View previous topic :: View next topic |
Author |
Message |
keith
Joined: 19 Sep 2005 Posts: 3355 Location: near Detroit, Michigan, USA
|
Posted: Tue Jan 13, 2009 12:40 am Post subject: |
|
|
Wapati,
I am enjoying this discussion!
However,
Do you have a real example? A swordfish (6 cells in 3 rows or columns) plus two essential extra "fin" cells that make an elimination not made by more elementary methods? Where both fins and the swordfish are required in the logic?
In my practical experience, almost all of these "almost fish" turn out to be simplified by line - box interactions. No higher theory needed.
Keith |
|
Back to top |
|
|
wapati
Joined: 10 Jun 2008 Posts: 472 Location: Brampton, Ontario, Canada.
|
Posted: Tue Jan 13, 2009 2:18 am Post subject: |
|
|
keith wrote: | Wapati,
I am enjoying this discussion!
However,
Do you have a real example? A swordfish (6 cells in 3 rows or columns) plus two essential extra "fin" cells that make an elimination not made by more elementary methods? Where both fins and the swordfish are required in the logic?
In my practical experience, almost all of these "almost fish" turn out to be simplified by line - box interactions. No higher theory needed.
Keith |
They are rare, partly because most people, and some solvers will not spot them. Further, there are always ways around big fish, so no-one needs to spot them.
I don't agree that skyscraper can be worked around easily. It cracks many puzzles formerly thought to be difficult. Skyscraper, sashimi x-wing, turbot ... all names for the same idea, and a powerful tool.
Any swordfish being needed is pretty rare.
I have no example at hand but will look. |
|
Back to top |
|
|
wapati
Joined: 10 Jun 2008 Posts: 472 Location: Brampton, Ontario, Canada.
|
Posted: Tue Jan 13, 2009 3:22 am Post subject: |
|
|
Here is a puzzle with some finned swordfish.
Code: | . . .|4 . .|. . 3
. . .|. 1 .|5 6 9
. . 6|. . .|4 2 .
-----+-----+-----
6 . .|. . .|9 3 .
. 2 .|. . 4|. . .
. . .|. 7 2|6 . .
-----+-----+-----
. 6 7|5 . 1|. . .
. 5 2|7 . .|. . .
1 4 .|. . .|. . 6
.------------------.------------------.------------------.
| 2 19 5 | 4 689 689 | 18 7 3 |
| 4 38 38 | 2 1 7 | 5 6 9 |
| 7 19 6 | 389 3589 3589 | 4 2 18 |
:------------------+------------------+------------------:
| 6 7 4 | 1 58 58 | 9 3 2 |
| 5 2 39 | 6 39 4 | 18 18 7 |
| 89 38 1 | 39 7 2 | 6 4 5 |
:------------------+------------------+------------------:
| 3 6 7 | 5 489 1 | 2 89 48 |
| 89 5 2 | 7 4689 689 | 3 189 148 |
| 1 4 89 | 389 2 389 | 7 5 6 |
'------------------'------------------'------------------' |
I've marked one of the swordfish below. As far as I'm concerned any one of the cells marked "@" may be considered as part of the pattern.
The elimination is made when all the fins are true or when the pattern is true. Code: |
.------------------.------------------.------------------.
| 2 19 5 | 4 689 689 | 18 7 3 |
| 4 38 38 | 2 1 7 | 5 6 9 |
| 7 19 6 | 389 3589 3589 | 4 2 18 |
:------------------+------------------+------------------:
| 6 7 4 | 1 58 58 | 9 3 2 |
| 5 2 39 | 6 39 4 | 18 18 7 |
|#89 38 1 |@39 7 2 | 6 4 5 |
:------------------+------------------+------------------:
| 3 6 7 | 5 @489 1 | 2 #89 48 |
|#89 5 2 | 7 @4689 @689 | 3 #189 148 |
| 1 4 89 | 38-9 2 389 | 7 5 6 |
'------------------'------------------'------------------' |
I can make the same elimination with other steps but none I'd call basic.
Edited to add: Here is another.
Code: | . . 6|8 . 7|. . 2
. . .|4 . .|. 5 3
8 . .|2 . .|. 9 .
-----+-----+-----
2 3 8|9 7 .|. . 5
. . .|6 . 5|. . .
