| View previous topic :: View next topic |
| Author |
Message |
tlanglet
Joined: 17 Oct 2007
Posts: 2468
Location: Northern California Foothills
|
 Posted: Fri Sep 05, 2008 6:42 pm Post subject: Vanhegan Fiendish 9/5 |
 |
|
Since we did not again have a VH locally today, I tried one at Vanhegan and found it very interesting and lengthy. Initially I noticed several choices for the identical deletions, and later I still used a variety of steps to solve the puzzle. In no order, I used :x-wings, xy-wings,two swordfish, kites, skyscrapper, UR, and BUG+2.
Overall it had a lot of VH moves.
| Code: |
*-----------*
|...|.7.|21.|
|1..|5..|8..|
|...|.1.|..6|
|---+---+---|
|..3|1..|68.|
|86.|...|.79|
|.27|..4|5..|
|---+---+---|
|3..|.5.|...|
|..6|..9|..5|
|.14|.6.|...|
*-----------*
|
Hope it enjoy it also.
Ted  |
|
| Back to top |
|
 |
storm_norm
Joined: 18 Oct 2007
Posts: 1741
|
| Posted: Sat Sep 06, 2008 6:33 am Post subject: |
|
|
| Code: | .------------------.------------------.------------------.
| 6 489 5 | 489 7 38 | 2 1 34 |
| 1 3 29 | 5 24 6 | 8 49 7 |
| 27 4789 289 | 2489 1 238 | 39 5 6 |
:------------------+------------------+------------------:
| 4 5 3 | 1 9 7 | 6 8 2 |
| 8 6 1 | 23 23 5 | 4 7 9 |
| 9 2 7 | 6 8 4 | 5 3 1 |
:------------------+------------------+------------------:
| 3 89 289 | 248 5 1 | 7 6 48 |
| 27 78 6 | 2348 234 9 | 1 24 5 |
| 5 1 4 | 7 6 28 | 39 29 38 |
'------------------'------------------'------------------' |
two step solution:
(8=3)r1c6-(3)r3c6=(3)r3c7-(3)r9c7=(3-8)r9c9=(8)r7c9-(8)r7c3=(8)r3c3; r3c4|r3c6 <> 8
after those 8's are gone, then the subsequent pointing pair eliminates the 8 in r1c2, just so no one is confused about this next move showing only a bivalue cell in r1c2.
(4=9)r1c2-(8=9)r7c2-(8=4)r7c9; r1c9 <> 4...
this is also the fancy way of writing out the xy-wing on {4,8,9}
done.
Last edited by storm_norm on Sat Sep 06, 2008 5:52 pm; edited 2 times in total |
|
| Back to top |
|
|
cgordon
Joined: 04 May 2007
Posts: 769
Location: ontario, canada
|
| Posted: Sat Sep 06, 2008 12:58 pm Post subject: |
|
|
I plodded and hacked away at this one because I am xy-wing challenged.
x-wing on 4
x-wing on 3
Tpe 4 UR on 23
ER on 2
Colouring on 2
Finned x-wing on 8
...then enfin an xy-wing for 238.
Good One ! |
|
| Back to top |
|
|
ravel
Joined: 21 Apr 2006
Posts: 536
|
| Posted: Sat Sep 06, 2008 1:22 pm Post subject: |
|
|
Nice puzzle again.
As cgordon says, a swordfish is not needed, e.g. kites and finned x-wing also do it (with x-wing and xy-wings).
Norms first chain can be viewed as M-wing 38 in r1c6/r9c9 (connected by the strong links for 3 in row 1 and column 9) with strong link for 8 in row 9 to r7c9, transported to r3c3 via the strong link in column 3.
A one-step solution is a finned M-wing 24 in r7c4/r8c8:
| Code: | .------------------.------------------.------------------.
| 6 489 5 | 489 7 38 | 2 1 34 |
| 1 3 29 | 5 24 6 | 8 49 7 |
| 27 4789 289 | 2489 1 238 | 39 5 6 |
:------------------+------------------+------------------:
| 4 5 3 | 1 9 7 | 6 8 2 |
| 8 6 1 | 23 23 5 | 4 7 9 |
| 9 2 7 | 6 8 4 | 5 3 1 |
:------------------+------------------+------------------:
| 3 89 @289 |#24+8 5 1 | 7 6 *48 |
|-27 78 6 | 2348 234 9 | 1 #24 5 |
| 5 1 4 | 7 6 28 | 39 29 38 |
'------------------'------------------'------------------' |
The fin is the 8 in r7c4.
