| View previous topic :: View next topic |
| Author |
Message |
Asellus
Joined: 05 Jun 2007
Posts: 865
Location: Sonoma County, CA, USA
|
 Posted: Fri Oct 24, 2008 10:02 pm Post subject: LA Times / Free Press - October 24, 2008 |
 |
|
| Code: | +-------+-------+-------+
| . . 2 | . 3 . | 7 . . |
| 5 . . | . 2 . | . . . |
| . . . | 5 . 6 | . . 9 |
+-------+-------+-------+
| . 9 . | 3 . . | 2 . . |
| 8 . 6 | . . . | . . 5 |
| . . 4 | . . 7 | . 6 . |
+-------+-------+-------+
| 7 . 8 | 6 . 2 | . . . |
| . . . | . 4 . | . . 7 |
| . . 3 | . 5 . | 6 . . |
+-------+-------+-------+ |
A one-stepper for me. |
|
| Back to top |
|
 |
tlanglet
Joined: 17 Oct 2007
Posts: 2468
Location: Northern California Foothills
|
| Posted: Sat Oct 25, 2008 1:13 am Post subject: |
|
|
My first elimination was also a one-stepper as follows........ | Quote: | | skyscraper on <8>, r2c67 |
Ted |
|
| Back to top |
|
|
storm_norm
Joined: 18 Oct 2007
Posts: 1741
|
| Posted: Sat Oct 25, 2008 7:53 am Post subject: |
|
|
| Code: | .------------------.------------------.------------------.
| 46 68 2 | 1489 3 1489 | 7 5 18 |
| 5 7 9 | 148 2 148 | 1348 138 6 |
| 34 38 1 | 5 7 6 | 48 2 9 |
:------------------+------------------+------------------:
| 1 9 7 | 3 6 5 | 2 48 48 |
| 8 23 6 | 1249 19 149 | 139 7 5 |
| 23 5 4 | 129 8 7 | 139 6 13 |
:------------------+------------------+------------------:
| 7 14 8 | 6 19 2 | 5 1349 134 |
| 26 26 5 | 189 4 3 | 18 189 7 |
| 9 14 3 | 7 5 18 | 6 148 2 |
'------------------'------------------'------------------' |
(8=4)r4c8-(4)r4c9=(4-3)r7c9=(3-1)r6c9=(1)r56c7-(1=8)r8c7; r89c8 <> 8
the chain eliminates the 8's in r89c8 for my one stepper. |
|
| Back to top |
|
|
Asellus
Joined: 05 Jun 2007
Posts: 865
Location: Sonoma County, CA, USA
|
| Posted: Sat Oct 25, 2008 9:17 am Post subject: |
|
|
Norm,
I believe you may have found the "anti-W-Wing" to my 18 W-Wing in c79 (that removes <8> from r23c7). Hah!
I don't see Ted's Skyscraper, though. |
|
| Back to top |
|
|
tlanglet
Joined: 17 Oct 2007
Posts: 2468
Location: Northern California Foothills
|
| Posted: Sat Oct 25, 2008 1:43 pm Post subject: |
|
|
| Asellus wrote: |
I don't see Ted's Skyscraper, though. |
Well, I don't see it either in the code posted by storm_norm 
I obviously made a mistake doing basics, because the skyscraper is "tall and proud" in my code after basics. I just did the puzzle again and got the correct outcome, but have no idea how I arrived at my initial result.
My turn to be 
Ted |
|
| Back to top |
|
|
keith
Joined: 19 Sep 2005
Posts: 3355
Location: near Detroit, Michigan, USA
|
| Posted: Sat Oct 25, 2008 3:04 pm Post subject: |
|
|
| Asellus wrote: | | ... my 18 W-Wing in c79 (that removes <8> from r23c7). |
I don't see it. My grid after basics is the same as that posted above by Norm. All I see so far is an XYZ-wing.
edit: There is an extended (pseudo-cell) XY-wing that takes out <3> in R6C1. After that, coloring on <8> finishes it.
