| View previous topic :: View next topic |
| Author |
Message |
daj95376
Joined: 23 Aug 2008
Posts: 3854
|
 Posted: Thu Aug 19, 2010 7:23 am Post subject: Puzzle 10/08/19: B |
 |
|
| Code: | +-----------------------+
| 5 . . | . . . | . . 9 |
| . 2 . | 1 9 5 | . . . |
| . . 7 | . . . | 5 . . |
|-------+-------+-------|
| . 6 . | . . 1 | . 5 . |
| . 9 . | . 7 . | 6 . . |
| . 5 . | 2 . 6 | 1 9 8 |
|-------+-------+-------|
| . . 6 | . 1 2 | 9 . 5 |
| . . . | 7 . 9 | . . 3 |
| 1 . . | . . 4 | 2 7 . |
+-----------------------+
|
Play this puzzle online at the Daily Sudoku site |
|
| Back to top |
|
 |
JC Van Hay
Joined: 13 Jun 2010
Posts: 494
Location: Charleroi, Belgium
|
| Posted: Thu Aug 19, 2010 12:09 pm Post subject: |
|
|
| Quote: | Mutant Swordfish : (3)R6B12/R3C35 : r4c5<>3, r3c5<>3 (autocannibalism)
X Wing : (4)R57/C18 : r2c8,r4c1<>4
XY Wing : (48)R4C5 (83)R4C1 (34)R6C3 : R6C5<>4 |
|
|
| Back to top |
|
|
peterj
Joined: 26 Mar 2010
Posts: 974
Location: London, UK
|
| Posted: Thu Aug 19, 2010 1:12 pm Post subject: |
|
|
Probably a cleaner wing in there but this was the first I saw...
| Quote: | | w-wing(34) with transport (4=3)r5c8 - r5c6=r3c6 - (3=4)r3c2 - r8c2=r7c1 ; r7c8<>4 |
So what is autocannibalism?! |
|
| Back to top |
|
|
JC Van Hay
Joined: 13 Jun 2010
Posts: 494
Location: Charleroi, Belgium
|
| Posted: Thu Aug 19, 2010 1:46 pm Post subject: |
|
|
Peter, nice one-step proof!
Autocannibalism is the elimination of a candidate in a SIS (base set), while an ordinary elimination occurs in a WIS (cover set). In principle, it is always an indication of a simpler move. Here, (3)r3c5 is eliminated by an Empty Rectangle (3)B1R6.
JC |
|
| Back to top |
|
|
peterj
Joined: 26 Mar 2010
Posts: 974
Location: London, UK
|
| Posted: Thu Aug 19, 2010 2:03 pm Post subject: |
|
|
| JC, thanks. I think I follow the SIS, WIS approach - I can see the advantage of brevity though sometimes the lack of explicitness seems a little opaque! Does it change the way you actually go about solving puzzles? |
|
| Back to top |
|
|
ronk
Joined: 07 May 2006
Posts: 398
|
| Posted: Thu Aug 19, 2010 4:09 pm Post subject: |
|
|
| JC Van Hay wrote: | | Autocannibalism is the elimination of a candidate in a SIS (base set) ... |
The "auto-" prefix is superfluous IMO. What other kind of sudoku cannibalism is there? |
|
| Back to top |
|
|
daj95376
Joined: 23 Aug 2008
Posts: 3854
|
| Posted: Thu Aug 19, 2010 4:57 pm Post subject: |
|
|
I don't check for cannibalism. Fortunately, the current FISH definitions make it extraneous.
| Quote: | The candidates forming the fish pattern are divided into vertices, fins and potential eliminations
depending on the number of Base/Cover sectors that the particular candidate exists in:
Fin (F): Base sectors > Cover sectors
Vertex (V): Base sectors = Cover sectors
Potential Elimination (PE): Base sectors < Cover sectors
|
| Quote: | Eventual Elimination(EE): A potential elimanation that can be linked (through the Pattern) to ALL fin cells. In the abscence of fins, EE(s)=PE(s)
|
A non-cannibalistic alternative; i.e., PEs have base sector counts of zero:
| Code: | Franken Swordfish r126\c35b3 ___ r34c5<>3
+--------------------------------------------------------------+
| 5 1 348 | 46 23 7 | 348 26 9 |
| 6 2 348 | 1 9 5 | 3478 348 47 |
| 9 34 7 | 46 238 38 | 5 26 1 |
|--------------------+--------------------+--------------------|
| 348 6 2 | 9 348 1 | 347 5 47 |
| 348 9 1 | 5 7 38 | 6 34 2 |
| 7 5 34 | 2 34 6 | 1 9 8 |
|--------------------+--------------------+--------------------|
| 34 7 6 | 38 1 2 | 9 48 5 |
| 2 48 5 | 7 6 9 | 48 1 3 |
| 1 38 9 | 38 5 4 | 2 7 6 |
+--------------------------------------------------------------+
# 41 eliminations remain
|
Besides, I always thought cannibalism was a one-sided discrimination because I'm unaware of any label being applied to a fin cell when its cover sector count is non-zero! _  _ |
|
| Back to top |
|
|
JC Van Hay
Joined: 13 Jun 2010
Posts: 494
Location: Charleroi, Belgium
|
| Posted: Thu Aug 19, 2010 8:23 pm Post subject: |
|
|
| ronk wrote: | | JC Van Hay wrote: | | Autocannibalism is the elimination of a candidate in a SIS (base set) ... |
The "auto-" prefix is superfluous IMO. What other kind of sudoku cannibalism is there? |
I used the term autocannibalism as it is defined in the Ultimate fish Guide here | Quote: | | AutoCannibalistic Fish: A fish with at least one Eventual Elimination in a base sector |
But, I would agree to drop the "auto-" prefix. I would indeed prefer the Allan Barker's expression "Cannibalized Logic". So, should "AutoCannibalistic Fish" be replaced by "Cannibalistic Fish" in the Ultimate Fish Guide  Besides, the "Cannibalized Logic" is not limited to Fish. Furthermore, it may generally be avoided by a simpler logic.
