| View previous topic :: View next topic |
| Author |
Message |
ravel
Joined: 21 Apr 2006
Posts: 536
|
 Posted: Mon Aug 18, 2008 2:32 pm Post subject: WALL-E by JPF |
 |
|
The puzzle was inspired by this poster.
| Code: | +-------------------+
| . . . . . . . . . |
| . . 1 . . . 2 . . |
| . . . 3 4 5 . . . |
| . . . . 6 . . . . |
| . . 6 7 2 3 8 . . |
| . 9 . . . . . 5 . |
| 4 . . . . . . . 3 |
| 2 . . . . . . . 6 |
| 3 7 . . . . . 9 5 |
+-------------------+ JPF
|
>>> play online
Half of the puzzle is quickly solved, but then ?
I needed to explore two (i.e. all) UR's. |
|
| Back to top |
|
 |
arkietech
Joined: 31 Jul 2008
Posts: 1834
Location: Northwest Arkansas USA
|
| Posted: Mon Aug 18, 2008 8:01 pm Post subject: |
|
|
I solved it with a m-wing and ur 
| Code: |
*-----------------------------------------------------------*
| 689 348 *47 | 1 79 2 | 5 3467 *478 |
| 59 345 1 | 6 79 8 | 2 347 *47 |
| 68 28 #27 | 3 4 5 | 9 16-7 18-7 |
|-------------------+-------------------+-------------------|
| 18 248 24 | 5 6 9 | 3 147 147 |
| 15 45 6 | 7 2 3 | 8 14 9 |
| 7 9 3 | 4 8 1 | 6 5 2 |
|-------------------+-------------------+-------------------|
| 4 6 9 | 8 5 7 | 1 2 3 |
| 2 1 5 | 9 3 4 | 7 8 6 |
| 3 7 8 | 2 1 6 | 4 9 5 |
*-----------------------------------------------------------*
m-wing
*--------------------------------------------------*
| 689 38 4 | 1 79 2 | 5 367 78 |
| 59 35 1 | 6 79 8 | 2 *347 *4-7 |
| 68 2 7 | 3 4 5 | 9 16 18 |
|----------------+----------------+----------------|
| 18 48 2 | 5 6 9 | 3 *4-7 *147 |
| 15 45 6 | 7 2 3 | 8 14 9 |
| 7 9 3 | 4 8 1 | 6 5 2 |
|----------------+----------------+----------------|
| 4 6 9 | 8 5 7 | 1 2 3 |
| 2 1 5 | 9 3 4 | 7 8 6 |
| 3 7 8 | 2 1 6 | 4 9 5 |
*--------------------------------------------------*
ur(47)
singles
|
Nice puzzle |
|
| Back to top |
|
|
Kdelle
Joined: 20 Mar 2008
Posts: 59
Location: Hudson, NH
|
| Posted: Mon Aug 18, 2008 8:39 pm Post subject: |
|
|
| arkietech wrote: | I solved it with a m-wing and ur
| Code: |
*-----------------------------------------------------------*
| 689 348 *47 | 1 79 2 | 5 3467 *478 |
| 59 345 1 | 6 79 8 | 2 347 *47 |
| 68 28 #27 | 3 4 5 | 9 16-7 18-7 |
|-------------------+-------------------+-------------------|
| 18 248 24 | 5 6 9 | 3 147 147 |
| 15 45 6 | 7 2 3 | 8 14 9 |
| 7 9 3 | 4 8 1 | 6 5 2 |
|-------------------+-------------------+-------------------|
| 4 6 9 | 8 5 7 | 1 2 3 |
| 2 1 5 | 9 3 4 | 7 8 6 |
| 3 7 8 | 2 1 6 | 4 9 5 |
*-----------------------------------------------------------*
m-wing
*--------------------------------------------------*
| 689 38 4 | 1 79 2 | 5 367 78 |
| 59 35 1 | 6 79 8 | 2 *347 *4-7 |
| 68 2 7 | 3 4 5 | 9 16 18 |
|----------------+----------------+----------------|
| 18 48 2 | 5 6 9 | 3 *4-7 *147 |
| 15 45 6 | 7 2 3 | 8 14 9 |
| 7 9 3 | 4 8 1 | 6 5 2 |
|----------------+----------------+----------------|
| 4 6 9 | 8 5 7 | 1 2 3 |
| 2 1 5 | 9 3 4 | 7 8 6 |
| 3 7 8 | 2 1 6 | 4 9 5 |
*--------------------------------------------------*
ur(47)
singles
|
Nice puzzle |
Dan, I'm not sure I see the M Wing. When I got to that point , I used a W Wing 47 connected by the 79s. That led to the elimination of the 4 in r2c2, which ultimately led to the elimination of the 4s in r1c89.