5 . .|. 2 .|. . 7
-----+-----+-----
. . .|. . .|. . .
. 4 5|. . .|. 3 .
7 1 .|5 . 8|. . 4
.---------------.---------------.---------------.
| 3 9 6 | 8 5 7 | 14 *14 2 |
| 1 2 7 | 4 69 69 | 8 5 3 |
| 8 5 4 | 2 13 13 | 7 9 6 |
:---------------+---------------+---------------:
| 2 3 8 | 9 7 14 | 6 @14 5 |
| 4 7 19 | 6 8 5 | 3 2 @19 |
| 5 6 19 |@13 2 134 | 49-1 8 7 |
:---------------+---------------+---------------:
| 69 8 23 |*13 4 269 | 5 7 *19 |
| 69 4 5 | 7 169 269 | 129 3 8 |
| 7 1 23 | 5 39 8 | 29 6 4 |
'---------------'---------------'---------------' |
|
|
Back to top |
|
|
daj95376
Joined: 23 Aug 2008 Posts: 3854
|
Posted: Tue Jan 13, 2009 5:01 am Post subject: |
|
|
The commonality between a Skyscraper, Sashimi X-Wings, and a Siamese Sashimi X-Wing.
Set H Puzzle 3 after basics:
Code: | +--------------------------------------------------------------+
| 1368 168 7 | 4 9 126 | 5 1236 16 |
| 4 5 16 | 8 3 126 | 7 126 9 |
| 1369 169 2 | 67 15 57 | 8 136 4 |
|--------------------+--------------------+--------------------|
| 2 4 8 | 1 6 3 | 9 7 5 |
| 7 3 5 | 9 2 8 | 14 146 16 |
| 16 16 9 | 5 7 4 | 3 8 2 |
|--------------------+--------------------+--------------------|
| 15 7 3 | 2 145 9 | 6 14 8 |
| 1568 2 16 | 67 1458 57 | 14 9 3 |
| 1689 1689 4 | 3 18 16 | 2 5 7 |
+--------------------------------------------------------------+
# 54 eliminations remain
|
Code: | Skyscraper c34
+-----------------------------------+
| 6 6 . | . . 6 | . 6 6 |
| . . *6 | . . -6 | . 6 . |
| -6 -6 . | *6 . . | . 6 . |
|-----------+-----------+-----------|
| . . . | . 6 . | . . . |
| . . . | . . . | . 6 6 |
| 6 6 . | . . . | . . . |
|-----------+-----------+-----------|
| . . . | . . . | 6 . . |
| 6 . *6 | *6 . . | . . . |
| 6 6 . | . . 6 | . . . |
+-----------------------------------+
|
Code: | Sashimi X-Wing c34\r28 w/fin [r3c4] Sashimi X-Wing c34\r38 w/fin [r2c3]
+-----------------------------------+ +-----------------------------------+
| 6 6 . | . . 6 | . 6 6 | | 6 6 . | . . 6 | . 6 6 |
| . . *6 | . . -6 | . 6 . | | . . 6# | . . 6 | . 6 . |
| 6 6 . | 6# . . | . 6 . | | -6 -6 . | *6 . . | . 6 . |
|-----------+-----------+-----------| |-----------+-----------+-----------|
| . . . | . 6 . | . . . | | . . . | . 6 . | . . . |
| . . . | . . . | . 6 6 | | . . . | . . . | . 6 6 |
| 6 6 . | . . . | . . . | | 6 6 . | . . . | . . . |
|-----------+-----------+-----------| |-----------+-----------+-----------|
| . . . | . . . | 6 . . | | . . . | . . . | 6 . . |
| 6 . *6 | *6 . . | . . . | | 6 . *6 | *6 . . | . . . |
| 6 6 . | . . 6 | . . . | | 6 6 . | . . 6 | . . . |
+-----------------------------------+ +-----------------------------------+
|
Code: | Siamese Sashimi X-Wing c34\r(23)8
identical to Skyscraper eliminations
+-----------------------------------+
| 6 6 . | . . 6 | . 6 6 |
| . . *6# | . . -6 | . 6 . |
| -6 -6 . | *6# . . | . 6 . |
|-----------+-----------+-----------|
| . . . | . 6 . | . . . |
| . . . | . . . | . 6 6 |
| 6 6 . | . . . | . . . |
|-----------+-----------+-----------|
| . . . | . . . | 6 . . |
| 6 . *6 | *6 . . | . . . |
| 6 6 . | . . 6 | . . . |
+-----------------------------------+
|
Each Sashimi X-Wing has only one fin cell. The Siamese Sashimi X-Wing combines the two Sashimi X-Wings because they are identical except for one cover set -- [r2] vs. [r3]. This produces eliminations identical to the Skyscraper.
