| View previous topic :: View next topic |
| Author |
Message |
keith
Joined: 19 Sep 2005
Posts: 3355
Location: near Detroit, Michigan, USA
|
 Posted: Tue Jan 18, 2011 12:33 pm Post subject: A vh Vanhegan |
 |
|
| Code: | Puzzle: VH4-892692vh
+-------+-------+-------+
| 1 . . | . . 5 | 4 . . |
| 4 . 8 | . 1 6 | . . 7 |
| . . 6 | . . 9 | . . . |
+-------+-------+-------+
| . . 9 | . . . | . 7 . |
| 7 . 3 | . 2 . | 8 . 1 |
| . 1 . | . . . | 6 . . |
+-------+-------+-------+
| . . . | 6 . . | 1 . . |
| 6 . . | 4 7 . | 5 . 9 |
| . . 5 | 9 . . | . . 4 |
+-------+-------+-------+ |
Keith |
|
| Back to top |
|
 |
cgordon
Joined: 04 May 2007
Posts: 769
Location: ontario, canada
|
| Posted: Tue Jan 18, 2011 3:57 pm Post subject: |
|
|
| That was a very weird puzzle ! I was mostly left with pairs of 23s. Couldn't find any wings - extended or otherwise - or any UR's. All I could manage was to make the odd ball pair (35) in R6C9 either a (3) or (5) and colour the chain of possibilities which left R7C9 a <2> in both cases. Not a very satisfying solution though. |
|
| Back to top |
|
|
Marty R.
Joined: 12 Feb 2006
Posts: 5770
Location: Rochester, NY, USA
|
| Posted: Tue Jan 18, 2011 5:40 pm Post subject: |
|
|
| Code: |
+----------+------------+------------+
| 1 9 7 | 238 38 5 | 4 23 6 |
| 4 23 8 | 23 1 6 | 9 5 7 |
| 23 5 6 | 7 4 9 | 23 1 8 |
+----------+------------+------------+
| 58 4 9 | 1 6 38 | 23 7 235 |
| 7 6 3 | 5 2 4 | 8 9 1 |
| 58 1 2 | 38 9 7 | 6 4 35 |
+----------+------------+------------+
| 9 7 4 | 6 5 238 | 1 238 23 |
| 6 38 1 | 4 7 238 | 5 238 9 |
| 23 238 5 | 9 38 1 | 7 6 4 |
+----------+------------+------------+
|
Play this puzzle online at the Daily Sudoku site
Looking at the potential 28 DP in boxes 89, one of the four cells must be a 3. The possibilities yield some common outcomes which solves it. |
|
| Back to top |
|
|
peterj
Joined: 26 Mar 2010
Posts: 974
Location: London, UK
|
| Posted: Tue Jan 18, 2011 6:10 pm Post subject: |
|
|
I think there be big fish swimming in this one! All those 3s...
Here's a Broken Wing solution...
| Code: | *--------------------------------------------------*
| 1 9 7 | 28+3 38 5 | 4 2-3 6 |
| 4 *23 8 |*23 1 6 | 9 5 7 |
|*23 5 6 | 7 4 9 |*2(3) 1 8 |
|----------------+----------------+----------------|
| 58 4 9 | 1 6 38 |*2(3) 7 25+3 |
| 7 6 3 | 5 2 4 | 8 9 1 |
| 58 1 2 |*38 9 7 | 6 4 *35 |
|----------------+----------------+----------------|
| 9 7 4 | 6 5 238 | 1 238 23 |
| 6 38 1 | 4 7 238 | 5 238 9 |
| 23 238 5 | 9 38 1 | 7 6 4 |
*--------------------------------------------------*
7-Cell Broken Wing with guardians (3)r1c4, (3)r4c9
BrokenWing(3)[(3)r1c4=r4c9] - r4c7=r3c7 ; r1c8<>3 |
Lots of x-chains that crack it also...
| Code: | | (3)r9c1=r3c1 - r3c7=r4c7 - r4c6=r78c6 ; r9c5<>3 |
I suppose that's a sashimi swordfish!