I then needed ravel's hint on the 2 URs. Took a while, but I got it!
Could you explain your M wing, please?
Thanks.
Kathy |
|
| Back to top |
|
|
arkietech
Joined: 31 Jul 2008
Posts: 1834
Location: Northwest Arkansas USA
|
| Posted: Mon Aug 18, 2008 9:24 pm Post subject: |
|
|
Kathy asked: | Quote: |
Could you explain your M wing, please?
|
oops 
As I looked to explain my m-wing I saw that the 4 in in column 9 was not strong (r12c9)-- there was a 4 in r4c9 that I did not see....blowing my m-wing to pieces. Thanks for keeping me straight. Don' know what I was thinking.
Back to the drawing board. |
|
| Back to top |
|
|
Marty R.
Joined: 12 Feb 2006
Posts: 5770
Location: Rochester, NY, USA
|
| Posted: Mon Aug 18, 2008 11:25 pm Post subject: |
|
|
Certainly an interesting pattern of clues.
I started with a W-Wing on 47. Then one which I'm not overly proud of, a Finned XY-Wing on 35-34-45 which solved r4c3. That allowed an XY-Chain which exposed an almost Type 6 UR on 47. Both deadly candidates had strong links, removing both from one cell which finished it off. |
|
| Back to top |
|
|
storm_norm
Joined: 18 Oct 2007
Posts: 1741
|
| Posted: Tue Aug 19, 2008 12:48 am Post subject: |
|
|
| Code: | .------------------.------------------.------------------.
| 689 348 47 | 1 79 2 | 5 3467 478 |
| 59 345 1 | 6 79 8 | 2 347 47 |
| 68 28 27 | 3 4 5 | 9 167 178 |
:------------------+------------------+------------------:
| 18 248 24 | 5 6 9 | 3 147 147 |
| 15 45 6 | 7 2 3 | 8 14 9 |
| 7 9 3 | 4 8 1 | 6 5 2 |
:------------------+------------------+------------------:
| 4 6 9 | 8 5 7 | 1 2 3 |
| 2 1 5 | 9 3 4 | 7 8 6 |
| 3 7 8 | 2 1 6 | 4 9 5 |
'------------------'------------------'------------------' |
w-wing {4,7}
then xy-chain (make that 3 xy-chains total) with a xy-wing thrown in there |
|
| Back to top |
|
|
nataraj
Joined: 03 Aug 2007
Posts: 1048
Location: near Vienna, Austria
|
| Posted: Tue Aug 19, 2008 1:02 am Post subject: |
|
|
After basics, I used the w-wing (47) with strong link 7 in col 5. It got rid of the 4 in r1c8,r1c9,r2c2.
Now there is a possible DP 47 in r24c89:
| Code: |
+--------------------------+--------------------------+--------------------------+
| 689 348 47 | 1 79 2 | 5 367 78 |
| 59 35 1 | 6 79 8 | 2 347 47 |
| 68 28 27 | 3 4 5 | 9 167 178 |
+--------------------------+--------------------------+--------------------------+
| 18 248 24 | 5 6 9 | 3 147 147 |
| 15 45 6 | 7 2 3 | 8 14 9 |
| 7 9 3 | 4 8 1 | 6 5 2 |
+--------------------------+--------------------------+--------------------------+
| 4 6 9 | 8 5 7 | 1 2 3 |
| 2 1 5 | 9 3 4 | 7 8 6 |
| 3 7 8 | 2 1 6 | 4 9 5 |
+--------------------------+--------------------------+--------------------------+
|
... which means either r2c8=3 or r4c89=1.
Transport both ends and we get
either r5c1=5 or r2c2=5 and thus r2c1=9, which solves the puzzle. |
|
| Back to top |
|
|
Asellus
Joined: 05 Jun 2007
Posts: 865
Location: Sonoma County, CA, USA
|
| Posted: Tue Aug 19, 2008 1:39 am Post subject: |
|
|
nataraj,
I can't see how you get r5c1=5 from the DP.
Meanwhile, after that W-Wing, the 17 UR in r34c89 provides a useful elimination. r4c3 is <2>, or it is <4>. If <4>, then the 17 UR is Type 3 (68 locked pair in r3) and r3c2 is <2>. Thus, r3c3 and r4c2 are not <2>.
After that, a 5-cell XY Chain with ends at r2c5 and r1c9 eliminates <7>s at r1c5 and r2c89, solving the puzzle. |
|
| Back to top |
|
|
tlanglet
Joined: 17 Oct 2007
Posts: 2468
Location: Northern California Foothills
|
| Posted: Tue Aug 19, 2008 3:41 pm Post subject: |
|
|
| Asellus wrote: |
Meanwhile, after that W-Wing, the 17 UR in r34c89 provides a useful elimination. r4c3 is <2>, or it is <4>. If <4>, then the 17 UR is Type 3 (68 locked pair in r3) and r3c2 is <2>. Thus, r3c3 and r4c2 are not <2>.
|
Asellus, I substituted the "odd" candidates, (6,8 & 4) into the UR and all forced r4c3=2. However I was stuck at that point. I need to work on finding chains.