Note: the concept of Siamese can be extended to larger fish with no problem. There is no larger pattern for Skyscraper.
wapati wrote: | They are rare, partly because most people, and some solvers will not spot them. Further, there are always ways around big fish, so no-one needs to spot them.
|
I would like to see your workaround for this elimination.
Code: | +--------------------------------------------------------------------------------+
| 6 78 2 | 3 478 9 | 1 5 478 |
| 17 3 18 | 2 4578 578 | 789 489 6 |
| 9 5 4 | 78 1 6 | 278 3 278 |
|--------------------------+--------------------------+--------------------------|
| 2 6 789 | 5 789 4 | 3 789 1 |
| 4 1 789 | 6789 2 3 | 5 6789 789 |
| 5 789 3 | 1 6789 78 | 2789 246789 24789 |
|--------------------------+--------------------------+--------------------------|
| 3 79 679 | 4 6789 2 | 789 1 5 |
| 8 4 15 | 79 3 15 | 6 279 279 |
| 17 2 1569 | 6789 56789 1578 | 4 789 3 |
+--------------------------------------------------------------------------------+
7-Fish r1458c16b7\r2c2389b5+r9|c5|b8 <> 7 [r9c5]
7-Fish r1458c1b27\r2c234589 <> 7 [r9c5]
7-Fish r18c167b47\r26c23b39+r9|c5|b8 <> 7 [r9c5]
7-Fish r18c16b467\r269c2389+r9|c5|b8 <> 7 [r9c5]
|
|
|
Back to top |
|
|
wapati
Joined: 10 Jun 2008 Posts: 472 Location: Brampton, Ontario, Canada.
|
Posted: Tue Jan 13, 2009 5:44 am Post subject: |
|
|
daj95376 wrote: |
I would like to see your workaround for this elimination.
Code: | +--------------------------------------------------------------------------------+
| 6 78 2 | 3 478 9 | 1 5 478 |
| 17 3 18 | 2 4578 578 | 789 489 6 |
| 9 5 4 | 78 1 6 | 278 3 278 |
|--------------------------+--------------------------+--------------------------|
| 2 6 789 | 5 789 4 | 3 789 1 |
| 4 1 789 | 6789 2 3 | 5 6789 789 |
| 5 789 3 | 1 6789 78 | 2789 246789 24789 |
|--------------------------+--------------------------+--------------------------|
| 3 79 679 | 4 6789 2 | 789 1 5 |
| 8 4 15 | 79 3 15 | 6 279 279 |
| 17 2 1569 | 6789 56789 1578 | 4 789 3 |
+--------------------------------------------------------------------------------+
7-Fish r1458c16b7\r2c2389b5+r9|c5|b8 <> 7 [r9c5]
7-Fish r1458c1b27\r2c234589 <> 7 [r9c5]
7-Fish r18c167b47\r26c23b39+r9|c5|b8 <> 7 [r9c5]
7-Fish r18c16b467\r269c2389+r9|c5|b8 <> 7 [r9c5]
|
|
SE does it this way. (It never finds big fish.)
Analysis results
Difficulty rating: 8.4
This Sudoku can be solved using the following logical methods:
39 x Hidden Single
2 x Claiming
2 x Naked Pair
1 x Hidden Pair
2 x XY-Wing
2 x Turbot Fish
2 x Bidirectional Cycle
13 x Forcing Chain
6 x Nishio Forcing Chains
1 x Cell Forcing Chains
3 x Region Forcing Chains
Soduko Cue finds Medusas and Nishio chains.
I see nothing that I know how to do, including whatever you are showing above. |
|
Back to top |
|
|
daj95376
Joined: 23 Aug 2008 Posts: 3854
|
Posted: Tue Jan 13, 2009 5:55 am Post subject: |
|
|
wapati: When your first puzzle is solved for basics and all smaller fish, the remaining Finned/Sashimi Swordfish have one fin cell each. (Your second puzzle contains an unfinned Swordfish after basics.)
