The NoFish List

Advanced methods and approaches for solving Sudoku puzzles

Re: The NoFish List

Postby ronk » Mon Sep 02, 2013 12:22 pm

here blue wrote:
daj95376 wrote:
For NoFish4, at the point that [edit: daj95376] show[s].
Here's one of the Obi-Wahn fish for the r9c8 elimination.
r347c15b77\r18999c3489b4

It's like the 5x5 fish, r347c15\r1c489b4, with r9c1 as a normal fin, and r3c3 and r8c5 as remote fins.
[ (r3c3-r8c3=r9c123-) and (r8c5-r8c3=r9c123-) ]

There are a few of things about that "Obi-Wahn" fish for NoFish4 that piques my interest.

1. Does anyone have a link to where Obi-Wahn posted fish (original fish, not mathematical transformations) with base sector repetitions?
2. In your r347c15b77\r18999c3489b4, would you please detail the reason for the second b7? With one less b7, couldn't there be one less r9 then too :?:

P.S. NoFish4 can be solved with a series of steps requiring no more than 3 strong links each, a.k.a. truths. It's ironic that it has garnered such a high degree of interest. :(
ronk
2012 Supporter
 
Posts: 4764
Joined: 02 November 2005
Location: Southeastern USA

Re: The NoFish List

Postby daj95376 » Mon Sep 02, 2013 4:40 pm

ronk wrote:
blue wrote:For NoFish4, at the point that [edit: daj95376] show[s].
Here's one of the Obi-Wahn fish for the r9c8 elimination.
r347c15b77\r18999c3489b4

It's like the 5x5 fish, r347c15\r1c489b4, with r9c1 as a normal fin, and r3c3 and r8c5 as remote fins.
[ (r3c3-r8c3=r9c123-) and (r8c5-r8c3=r9c123-) ]

There are a few things about that "Obi-Wahn" fish for NoFish4 that piques my interest.

1. Does anyone have a link to where Obi-Wahn posted fish (original fish, not mathematical transformations) with base sector repetitions?

I don't know of any Obi-Wahn examples for base sector repetition. However, he only posted two examples using his arithmetic ... not an exhaustive set of examples. Bottom Line: his arithmetic does not disallow it.

ronk wrote:2. In your r347c15b77\r18999c3489b4, would you please detail the reason for the second b7? With one less b7, couldn't there be one less r9 then too :?:

Your conjecture is not supported by Obi-Wahn's arithmetic because you'd still have K = 3, but one less cover count in r9c8.

Code: Select all
 blue's results
 r347c15b77\r18999c3489b4   7x10   K = 3   =>   r9c8<>9
 +-----------------------------------------------+
 |  1/1  .   .   |   .  1/1  .   |   .   .   .   |
 |   .  0/0 0/1  |   .   .   .   |  0/0  .   .   |
 |   .   .  1/1  |  1/1  .   .   |   .  1/1 1/1  |
 |---------------+---------------+---------------|
 |   .  1/1  .   |   .   .   .   |   .  1/1 1/1  |
 |  1/1  .  0/2  |   .   .   .   |  0/0  .   .   |
 |   .   .   .   |   .   .  0/0  |   .   .   .   |
 |---------------+---------------+---------------|
 |   .   .   .   |  1/1  .   .   |   .  1/1  .   |
 |   .   .  2/2  |   .  1/1  .   |   .  0/2  .   |
 |  3/3 2/3 2/4  |   .   .   .   |  0/3 0/4  .   |
 +-----------------------------------------------+

 w/o one each of [b7]\[r9]
 r347c15b7\r1899c3489b4     6x9    K = 3   =>   nada
 +-----------------------------------------------+
 |  1/1  .   .   |   .  1/1  .   |   .   .   .   |
 |   .  0/0 0/1  |   .   .   .   |  0/0  .   .   |
 |   .   .  1/1  |  1/1  .   .   |   .  1/1 1/1  |
 |---------------+---------------+---------------|
 |   .  1/1  .   |   .   .   .   |   .  1/1 1/1  |
 |  1/1  .  0/2  |   .   .   .   |  0/0  .   .   |
 |   .   .   .   |   .   .  0/0  |   .   .   .   |
 |---------------+---------------+---------------|
 |   .   .   .   |  1/1  .   .   |   .  1/1  .   |
 |   .   .  1/2  |   .  1/1  .   |   .  0/2  .   |
 |  2/2 1/2 1/3  |   .   .   .   |  0/2 0/3  .   |
 +-----------------------------------------------+


ronk wrote:P.S. NoFish4 can be solved with a series of steps requiring no more than 3 strong links each, a.k.a. truths. It's ironic that it has garnered such a high degree of interest. :(

Please provide the full logic for these steps ... without using XSUDO diagrams or anything mentioning 3-linkset.
daj95376
2014 Supporter
 
Posts: 2624
Joined: 15 May 2006

Re: The NoFish List

Postby daj95376 » Mon Sep 02, 2013 6:23 pm

Soap Box Time:

To me, the NoFish list is a misnomer because I don't recall anyone saying they had tested the puzzles for Kraken Fish.