Without the fin the 2 strong links for 4 in row 7/box 9 connect the pairs (both must be 2 or both 4). With the strong link for 2 in row 7 one of r7c3 and r8c8 must be 2.
If the fin is true, r7c4=8 implies r7c9=4 and r8c8=2 also.
So r8c1 cannot be 2. |
|
| Back to top |
|
|
daj95376
Joined: 23 Aug 2008
Posts: 3854
|
| Posted: Sat Sep 06, 2008 3:40 pm Post subject: |
|
|
| 2x Locked Candidate 1, Locked Candidate 2, X-Wing (4), Swordfish (2), XY-Wing |
|
| Back to top |
|
|
arkietech
Joined: 31 Jul 2008
Posts: 1834
Location: Northwest Arkansas USA
|
| Posted: Sat Sep 06, 2008 3:56 pm Post subject: |
|
|
| ravel wrote: | A one-step solution is a finned M-wing 24 in r7c4/r8c8:
| Code: | .------------------.------------------.------------------.
| 6 489 5 | 489 7 38 | 2 1 34 |
| 1 3 29 | 5 24 6 | 8 49 7 |
| 27 4789 289 | 2489 1 238 | 39 5 6 |
:------------------+------------------+------------------:
| 4 5 3 | 1 9 7 | 6 8 2 |
| 8 6 1 | 23 23 5 | 4 7 9 |
| 9 2 7 | 6 8 4 | 5 3 1 |
:------------------+------------------+------------------:
| 3 89 @289 |#24+8 5 1 | 7 6 *48 |
|-27 78 6 | 2348 234 9 | 1 #24 5 |
| 5 1 4 | 7 6 28 | 39 29 38 |
'------------------'------------------'------------------' |
The fin is the 8 in r7c4.
|
Bird or shark? a finned extended wing. What fun! |
|
| Back to top |
|
|
tlanglet
Joined: 17 Oct 2007
Posts: 2468
Location: Northern California Foothills
|
| Posted: Sat Sep 06, 2008 10:20 pm Post subject: |
|
|
As I indicated in my original post, I enjoyed this puzzle because of the number and variety of VH techniques possible/necessary for me to complete it.
I can now say that I am also envious of the elegance of some of the further advanced techniques presented that solve the puzzle in fewer moves. I also greatly appreciate the detailed explanations of these more complex solutions as an aid for learning them myself. Thanks to all .......
Ted  |
|
| Back to top |
|
|
Johan
Joined: 25 Jun 2007
Posts: 206
Location: Bornem Belgium
|
| Posted: Sat Sep 06, 2008 10:26 pm Post subject: |
|
|
Another one step is the 23 UR* in R58C45, for avoiding the 23 DP either/or both R8C45=4, R8C4=8, which eliminates both <2>'s in R8C45 and <2> in R9C8.
R8C45=4 => R8C8=2(a)
R8C4=8 => R9C6=2(b)
| Code: | +------------------------+------------------------+------------------------+
| 6 489 5 | 489 7 38 | 2 1 34 |
| 1 3 29 | 5 24 6 | 8 49 7 |
| 27 4789 289 | 2489 1 238 | 39 5 6 |
+------------------------+------------------------+------------------------+
| 4 5 3 | 1 9 7 | 6 8 2 |
| 8 6 1 |*23 *23 5 | 4 7 9 |
| 9 2 7 | 6 8 4 | 5 3 1 |
+------------------------+------------------------+------------------------+
| 3 89 289 | 248 5 1 | 7 6 48 |
| 27 78 6 |*23+48 *23+4 9 | 1 a =24 5 |
| 5 1 4 | 7 6 b =28 | 39 29 38 |
+------------------------+------------------------+------------------------+ |
|
|
| Back to top |
|
|
cgordon
Joined: 04 May 2007
Posts: 769
Location: ontario, canada
|
| Posted: Sun Sep 07, 2008 12:28 am Post subject: |
|
|
| Quote: | | Another one step is the 23 UR* in R58C45, |
I mentioned that above damnit |
|
| Back to top |
|
|
Asellus
Joined: 05 Jun 2007
Posts: 865
Location: Sonoma County, CA, USA
|
| Posted: Sun Sep 07, 2008 6:31 am Post subject: |
|
|
Hmmm... just looking over this thread and have some comments:
ravel's "finned" M-Wing is really just an "M-Wing". The non-pincer bivalue cell of an M-Wing (as discussed a few times before) does not actually need to be a bivalue cell because it serves a weak link function. Even if that 248 cell were 124689 (not likely, of course), the M-Wing elimination of the <2> would work the same and there is no need to think of it as having a 1689 fin! In Eureka:
(2=4)r8c8 - (4)r7c9=(4-2)r7c4=(2)r7c3
The bit in red shows that the 24 link in r7c4 is weak and so any other digits may share the cell. Of course, this is harder to spot because you don't see the matching pair of bivalue cells. (Someone ... nataraj? ... came up with another name for these things that I've forgotten since one name is all I prefer to remember.) Maybe if one looks for "fin" digits in the otherwise bivalue cell that is not the pincer while remembering that it's not really a fin, that can help spot the things.
