb/rrcc

and a negative fish that doesn't really subtract anything:

Code: Select all: . . . | . . . | . . . . . . | . . . | . . . . . . | . . . | . . . ----------+----------+--------- X X X | X X X | X X X X X X | X X X | X X X X X X | X X X | X X X ----------+----------+--------- . . . | . . . | . . . . . . | . . . | . . . . . . | . . . | . . . bbb/rrr

Ron's desire to have both the fish and its inverse in the same set begs the question of what an identity fish looks like. Maybe

Code: Select all: * * * | * * * | * * * * * * | * * * | * * * * * * | * * * | * * * ----------+----------+--------- * * * | * * * | * * * * * * | * * * | * * * * * * | * * * | * * * ----------+----------+--------- * * * | * * * | * * * * * * | * * * | * * * * * * | * * * | * * *

Then theres the unit fish

Code: Select all: . . . | . * . | . . . . . . | . * . | . . . . . . | . * . | . . . ----------+----------+--------- . . . | * * * | . . . / / / | / X / | / / / . . . | * * * | . . . ----------+----------+--------- . . . | . * . | . . . . . . | . * . | . . . . . . | . * . | . . .

and the null fish

Code: Select all: . . . | . . . | . . . . . . | . . . | . . . . . . | . . . | . . . ----------+----------+--------- . . . | . . . | . . . . . . | . . . | . . . . . . | . . . | . . . ----------+----------+--------- . . . | . . . | . . . . . . | . . . | . . . . . . | . . . | . . .

Enough, I fear I may have degenerated too far! Happy fishing.

by **Havard** » Tue Nov 14, 2006 12:50 pm

ronk wrote:You're both correct ... my mistake. But I still think the franken fish has only one box ... as either a base unit or a cover unit.

The original Franken Fish had two boxes in the covering unit, and I thought that the new definition of a Frankenfish was one that uses one or more boxes as covering units http://forum.enjoysudoku.com/viewtopic.php?p=36826#p36826

Boxes in the base-units was concidered Mutant Fish, or?

Havard

The New Sudoku Players' Forum

The Ultimate FISH Guide

Re: The Ultimate FISH Guide