Code: | *-----------------------------------------------------------*
| 2 19 5 | 4 69 68 | 18 7 3 |
| 4 38 38 | 2 1 7 | 5 6 9 |
| 7 19 6 | 389 3589 358 | 4 2 18 |
|-------------------+-------------------+-------------------|
| 6 7 4 | 1 58 58 | 9 3 2 |
| 5 2 39 | 6 39 4 | 18 18 7 |
| 89 38 1 | 39 7 2 | 6 4 5 |
|-------------------+-------------------+-------------------|
| 3 6 7 | 5 48 1 | 2 9 48 |
| 89 5 2 | 7 468 689 | 3 18 148 |
| 1 4 89 | 38 2 389 | 7 5 6 |
*-----------------------------------------------------------*
Sashimi Swordfish r147\c567 w/fin [r7c9] [r3c9]<>8
Sashimi Swordfish r147\c569 w/fin [r1c7] [r3c9]<>8
Finned Swordfish r347\c569 w/fin [r1c4] [r1c6]<>8
|
Keith: Consider the following Empty Rectangle elimination. It can also be viewed as ...
Code: | finned Franken X-Wing r8b1\c28 w/fins [r2c13]
[r2c13] are in base set [b1] but not in cover sets [c28]
+-----------------------------------+
| / 6 / | . . . | . . . |
| #6 6 #6 | . . . | . -6 . |
| / 6 / | . . . | . . . |
|-----------+-----------+-----------|
| . . . | . . . | . . . |
| . . . | . . . | . . . |
| . . . | . . . | . . . |
|-----------+-----------+-----------|
| . . . | . . . | . . . |
| / 6 / | / / / | / 6 / |
| . . . | . . . | . . . |
+-----------------------------------+
|
Marty: My apologies for helping to send your thread off on a tangent. |
|
Back to top |
|
|
Asellus
Joined: 05 Jun 2007 Posts: 865 Location: Sonoma County, CA, USA
|
Posted: Tue Jan 13, 2009 10:59 pm Post subject: |
|
|
I think this question about the "number of fins" is semantics. Any sort of finned fish has just one fin, but that fin can be comprised of multiple cells. Think of it as a "grouped fin."
wapati's Sashimi Swordfish might be clearer marked this way:
Code: | .------------------.------------------.------------------.
| 2 19 5 | 4 689 689 | 18 7 3 |
| 4 38 38 | 2 1 7 | 5 6 9 |
| 7 19 6 | 389 3589 3589 | 4 2 18 |
:------------------+------------------+------------------:
| 6 7 4 | 1 58 58 | 9 3 2 |
| 5 2 39 | 6 39 4 | 18 18 7 |
|#89 38 1 |#39 7 2 | 6 4 5 |
:------------------+------------------+------------------:
| 3 6 7 |*5 F489 1 | 2 #89 48 |
|#89 5 2 |*7 F4689 F689 | 3 #189 148 |
| 1 4 89 | 38-9 2 389 | 7 5 6 |
'------------------'------------------'------------------' |
The Swordfish part is marked #, with the missing bits marked *. The 3-cell fin is marked F. However...
wapati wrote: | I can make the same elimination with other steps but none I'd call basic. |
Not even a Kite? The one with pincers at r6c4 and r9c3 with pivot in b4 seems much simpler than the Sashimi Swordfish to me. Or, how about the ER in b5 and the conjugate <9>s in c3? Also simpler. |
|
Back to top |
|
|
keith
Joined: 19 Sep 2005 Posts: 3355 Location: near Detroit, Michigan, USA
|
Posted: Wed Jan 14, 2009 12:27 am Post subject: |
|
|
Asellus: Quote: | Any sort of finned fish has just one fin, but that fin can be comprised of multiple cells. |
Aha! A grouped fin!