(Okay... that last sentence definitely sounds funny! But, it's a serious idea.)
Johan did a nice job of exploiting the grouped strong link
(4)r8c45=(8)r8c4
induced by the 23 UR. In this case, as in a couple of others recently, this strong link functions as a 48 bivalue "pseudo-cell" to complete the 248 XY-Wing for the <2> elimination. It's a good demonstration of the power of thinking about those induced strong links rather than just following the "Type #" approach. |
|
| Back to top |
|
|
ravel
Joined: 21 Apr 2006
Posts: 536
|
| Posted: Sun Sep 07, 2008 10:49 am Post subject: |
|
|
| Asellus wrote: | | ravel's "finned" M-Wing is really just an "M-Wing". |
You are very right, its not the same as for almost w-wings, where i had to look at the fins also.
So this time i also missed Johan's nice DP solution, because i thought, i could not do more than the type 4 eliminations. |
|
| Back to top |
|
|
daj95376
Joined: 23 Aug 2008
Posts: 3854
|
| Posted: Sun Sep 07, 2008 8:44 pm Post subject: |
|
|
| Asellus wrote: | Johan did a nice job of exploiting the grouped strong link
(4)r8c45=(8)r8c4
induced by the 23 UR. In this case, as in a couple of others recently, this strong link functions as a 48 bivalue "pseudo-cell" to complete the 248 XY-Wing for the <2> elimination. It's a good demonstration of the power of thinking about those induced strong links rather than just following the "Type #" approach. |
I think it's time to give this puppy a name instead of describing it repeatedly. How about UR-Wing? Not only for the XY-Wing variant, but in case a more complex variant occurs sometime. |
|
| Back to top |
|
|
Asellus
Joined: 05 Jun 2007
Posts: 865
Location: Sonoma County, CA, USA
|
| Posted: Sun Sep 07, 2008 9:33 pm Post subject: |
|
|
| daj95376 wrote: | | I think it's time to give this puppy a name instead of describing it repeatedly. How about UR-Wing? Not only for the XY-Wing variant, but in case a more complex variant occurs sometime. |
Actually, I'm sort of surprised that 3 or 4 of these "UR XY-Wings" have come up in quick succession lately since I hadn't thought them very common. Perhaps if we name them, we'll stop finding them!
Regardless of a name, the essential idea is of a UR-induced strong inference between two different digits (grouped or not) that substitutes for a bivalue cell in some sort of wing. However there's a caveat: this is a strong only link and so cannot replace a bivalue cell that is serving a weak inference role. The only such case I can think of in commonly named wings discussed on this board is the one mentioned above: the non-pincer bivalue cell of an M-Wing. A UR induced link cannot be used in place of that particular bivalue cell.
In my experience, UR-induced strong links usually require AIC chains that are not so directly analogous to standard wings. So, I'm not sure how generally useful such a name would be. However, the concept is quite useful for those comfortable with AICs more generally. |
|
| Back to top |
|
|
keith
Joined: 19 Sep 2005
Posts: 3355
Location: near Detroit, Michigan, USA
|
| Posted: Sun Sep 07, 2008 10:04 pm Post subject: |
|
|
My own opinion, for what it's worth, is that the extra digits on the UR are a "pseudo-cell" that can often replace a regular cell in the logic of a number of named patterns.
Consider the following chain: <12>-<23>-<34>-<14>. Any cell that sees both ends cannot be <1>. We can, for example, regard the center cells as a pseudo-cell <24>, and we have, in my bad terminology, an "extended" XY-wing, <12>-<24>-<14>.
I have found this idea quite useful in looking for short chains. Pseudo-cells are everywhere, not just on a side of a UR.
Keith |
|
| Back to top |
|
|
|
FAQ
Search
Memberlist
Usergroups
Register
Profile
Log in to check your private messages
Log in