Does anyone even have a solver that finds Kraken Fish :?:
daj95376
2014 Supporter
 
Posts: 2624
Joined: 15 May 2006

Re: The NoFish List

Postby Leren » Mon Sep 02, 2013 11:14 pm

ronk wrote:P.S. NoFish4 can be solved with a series of steps requiring no more than 3 strong links each, a.k.a. truths.

Please provide the full logic for these steps ... without using XSUDO diagrams or anything mentioning 3-linkset.

I know it's not my question but I couldn't resist finding out whether ronk was right. As far as I can see he is correct.

I have with a solution to the NoFish4 puzzle that requires (in addition to basics at various stages) : 1 Kite, 1 M Ring Type C, 5 more Kites, 1 W Wing, 1 Skyscraper, 1 L2 Wing and 1 M Ring Type A

Leren
Leren
 
Posts: 2930
Joined: 03 June 2012

Re: The NoFish List

Postby StrmCkr » Tue Sep 03, 2013 2:03 am

know it's not my question but I couldn't resist finding out whether ronk was right. As far as I can see he is correct.
yes I/we know the puzzle solves with successive simple techniques. the real pondering here is how and if we can identify the fish accomplished by the template deletion that is currently unknown: hence the no-fish.
Some do, some teach, the rest look it up.
User avatar
StrmCkr
 
Posts: 663
Joined: 05 September 2006

Re: The NoFish List

Postby ronk » Tue Sep 03, 2013 2:56 am

daj95376 wrote:
ronk wrote:
blue wrote:For NoFish4, at the point that [edit: daj95376] show[s].
Here's one of the Obi-Wahn fish for the r9c8 elimination.
r347c15b77\r18999c3489b4

It's like the 5x5 fish, r347c15\r1c489b4, with r9c1 as a normal fin, and r3c3 and r8c5 as remote fins.
[ (r3c3-r8c3=r9c123-) and (r8c5-r8c3=r9c123-) ]

There are a few things about that "Obi-Wahn" fish for NoFish4 that piques my interest.

1. Does anyone have a link to where Obi-Wahn posted fish (original fish, not mathematical transformations) with base sector repetitions?

I don't know of any Obi-Wahn examples for base sector repetition. However, he only posted two examples using his arithmetic ... not an exhaustive set of examples. Bottom Line: his arithmetic does not disallow it.

You and others are hung-up on this [edit: non-traditional] mathematical "Obi-Wahn fish" which is not a real fish IMO. Have you ever seen a fish with two or three dorsal fins [edit: in the same location]? Have you ever seen a sudoku grid with more than one r9? I think not.

The real fish in this case is r347c15b7\r189c3489b4. No duplicates ... and Xsudo likes it just fine. I like it a lot better too. Starting with this real fish, can't someone develop modified mathematical rules?

daj95376 wrote:
ronk wrote:P.S. NoFish4 can be solved with a series of steps requiring no more than 3 strong links each, a.k.a. truths. It's ironic that it has garnered such a high degree of interest. :(
Please provide the full logic for these steps ... without using XSUDO diagrams or anything mentioning 3-linkset.

Considering Leren's post, do you still want a reply from me?

[edit: Add "in the same location" for the literal minded. Make it clear this is for a "non-traditional" fish.]
Last edited by ronk on Tue Sep 03, 2013 9:47 am, edited 1 time in total.
ronk
2012 Supporter
 
Posts: 4764
Joined: 02 November 2005
Location: Southeastern USA

Re: The NoFish List

Postby StrmCkr » Tue Sep 03, 2013 4:50 am

Have you ever seen a fish with two or three dorsal fins?
YES!
http://en.wikipedia.org/wiki/Fish_anatomy
Dorsal fins are located on the back. Most fishes have one dorsal fin, but some fishes have two or three

for example HADDOCK HAS 3.