Keith
(I think) |
|
Back to top |
|
|
ronk
Joined: 07 May 2006 Posts: 398
|
Posted: Wed Jan 14, 2009 3:54 pm Post subject: |
|
|
Asellus wrote: | I think this question about the "number of fins" is semantics. Any sort of finned fish has just one fin, but that fin can be comprised of multiple cells. Think of it as a "grouped fin." |
In this case, that's correct. However, sometimes all the fin cells don't "see" each other. |
|
Back to top |
|
|
wapati
Joined: 10 Jun 2008 Posts: 472 Location: Brampton, Ontario, Canada.
|
Posted: Wed Jan 14, 2009 4:20 pm Post subject: |
|
|
keith wrote: | Wapati,
I am enjoying this discussion!
However,
Do you have a real example? A swordfish (6 cells in 3 rows or columns) plus two essential extra "fin" cells that make an elimination not made by more elementary methods? Where both fins and the swordfish are required in the logic?
In my practical experience, almost all of these "almost fish" turn out to be simplified by line - box interactions. No higher theory needed.
Keith |
Found one, it took a while!
Code: |
2 . .|. . 1|. 4 .
. . 4|9 . 3|. . .
. 6 3|8 4 .|2 . .
-----+-----+-----
. 4 1|. 7 .|5 3 .
. . 8|4 . .|. . .
6 7 .|. . .|. . .
-----+-----+-----
. . 9|1 . .|. . .
4 . .|3 . .|. . 6
. . .|. . .|. 2 9
.---------------------.---------------------.---------------------.
| 2 9 *57 |*567 *56 1 | 38 4 38 |
| 1578 158 4 | 9 2 3 | 167 1567 157 |
| 157 6 3 | 8 4 57 | 2 9 157 |
:---------------------+---------------------+---------------------:
| 9 4 1 | 26 7 68 | 5 3 28 |
| 35 235 8 | 4 1 59 | 679 67 27 |
| 6 7 *25 |*25 3 89 | 1489 18 148 |
:---------------------+---------------------+---------------------:
| 3578 2358 9 | 1 #568 24567 | 3478 578 34578 |
| 4 1258 *257 | 3 9 27-5 | 178 1578 6 |
| 13578 1358 6 |#57 #58 457 | 13478 2 9 |
'---------------------'---------------------'---------------------' |
|
|
Back to top |
|
|
daj95376
Joined: 23 Aug 2008 Posts: 3854
|
Posted: Wed Jan 14, 2009 5:44 pm Post subject: |
|
|
ronk: Does wapati's Sashimi Swordfish c345\r168 contains two fins (aka a fin sector, a fin unit) -- [c4] & [c5] -- and three fin cells ?
everyone else: Sorry, this discussion is a carry-over from the Players' Forums. |
|
Back to top |
|
|
keith
Joined: 19 Sep 2005 Posts: 3355 Location: near Detroit, Michigan, USA
|
Posted: Wed Jan 14, 2009 8:36 pm Post subject: |
|
|
daj95376 wrote: | everyone else: Sorry, this discussion is a carry-over from the Players' Forums. |
Can you post a link?
Keith |
|
Back to top |
|
|
daj95376
Joined: 23 Aug 2008 Posts: 3854
|
Posted: Wed Jan 14, 2009 9:50 pm Post subject: |
|
|
keith wrote: | daj95376 wrote: | everyone else: Sorry, this discussion is a carry-over from the Players' Forums. |
Can you post a link?
|
It starts here ... and the head message of the thread has been updated. I think it will be very confusing reading. ronk and I seemed to be on different wavelengths.
http://www.sudoku.com/boards/viewtopic.php?p=65864#65864
Last edited by daj95376 on Wed Jan 14, 2009 9:59 pm; edited 1 time in total |
|
Back to top |
|
|
ronk
Joined: 07 May 2006 Posts: 398
|
Posted: Wed Jan 14, 2009 9:57 pm Post subject: |
|
|
daj95376 wrote: | ronk: Does wapati's Sashimi Swordfish c345\r168 contains two fins (aka a fin sector, a fin unit) -- [c4] & [c5] -- and three fin cells ? |
Three fin cells (as shown on your pencilmarks) in one fin unit b8.
As not one of the three fin cells sees the elimination cell via a row or a column, there's not even another possibility with that interpretation.
Using the franken fish interpretation, c345\r16b8, there would only be one fin cell r8c3.