Ted |
|
| Back to top |
|
|
Asellus
Joined: 05 Jun 2007
Posts: 865
Location: Sonoma County, CA, USA
|
| Posted: Tue Aug 19, 2008 5:22 pm Post subject: |
|
|
Ted,
I don't care for the T&E aspect of "trying" all the extra digits of a UR. So, I look for exploitable structures. Though I described it as a Type 3 UR induced by the start of an XY chain, that isn't how I actually spotted it. (It was just easier to describe that way.) I first noted the strong link induced by the UR between the grouped <4>s in r4c89 and the grouped {68} in r3c89, or (4)r4c89=({68})r3c89. With a 24 bivalue in r4 and 68 and 28 bivalues (or a 268 ALS) in r3, the AIC was immediately obvious to me:
(2=4)r4c3 - UR[(4)r4c89=({68})r3c89] - ALS[({68})r3c12=(2)r3c2]
These AICs for me are not T&E but are about spotting certain useful structure patterns in the grid. It is a skill that is acquired somewhat gradually. One must get comfortable with the notions of strong and weak inferences and learn how to spot these in the various sorts of structures that occur in the grid. At the beginning, this takes time and reflection on the logic involved. After a while, the logic becomes second nature and the links much more obvious. But for some, this is just too much. Medusa multi-coloring, where you take time to trace the links of any elimination to discover the underlying AIC, is one way to get started. Also, learning to add a multi-cell ALS to an XY chain is another helpful starting place. At some point, UR-induced strong links (often simpler than the one above) get added to the arsenal.
As for finding XY Chains, it is not totally arbitrary. In this case, <7> was one of the few candidates still well distributed in the grid AND also involved in a number of bivalues. It was natural to check for an XY chain with <7> pincers. |
|
| Back to top |
|
|
tlanglet
Joined: 17 Oct 2007
Posts: 2468
Location: Northern California Foothills
|
| Posted: Tue Aug 19, 2008 6:14 pm Post subject: |
|
|
| Asellus wrote: | Ted,
I don't care for the T&E aspect of "trying" all the extra digits of a UR. So, I look for exploitable structures. Though I described it as a Type 3 UR induced by the start of an XY chain, that isn't how I actually spotted it. (It was just easier to describe that way.) I first noted the strong link induced by the UR between the grouped <4>s in r4c89 and the grouped {68} in r3c89, or (4)r4c89=({68})r3c89. With a 24 bivalue in r4 and 68 and 28 bivalues (or a 268 ALS) in r3, the AIC was immediately obvious to me:
(2=4)r4c3 - UR[(4)r4c89=({68})r3c89] - ALS[({68})r3c12=(2)r3c2]
These AICs for me are not T&E but are about spotting certain useful structure patterns in the grid. It is a skill that is acquired somewhat gradually. One must get comfortable with the notions of strong and weak inferences and learn how to spot these in the various sorts of structures that occur in the grid. At the beginning, this takes time and reflection on the logic involved. After a while, the logic becomes second nature and the links much more obvious. But for some, this is just too much. Medusa multi-coloring, where you take time to trace the links of any elimination to discover the underlying AIC, is one way to get started. Also, learning to add a multi-cell ALS to an XY chain is another helpful starting place. At some point, UR-induced strong links (often simpler than the one above) get added to the arsenal.
|
Yes, I agree that trying all candidates of a UR has an aspect of T & E, but it is focused on the extra digits in specific cells. My present "Sudoku Life" level of knowledge allows me to solve most VH puzzles on this site but I struggle on more difficult ones and will use any "insights" to help me find a solution. With more effort and lots of luck, things will hopefully improve.
Thanks for your comments and encouragement.
Ted |
|
| Back to top |
|
|
nataraj
Joined: 03 Aug 2007
Posts: 1048
Location: near Vienna, Austria
|
| Posted: Tue Aug 19, 2008 11:57 pm Post subject: |
|
|
| Asellus wrote: | nataraj,
I can't see how you get r5c1=5 from the DP. |
I don't blame you, that was a terrible blunder . Thx for pointing it out. From r4c89=1 we get r5c1=1 (which is of no help) and not r5c5=5. I must still have been suffering from jet lag or whatever... |
|
| 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
|
FAQ
Search
Memberlist
Usergroups
Register
Profile
Log in to check your private messages
Log in