The real fish in this case is r347c15b7\r189c3489b4

pretty hard to find that fish in nxn fish mathematics as well in fact it cannot.

xsudoku dose something else to reduce its rank class: what exactly i don't know.
can't someone develop modified mathematical rules

trying as my nxn+k fish finder cannot use multiple same base/cover sectors. so far no rule i can devise allows for these type eliminations.
Last edited by StrmCkr on Tue Sep 03, 2013 5:53 am, edited 4 times in total.
Some do, some teach, the rest look it up.
User avatar
StrmCkr
 
Posts: 663
Joined: 05 September 2006

Re: The NoFish List

Postby daj95376 » Tue Sep 03, 2013 5:08 am

ronk wrote:You and others are hung-up on this mathematical "Obi-Wahn fish" which is not a real fish IMO. Have you ever seen a fish with two or three dorsal fins? Have you ever seen a sudoku grid with more than one r9? I think not.

You are absolutely right about my being hung-up on Obi-Wahn's arithmetic ... because it is precise down to the cell level.

ronk wrote:The real fish in this case is r347c15b7\r189c3489b4. No duplicates ... and Xsudo likes it just fine. I like it a lot better too. Starting with this real fish, can't someone develop modified mathematical rules?

If someone is going to develop modified mathematical rules for you, then they can start with these base/cover cell-counts from the "fish" that you and Xsudo like so much.

Code: Select all
 r347c15b7\r189c3489b4     6x8     K = 2
 +-----------------------------------------------+
 |  1/1  .   .   |   .  1/1  .   |   .   .   .   |
 |   .  0/0 0/1  |   .   .   .   |  0/0  .   .   |
 |   .   .  1/1  |  1/1  .   .   |   .  1/1 1/1  |
 |---------------+---------------+---------------|
 |   .  1/1  .   |   .   .   .   |   .  1/1 1/1  |
 |  1/1  .  0/2  |   .   .   .   |  0/0  .   .   |
 |   .   .   .   |   .   .  0/0  |   .   .   .   |
 |---------------+---------------+---------------|
 |   .   .   .   |  1/1  .   .   |   .  1/1  .   |
 |   .   .  1/2  |   .  1/1  .   |   .  0/2  .   |
 |  2/1 1/1 1/2  |   .   .   .   |  0/1 0/2  .   |
 +-----------------------------------------------+

I'll even make it easier for you. Simple Sudoku says r9c1<>9 in the solution. So, let's imagine that it's been eliminated by an earlier step.

Code: Select all
 r347c15b7\r189c3489b4     6x8     K = 2     w/o fin cell r9c1
 +-----------------------------------------------+
 |  1/1  .   .   |   .  1/1  .   |   .   .   .   |
 |   .  0/0 0/1  |   .   .   .   |  0/0  .   .   |
 |   .   .  1/1  |  1/1  .   .   |   .  1/1 1/1  |
 |---------------+---------------+---------------|
 |   .  1/1  .   |   .   .   .   |   .  1/1 1/1  |
 |  1/1  .  0/2  |   .   .   .   |  0/0  .   .   |
 |   .   .   .   |   .   .  0/0  |   .   .   .   |
 |---------------+---------------+---------------|
 |   .   .   .   |  1/1  .   .   |   .  1/1  .   |
 |   .   .  1/2  |   .  1/1  .   |   .  0/2  .   |
 |   .  1/1 1/2  |   .   .   .   |  0/1 0/2  .   |
 +-----------------------------------------------+

You now have 0/2 in r5c3, r8c8, and r9c8 with no fin cells present. According to Simple Sudoku, you're okay on claiming r8c8<>9 and r9c8<>9, but you (and Xsudo) are going to need a great fish explanation for why r5c3=9 occurs in the solution.
daj95376
2014 Supporter
 
Posts: 2624
Joined: 15 May 2006

Re: The NoFish List

Postby ronk » Tue Sep 03, 2013 10:31 am

StrmCkr wrote:
Have you ever seen a fish with two or three dorsal fins?
YES!
http://en.wikipedia.org/wiki/Fish_anatomy
Dorsal fins are located on the back. Most fishes have one dorsal fin, but some fishes have two or three
for example HADDOCK HAS 3.

Didn't know that, but had you noted and quoted my follow-on sentence it should have been clear I was talking about two (or three) dorsal fins in the same location.

StrmCkr wrote:
The real fish in this case is r347c15b7\r189c3489b4
pretty hard to find that fish in nxn fish mathematics as well in fact it cannot.
...
can't someone develop modified mathematical rules
trying as my nxn+k fish finder cannot use multiple same base/cover sectors. so far no rule i can devise allows for these type eliminations.[/

I acknowledged that point, and that's why I spoke of modified rules for the "Obi-Wahn fish", at least for when the fish is non-traditional.