cgordon, welcome to the other parts of the forum! Did you try just normal multi-colouring on 3 because that should catch these x-chains? |
|
| Back to top |
|
|
keith
Joined: 19 Sep 2005
Posts: 3355
Location: near Detroit, Michigan, USA
|
| Posted: Tue Jan 18, 2011 7:58 pm Post subject: |
|
|
| Marty R. wrote: | | Looking at the potential 28 DP in boxes 89, one of the four cells must be a 3. The possibilities yield some common outcomes which solves it. |
Wow!  To explain what I think Marty is saying:
After basics: | Code: | +-------------+-------------+-------------+
| 1 9 7 | 238 38c 5 | 4 23d 6 |
| 4 23 8 | 23 1 6 | 9 5 7 |
| 23 5 6 | 7 4 9 | 23 1 8 |
+-------------+-------------+-------------+
| 58 4 9 | 1 6 38 | 23 7 235 |
| 7 6 3 | 5 2 4 | 8 9 1 |
| 58 1 2 | 38 9 7 | 6 4 35 |
+-------------+-------------+-------------+
| 9 7 4 | 6 5 238 | 1 -238 23a |
| 6 38 1 | 4 7 238 | 5 -238 9 |
| 23 238 5 | 9 38b 1 | 7 6 4 |
+-------------+-------------+-------------+ |
One of R78C68 is 3, so ab are a pseudocell 28. cd complete an XY-wing -238, making the eliminations shown and solving the puzzle.
Otherwise | Code: | +-------------+-------------+-------------+
| 1 9 7 |2-38 38* 5 | 4 23 6 |
| 4 23* 8 | 23* 1 6 | 9 5 7 |
| 23* 5 6 | 7 4 9 | 23 1 8 |
+-------------+-------------+-------------+
| 58 4 9 | 1 6 38 | 23 7 235 |
| 7 6 3 | 5 2 4 | 8 9 1 |
| 58 1 2 | 38 9 7 | 6 4 35 |
+-------------+-------------+-------------+
| 9 7 4 | 6 5 238 | 1 238 23 |
| 6 38 1 | 4 7 238 | 5 238 9 |
| 23* 2-38 5 | 9 38* 1 | 7 6 4 |
+-------------+-------------+-------------+ |
there is a Franken swordfish, eliminating a couple of 3s. Which does not solve it. I still need an extended XY-wing 2-38.
Keith |
|
| Back to top |
|
|
cgordon
Joined: 04 May 2007
Posts: 769
Location: ontario, canada
|
| Posted: Tue Jan 18, 2011 9:07 pm Post subject: |
|
|
| Quote: | | Looking at the potential 28 DP in boxes 89, one of the four cells must be a 3. |
Don't get that ! There are other places for 3s that don't include the four cells.
| Quote: | | Did you try just normal multi-colouring on 3 |
Not sure what multi-colouring is - but I tried regular coluring on every thing and it didn't work. |
|
| Back to top |
|
|
Marty R.
Joined: 12 Feb 2006
Posts: 5770
Location: Rochester, NY, USA
|
| Posted: Tue Jan 18, 2011 10:17 pm Post subject: |
|
|
| Quote: | | Don't get that ! There are other places for 3s that don't include the four cells. |
If one of those four cells isn't a 3, then you get the DP 28-28-28-28. |
|
| Back to top |
|
|
Marty R.
Joined: 12 Feb 2006
Posts: 5770
Location: Rochester, NY, USA
|
| Posted: Tue Jan 18, 2011 10:32 pm Post subject: |
|
|
| Quote: | | To explain what I think Marty is saying: |
No pseudo cells or other cerebral stuff.
R78c6=3-->r9c5=8-->r1c5=3-->r1c8=2-->r3c7=3
R78c8=3-->r7c9=2-->r4c7=2-->r3c7=3 |
|
| Back to top |
|
|
peterj
Joined: 26 Mar 2010
Posts: 974
Location: London, UK
|
| Posted: Tue Jan 18, 2011 11:34 pm Post subject: |
|
|
| Marty R. wrote: | | The possibilities yield some common outcomes which solves it. |
Hard to know exactly what path was envisaged 
Another interpretation is an extended w-wing(23)
| Code: | | (2=3)r1c8 - r1c5=r9c5 - UR(38)[(3)r78c6=r78c8] - (3=2)r7c9 ; r78c8<>2 |
[Edit] Oops! No, crossed posts, that isn't what Marty had in mind. |
|
| Back to top |
|
|
keith
Joined: 19 Sep 2005
Posts: 3355
Location: near Detroit, Michigan, USA
|
| Posted: Tue Jan 18, 2011 11:44 pm Post subject: |
|
|
| Marty R. wrote: | | Quote: | | To explain what I think Marty is saying: |
No pseudo cells or other cerebral stuff.