Keith |
|
| Back to top |
|
|
Marty R.
Joined: 12 Feb 2006
Posts: 5770
Location: Rochester, NY, USA
|
| Posted: Sat Oct 25, 2008 4:08 pm Post subject: |
|
|
| keith wrote: | | Asellus wrote: | | ... my 18 W-Wing in c79 (that removes <8> from r23c7). |
I don't see it. My grid after basics is the same as that posted above by Norm. All I see so far is an XYZ-wing.
Keith |
Are they connected by the 1s in box 6? |
|
| Back to top |
|
|
daj95376
Joined: 23 Aug 2008
Posts: 3854
|
| Posted: Sat Oct 25, 2008 4:39 pm Post subject: |
|
|
| Marty R. wrote: | | keith wrote: | | Asellus wrote: | | ... my 18 W-Wing in c79 (that removes <8> from r23c7). |
I don't see it. |
Are they connected by the 1s in box 6? |
Possibly ... or the 1s in column 8. Either way, it's an extension to the W-Wing definition using a grouped strong link. |
|
| Back to top |
|
|
keith
Joined: 19 Sep 2005
Posts: 3355
Location: near Detroit, Michigan, USA
|
| Posted: Sat Oct 25, 2008 5:03 pm Post subject: |
|
|
Grouped coloring. I see it now!  |
|
| Back to top |
|
|
Marty R.
Joined: 12 Feb 2006
Posts: 5770
Location: Rochester, NY, USA
|
| Posted: Sat Oct 25, 2008 5:11 pm Post subject: |
|
|
| Can someone provide a definition of "grouped strong link" or "grouped coloring"? |
|
| Back to top |
|
|
daj95376
Joined: 23 Aug 2008
Posts: 3854
|
| Posted: Sat Oct 25, 2008 6:54 pm Post subject: |
|
|
[resubmitted]
A grouped strong link exists when all candidate cells in a house/unit are partitioned into two subsets.
A traditional 2-cell strong link is a grouped strong link where each partition contains only one cell.
Only some grouped strong link patterns are useful. Here are examples that come readily to mind. The example in [b9] should look familiar to you.
| Code: | +-----------------------+
| . . . | . . . | . . . | traditional 2-cell strong link
| / X / | / / / | / X / | [r2c2]=X=[r2c8]
| . . . | . . . | . . . |
|-------+-------+-------|
| . . . | . . . | . . . | grouped strong link in a row
| / X / | / / / | X X X | [r5c2]=X=[r5c789]
| . . . | . . . | . . . |
|-------+-------+-------|
| . . . | / X / | / / / | grouped strong link in a box
| . . . | X / X | Y / / | [r8c46]=X=[r79c5]
| . . . | / X / | Y / Y | [r9c9]=Y=[r89c7]
+-----------------------+
|
This pattern remains after a mutant Swordfish r2b79\r8c28 performed eliminations [r13456c28]<>X, [r8c24568]<>X:
| Code: | strong link in [r2]
grouped strong links in [r8], [c2], [c8], [b7], and [b9]
+-----------------------+
| . / . | . . . | . / . |
| / X / | / / / | / X / |
| . / . | . . . | . / . |
|-------+-------+-------|
| . / . | . . . | . / . |
| . / . | . . . | . / . |
| . / . | . . . | . / . |
|-------+-------+-------|
| / X / | . . . | / X / |
| X / X | / / / | X / X |
| / X / | . . . | / X / |
+-----------------------+
|
Last edited by daj95376 on Sat Oct 25, 2008 8:35 pm; edited 6 times in total |
|
| Back to top |
|
|
storm_norm
Joined: 18 Oct 2007
Posts: 1741
|
| Posted: Sat Oct 25, 2008 6:55 pm Post subject: |
|
|
| Quote: | | Can someone provide a definition of "grouped strong link" or "grouped coloring"? |
in my image, you will notice that the green circle contains both 1's in column 7. that is a grouped link because both of them essentially act like a single candidate for the purposes of that particular chain.