Last edited by ronk on Wed Jan 14, 2009 10:05 pm; edited 1 time in total |
|
Back to top |
|
|
Asellus
Joined: 05 Jun 2007 Posts: 865 Location: Sonoma County, CA, USA
|
Posted: Thu Jan 15, 2009 11:23 pm Post subject: |
|
|
ronk wrote: | In this case, that's correct. However, sometimes all the fin cells don't "see" each other. |
Short of an example, I don't see that that makes a difference. Each fin cell needs to destroy the fish and each fin cell must "see" the victim(s) in some manner or other. So, they would still function as a group even if not peers. Or, is there some other sort of multi-finned fish that works in some other way? |
|
Back to top |
|
|
ronk
Joined: 07 May 2006 Posts: 398
|
Posted: Fri Jan 16, 2009 12:24 am Post subject: |
|
|
Asellus wrote: | ronk wrote: | In this case, that's correct. However, sometimes all the fin cells don't "see" each other. |
Short of an example, I don't see that that makes a difference. Each fin cell needs to destroy the fish and each fin cell must "see" the victim(s) in some manner or other. So, they would still function as a group even if not peers. Or, is there some other sort of multi-finned fish that works in some other way? |
It's just one way of looking at it, but it's also a matter of convention and notational convenience. For the below ... Code: |
* * X | . . . | . * .
X *X / | / / / | # X /
* * X | . . . | . * .
---------+----------+----------
. * / | . . . | . * .
* * X | * * * | ** * *
/ X / | / / / | # X #
---------+----------+----------
. * / | . . . | . * .
/ X / | / / / | # X /
. * / | . . . | . * . |
... the fish is completely defined by r268c3\r5c278b16 ==> r5c7<>X
No need to write r268c7 and r6c9 to detail the fin locations in c7 and b6. |
|
Back to top |
|
|
keith
Joined: 19 Sep 2005 Posts: 3355 Location: near Detroit, Michigan, USA
|
Posted: Sat Jan 17, 2009 1:22 am Post subject: |
|
|
So, in simple language, what is a "fin"?
I think it is easy: Either we have a pattern (X-wing, XY-wing, ...) or we have one other feature (the "fin") that invalidates the pattern.
Either the pattern is true, or the fin is true.
My original point, and I'm sticking to it! If the logic of a fin is A OR B is true, what does C have to do with it?
What is a "fin"?
Keith |
|
Back to top |
|
|
ronk
Joined: 07 May 2006 Posts: 398
|
Posted: Sat Jan 17, 2009 2:09 am Post subject: |
|
|
keith wrote: | So, in simple language, what is a "fin"?
I think it is easy: Either we have a pattern (X-wing, XY-wing, ...) or we have one other feature (the "fin") that invalidates the pattern.
Either the pattern is true, or the fin is true.
My original point, and I'm sticking to it! If the logic of a fin is A OR B is true, what does C have to do with it?
What is a "fin"? |
Extra candidate(s) in the same sector (unit) as elimination candidate(s). |
|
Back to top |
|
|
keith
Joined: 19 Sep 2005 Posts: 3355 Location: near Detroit, Michigan, USA
|
Posted: Sat Jan 17, 2009 3:40 am Post subject: |
|
|
ronk wrote: | keith wrote: | So, in simple language, what is a "fin"?
I think it is easy: Either we have a pattern (X-wing, XY-wing, ...) or we have one other feature (the "fin") that invalidates the pattern.
Either the pattern is true, or the fin is true.
My original point, and I'm sticking to it! If the logic of a fin is A OR B is true, what does C have to do with it?
What is a "fin"? |
Extra candidate(s) in the same sector (unit) as elimination candidate(s). |
Really?
So my statement, "Either the pattern is true, or the fin is true", is not correct?
Keith
PS: Please define: "Extra candidate(s)", "sector", "unit", and "elimination candidate(s)". |
|
Back to top |
|
|
ronk
Joined: 07 May 2006 Posts: 398
|
Posted: Sat Jan 17, 2009 4:11 am Post subject: |
|
|
keith wrote: | Please define: "Extra candidate(s)", "sector", "unit", and "elimination candidate(s)". |
I'm confident you know the definitions for at least three of those terms, so I think you are just trolling. Adios. |
|
Back to top |
|
|
|
|
You cannot post new topics in this forum You cannot reply to topics in this forum You cannot edit your posts in this forum You cannot delete your posts in this forum You cannot vote in polls in this forum
|
Powered by phpBB © 2001, 2005 phpBB Group
|