In the present puzzle, I just condensed into one step two simple moves that explain the previously determined extraneous candidates (3)r34c5: an Empty Recangle (3)R6B1 => r4c5<>3 and a Grouped X-Chain (3)C6R6B1 => r3c5<>3. I however overlooked the Rank 0 Mutant Swordfish (as it appears now clearly while writing this post) (3)R6C6B1/R3C3B5. Instead I was satisfied by the rapidly found "Cannibalistic Mutant Swordfish". The Rank 0 Franken Swordfish found by Danny is even simpler 
Peter, I hope you will find in the above details some elements of answer to your question.
JC |
|
| Back to top |
|
|
ronk
Joined: 07 May 2006
Posts: 398
|
| Posted: Fri Aug 20, 2010 12:40 am Post subject: |
|
|
| JC Van Hay wrote: | | ronk wrote: | | The "auto-" prefix is superfluous IMO. What other kind of sudoku cannibalism is there? |
I used the term autocannibalism as it is defined in the Ultimate fish Guide here | Quote: | | AutoCannibalistic Fish: A fish with at least one Eventual Elimination in a base sector |
|
I know ... and I guess I should try to persuade tarek to drop the "auto" prefix.
| JC Van Hay wrote: | | ... the "Cannibalized Logic" is not limited to Fish. Furthermore, it may generally be avoided by a simpler logic. |
I agree with both points. However, on occasion it's more elegant to use cannibalistic rank 0 logic ... rather than first pick off cannibalized candidate(s) wth rank 1 logic.
| JC Van Hay wrote: | | In the present puzzle ... I however overlooked the Rank 0 Mutant Swordfish (as it appears now clearly while writing this post) (3)R6C6B1/R3C3B5. Instead I was satisfied by the rapidly found "Cannibalistic Mutant Swordfish". The Rank 0 Franken Swordfish found by Danny is even simpler |
Because it can also be written as a chain (a continuous loop), I like your mutant swordfsh the best.
(3)r3c2 = (3)r12c3 - (3)r6c3 = (3)r6c5 - (3)r5c6 = (3)r3c6 - loop |
|
| Back to top |
|
|
tlanglet
Joined: 17 Oct 2007
Posts: 2468
Location: Northern California Foothills
|
| Posted: Fri Aug 20, 2010 2:44 am Post subject: |
|
|
A different three stepper...........
| Quote: | 6-cell ADP (38)r345c156; to prevent DP, r3c5=2, r45c1=4 or r4c5=4
(2)r3c5-(2=6)r3c8-(6=4)r3c4-(4=3)r3c2-r3c56=r1c5; r34c5<>3
(4)r45c1-r7c1=r8c2-(4=3)r3c2-r3c56=r1c5; r34c5<>3
(4)r4c5-(4=3)r6c5; r34c5<>3
x-wing (4)r57c18; r2c8,r4c1<>4
xy-wing 3-48 vertex (38)r4c1; r6c5<>4 |
Ted |
|
| Back to top |
|
|
Marty R.
Joined: 12 Feb 2006
Posts: 5770
Location: Rochester, NY, USA
|
| Posted: Fri Aug 20, 2010 10:26 pm Post subject: |
|
|
| I played a W-Wing (34) with a couple of transports but it didn't take me as far as it took Peter. A Type 4 UR (47), then looking at the potential DP (38) in boxes 13; r12c3<>4. |
|
| Back to top |
|
|
|
FAQ
Search
Memberlist
Usergroups
Register
Profile
Log in to check your private messages
Log in