StrmCkr wrote:xsudoku dose something else to reduce its rank class: what exactly i don't know.

I assume you mean "effective rank" and I don't know either.
ronk
2012 Supporter
 
Posts: 4764
Joined: 02 November 2005
Location: Southeastern USA

Re: The NoFish List

Postby ronk » Tue Sep 03, 2013 10:40 am

daj95376 wrote:I'll even make it easier for you. Simple Sudoku says r9c1<>9 in the solution. So, let's imagine that it's been eliminated by an earlier step.

Code: Select all
 r347c15b7\r189c3489b4     6x8     K = 2     w/o fin cell r9c1
 +-----------------------------------------------+
 |  1/1  .   .   |   .  1/1  .   |   .   .   .   |
 |   .  0/0 0/1  |   .   .   .   |  0/0  .   .   |
 |   .   .  1/1  |  1/1  .   .   |   .  1/1 1/1  |
 |---------------+---------------+---------------|
 |   .  1/1  .   |   .   .   .   |   .  1/1 1/1  |
 |  1/1  .  0/2  |   .   .   .   |  0/0  .   .   |
 |   .   .   .   |   .   .  0/0  |   .   .   .   |
 |---------------+---------------+---------------|
 |   .   .   .   |  1/1  .   .   |   .  1/1  .   |
 |   .   .  1/2  |   .  1/1  .   |   .  0/2  .   |
 |   .  1/1 1/2  |   .   .   .   |  0/1 0/2  .   |
 +-----------------------------------------------+

You now have 0/2 in r5c3, r8c8, and r9c8 with no fin cells present. According to Simple Sudoku, you're okay on claiming r8c8<>9 and r9c8<>9, but you (and Xsudo) are going to need a great fish explanation for why r5c3=9 occurs in the solution.

Looks like all the rules are unchanged to this point, especially as to how K is determined and applied. I think it's going to take a bit more "thinking outside the box" than that, so let's just agree to disagree.
ronk
2012 Supporter
 
Posts: 4764
Joined: 02 November 2005
Location: Southeastern USA

Re: The NoFish List

Postby daj95376 » Thu Sep 05, 2013 7:29 am

sultan vinegar wrote:Hi everyone, this is my first post after rapidly getting up to speed on advanced techniques. For this elimination:

Code: Select all
 .  .  . |  .  .  8 |  .  .  .
 8  8  . |  .  .  . |  .  8  .
 .  .  8 |  .  .  . |  .  .  8
---------+----------+---------
 .  .  8 |  .  8  . |  .  8  .
 8  .  . |  8  .  . |  .  8  .
 .  8  8 |  .  8  . |  8  .  .
---------+----------+---------
 F  .  8 |  .  8  . |  8  8  8
 .  G  . |  .  8  . |  .  8  T
 8  G  . |  8  .  . |  8  .  .


I see two overlapping fish (both for digit 8), at least one of which is true, and both fish eliminate the target. Bear with me.

F is weakly linked to G, so at least one of F,G is false.

If F is false, then the 5 truth, 7 link fish c1247b3\r25689c9b9 is true. The target T is in 0 base sets and 3 link sets, so is eliminated automatically under rank 2 logic.

If G is false, then the 3 truth, 4 link fish (grouped x-chain) c27b3\r26c9b9 is true. The target T is in 0 base sets and 2 link sets, so is eliminated automatically under rank 1 logic.

Either way, the target T is eliminated.

Interesting. So, I decided to try NxN Fish. Turns out, there are numerous combinations using your cells. Here is one combination:

Code: Select all
  scenario: r7c1 assumed empty

  4-Fish r347b3\c3589   w/fin cells r3c9,r7c7  <> 8  r8c9
  +-----------------------------------+
  |  .  .  *  |  .  *  .  |  /  X  X  |
  |  .  .  *  |  .  *  .  |  /  X  X  |
  |  /  /  X  |  /  X  /  |  /  /  #  |
  |-----------+-----------+-----------|
  |  /  /  X  |  /  X  /  |  /  X  X  |
  |  .  .  *  |  .  *  .  |  .  *  *  |
  |  .  .  *  |  .  *  .  |  .  *  *  |
  |-----------+-----------+-----------|
  |  @  /  X  |  /  X  /  |  #  X  X  |
  |  .  .  *  |  .  *  .  |  .  * **  |
  |  .  .  *  |  .  *  .  |  .  *  *  |
  +-----------------------------------+