R78c6=3-->r9c5=8-->r1c5=3-->r1c8=2-->r3c7=3
R78c8=3-->r7c9=2-->r4c7=2-->r3c7=3 |
Marty, I think your logic is the same as I had in my message.
Keith |
|
| Back to top |
|
|
cgordon
Joined: 04 May 2007
Posts: 769
Location: ontario, canada
|
| Posted: Thu Jan 20, 2011 11:03 pm Post subject: |
|
|
Sorry Chaps:
The idea of those four 238 cells having to contain at least one <3> to avoid a DP was brilliant. But please correct me if I am wrong - I found the ensuing "path" solutions arbitrary. Nothing seemed to be tangibly related to any "definitive" techniques - just a case of "try this and let's see".
Could it be that for some Sudokos - this is the only solution. |
|
| Back to top |
|
|
keith
Joined: 19 Sep 2005
Posts: 3355
Location: near Detroit, Michigan, USA
|
| Posted: Fri Jan 21, 2011 3:54 am Post subject: |
|
|
| cgordon wrote: | Sorry Chaps:
The idea of those four 238 cells having to contain at least one <3> to avoid a DP was brilliant. But please correct me if I am wrong - I found the ensuing "path" solutions arbitrary. Nothing seemed to be tangibly related to any "definitive" techniques - just a case of "try this and let's see".
Could it be that for some Sudokos - this is the only solution. |
Chaps? :wink:
Let's start here: | Code: | +-------------+-------------+-------------+
| 1 9 7 | 238 38 5 | 4 23 6 |
| 4 23 8 | 23 1 6 | 9 5 7 |
| 23 5 6 | 7 4 9 | 23 1 8 |
+-------------+-------------+-------------+
| 58 4 9 | 1 6 38 | 23 7 235 |
| 7 6 3 | 5 2 4 | 8 9 1 |
| 58 1 2 | 38 9 7 | 6 4 35 |
+-------------+-------------+-------------+
| 9 7 4 | 6 5 238 | 1 238 23 |
| 6 38 1 | 4 7 238 | 5 238 9 |
| 23 238 5 | 9 38 1 | 7 6 4 |
+-------------+-------------+-------------+ |
Note that R4C1 is either 5 or it is 8. If so, ... That is what we all do all the time.
I learned the following kind of logic from Myth Jellies and re'born, although Marty seems to be more of a master than I am.
In this case, there is a potential Deadly Pattern in R78C68. It is either 23, 28, or 38. To destroy the DP:
The DP is 38:
One of R7C68 and/or one of R8C68 is 2
or
One of R78C6 and/or one of R78C8 is 2
Yes, because of the strong links on 2 in C6 and R8.
Or:
The DP is 23:
One of R7C68 and/or one of R8C68 is 8
or
One of R78C6 and/or one of R78C8 is 8
Yes, because of the strong links on 8 in C8 and R7.
Or:
The DP is 28:
One of R7C68 and/or one of R8C68 is 3
=> R7C9 is 2 and/or R8C2 is 8 (They are a pseudocell 28).
or
One of R78C6 and/or one of R78C8 is 3
=> (R9C5 and R4C6 are 8) and/or (R7C9 and R1C8 are 2)
Making four potential pseudocells 28.
I have no logical or intellectual problem fitting one of these 28 pseudocells into a pattern like an XY-wing. In fact, I think it is rather elegant.
Keith |
|
| Back to top |
|
|
cgordon
Joined: 04 May 2007
Posts: 769
Location: ontario, canada
|
| Posted: Fri Jan 21, 2011 3:17 pm Post subject: |
|
|
Gorrit!! One of the 238 cells has to be a 3. One of them in in C6 has to be a 28. Ergo there's an xy wing 238 (though didn't see the R4C1 connection).
Surgite. |
|
| Back to top |
|
|
|
FAQ
Search
Memberlist
Usergroups
Register
Profile
Log in to check your private messages
Log in