the 1's basically act like a single 1.
so you can see Asellus's w-wing the same way.
| Code: | .------------------.------------------.------------------.
| 46 68 2 | 1489 3 1489 | 7 5 #18 |
| 5 7 9 | 148 2 148 |134-8 138 6 |
| 34 38 1 | 5 7 6 |4-8 2 9 |
:------------------+------------------+------------------:
| 1 9 7 | 3 6 5 | 2 48 48 |
| 8 23 6 | 1249 19 149 |*139 7 5 |
| 23 5 4 | 129 8 7 |*139 6 *13 |
:------------------+------------------+------------------:
| 7 14 8 | 6 19 2 | 5 1349 134 |
| 26 26 5 | 189 4 3 |#18 189 7 |
| 9 14 3 | 7 5 18 | 6 148 2 |
'------------------'------------------'------------------' |
the 1's in column 7 act like one digit because if both are true, then the 1 in r7c9 is false and vice versa. |
|
| Back to top |
|
|
arkietech
Joined: 31 Jul 2008
Posts: 1834
Location: Northwest Arkansas USA
|
| Posted: Sat Oct 25, 2008 8:53 pm Post subject: |
|
|
| storm_norm wrote: | | ...in my image... |
What software did you use to create your drawing or img? |
|
| Back to top |
|
|
Marty R.
Joined: 12 Feb 2006
Posts: 5770
Location: Rochester, NY, USA
|
| Posted: Sat Oct 25, 2008 9:17 pm Post subject: |
|
|
| Thank you both. It turns out I've used grouped strong links on a number of occasions without knowing what they were called or that they even had a name. |
|
| Back to top |
|
|
Asellus
Joined: 05 Jun 2007
Posts: 865
Location: Sonoma County, CA, USA
|
| Posted: Sat Oct 25, 2008 10:30 pm Post subject: |
|
|
| daj95376 wrote: | | Either way, it's an extension to the W-Wing definition using a grouped strong link. |
If it is a matter of an "originalist" fidelity to the W-Wing definition, I haven't extended anything. Here is part of the post by TexCat from the thread on this board in which the term W-Wing ("Woods"-Wing) was first coined:
| TexCat wrote: | But the w-wing!!! Very exciting new technique! Let's see if I understand it....
| Code: |
. . . | NotG NotG NotG | . . .
GW . . | . . . | . . .
. . . | . . . | . . GW
|
If you have an GW pair in boxes either in the same column or the same row of boxes, and the row/column which is not part of the GW pair cannot contain G, then you can eliminate W from the cells that the GW pair have in common.
And so you end up with
| Code: |
. . . | NotG NotG NotG | . . .
GW . . | . . . |NotW NotW NotW
NotW NotW NotW| . . . | . . GW |
|
The 18 W-Wing above conforms exactly to this definition, which clearly allows for grouped strong links. The discussion of the nature of the external strong links involved in W-Wings evolved in the days and weeks immediately following this original post and examples were posted with all sorts of such links, including ER sequences and Wing-ER sequences, etc., few of which any longer conformed to TexCat's graphic of George Woods's original observation. I don't see how one can draw some sort of originalist definitional line-in-the-sand for the nature of the link given this history. |
|
| Back to top |
|
|
daj95376
Joined: 23 Aug 2008
Posts: 3854
|
| Posted: Sun Oct 26, 2008 12:39 am Post subject: |
|
|
It would appear that keith's (later) thread has W-Wing over-constrained.
Sudopedia has it over-constrained as well. |
|
| Back to top |
|
|
storm_norm
Joined: 18 Oct 2007
Posts: 1741
|
| Posted: Sun Oct 26, 2008 7:46 am Post subject: |
|
|
| arkietech wrote: | | storm_norm wrote: | | ...in my image... |
What software did you use to create your drawing or img? |
i sent you a PM with all the info |
|
| Back to top |
|
|
|
FAQ
Search
Memberlist
Usergroups
Register
Profile
Log in to check your private messages
Log in