Code: Select all
  scenario: r89c2 assumed empty

  3-Fish c27b3\r26b9   w/fin cell  r3c9        <> 8  r78c9
  +-----------------------------------+
  |  .  /  .  |  .  .  .  |  /  /  /  |
  |  *  X  *  |  *  *  *  |  /  X  X  |
  |  .  /  .  |  .  .  .  |  /  /  #  |
  |-----------+-----------+-----------|
  |  .  /  .  |  .  .  .  |  /  .  .  |
  |  .  /  .  |  .  .  .  |  /  .  .  |
  |  *  X  *  |  *  *  *  |  X  *  *  |
  |-----------+-----------+-----------|
  |  .  /  .  |  .  .  .  |  X  * **  |
  |  .  @  .  |  .  .  .  |  X  * **  |
  |  .  @  .  |  .  .  .  |  X  *  *  |
  +-----------------------------------+

  essentially equivalent to your second fish

  In addition, there are 4-Fish that perform r8c9<>8 and only need r9c2 to be assumed empty.

Too bad I couldn't find a combination where both fish patterns had the same base set and/or cover set.
daj95376
2014 Supporter
 
Posts: 2624
Joined: 15 May 2006

re: "NoFish #4"

Postby Pat » Mon Sep 09, 2013 1:53 pm

blue wrote:
    to exclude 9 at r9c8:
      7\10
      (9) r347c15b77\r18999c3489b4


or a "finned" 7-fish
    (9) r347c15b77\r18c3489b4
    with fin = r9c123

i call it a "finned" fish,
ronk rejects it,
perhaps he's right,
i don't own the term fish,
so:
    it's a "finned" eel
and we can stop debating terminology
and get back to the substance

the substance:
the arithmetic works just fine
User avatar
Pat
 
Posts: 3448
Joined: 18 July 2005

Postby Pat » Tue Sep 10, 2013 12:26 pm

daj95376 wrote:
I don't recall anyone saying they had tested the puzzles for Kraken

Does anyone even have a solver that finds Kraken ?


i don't even have a good definition of Kraken

somehow i suspect that any Kraken will also have an equivalent using Obi-Wahn's arithmetic

    blue has become very quiet,
    possibly working on your request
    ( and coming up empty-handed if my conjecture is true )
User avatar
Pat
 
Posts: 3448
Joined: 18 July 2005

Postby Pat » Tue Sep 10, 2013 12:47 pm

StrmCkr wrote:
my nxn+k fish finder cannot use multiple same base/cover sectors


i'm not quite sure how to read your comment

is this a programming issue ?

yes the puzzle only has 27 houses
but
the eels are very talented, they can re-use houses as necessary

so, just clone any house you wish to re-use in the Cover

    e.g.
    • r9
    • r9'
    • r9''
    are identical triplets
    and can be used as "separate" houses in the Cover

likewise for the Base

    e.g.
    • b7
    • b7'
    are identical twins
    and can be used as "separate" houses in the Base
User avatar
Pat
 
Posts: 3448
Joined: 18 July 2005

Postby daj95376 » Tue Sep 10, 2013 3:56 pm

Pat wrote:
daj95376 wrote:
I don't recall anyone saying they had tested the puzzles for Kraken

Does anyone even have a solver that finds Kraken ?


i don't even have a good definition of Kraken

somehow i suspect that any Kraken will also have an equivalent using Obi-Wahn's arithmetic

    blue has become very quiet,
    possibly working on your request
    ( and coming up empty-handed if my conjecture is true )

StrmCkr sent me a pm and reminded me that HoDoKu has a Kraken fish option. I haven't tried it yet.

I must admit that TUFG's Kraken fish definition (in the head post) seems nebulous. Until recently, I confused ronk's remote fin cell definition as being associated with Kraken fish only. However, I now believe that remote fin cells can be present in any fish shape. Remote fin cells are a subset of what I've been calling indirect fin cells. Indirect fin cells do not directily see all EE cells, but include all Kraken fin cells and remote fin cells.

Obi-Wahn's arithmetic would accurately identify the existence of indirect fin cells in a fish pattern, but it wouldn't show how difficult it would be to link them back to the fish eliminations ... and that's the most important part of identifying such cells.
daj95376
2014 Supporter
 
Posts: 2624
Joined: 15 May 2006

PreviousNext

Return to Advanced solving techniques