| View previous topic :: View next topic |
| Author |
Message |
daj95376
Joined: 23 Aug 2008
Posts: 3854
|
 Posted: Tue Jan 20, 2009 2:10 am Post subject: Set Keith_1 Puzzle 2 |
 |
|
My first attempt to purposely generate VH+AE puzzles (as suggested by Keith) has produced some interesting failures. Here's one.
| Code: | +-----------------------+
| . 9 7 | . 4 . | 5 . 1 |
| 6 . . | 8 1 . | . . 3 |
| 8 . . | 5 3 . | 6 . . |
|-------+-------+-------|
| . 4 2 | 9 . . | 1 . . |
| 5 1 3 | . 8 . | 4 . . |
| . . . | . . . | . . . |
|-------+-------+-------|
| 4 . 9 | 1 2 . | 3 . . |
| . . . | . . . | . . . |
| 1 7 . | . . . | . . 4 |
+-----------------------+
|
Play this puzzle online at the Daily Sudoku site
===== ===== ===== ===== ===== ===== ===== ===== ===== ===== ===== ===== =====
Can be solved using steps from ...
Basics: Naked/Hidden Single, Naked Pair/Triple, Locked Candidate 1/2
Basics+: Naked Quad, Hidden Pair/Triple/Quad
VH: BUG+1, Remote Pair, UR Type 1, X-Wing, XY-Wing
Advanced: 2-String Kite, Empty Rectangle, Skyscraper, UR Type 2, XYZ-Wing
Extreme: finned X-Wing, Multiple Colors, M-Wing, W-Wing, XY-Chain
VH+AE: multi-step VH plus one Advanced and/or one Extreme step.
Extreme+: Swordfish, Jellyfish, (but mostly) Chain |
|
| Back to top |
|
 |
keith
Joined: 19 Sep 2005
Posts: 3355
Location: near Detroit, Michigan, USA
|
| Posted: Tue Jan 20, 2009 4:24 am Post subject: Re: Set Keith_1 Puzzle 2 |
|
|
| daj95376 wrote: | My first attempt to purposely generate VH+AE puzzles (as suggested by Keith) has produced some interesting failures. Here's one.
|
OMG! They're naming puzzles for me.
At first glance, this can be solved with two of my four-cell extended XY-wings. It is not a "failure".
I will only be able to kick back with pencil and paper and a cup of coffee to really figure this one out later on Tuesday.
Keith |
|
| Back to top |
|
|
Marty R.
Joined: 12 Feb 2006
Posts: 5770
Location: Rochester, NY, USA
|
| Posted: Tue Jan 20, 2009 5:14 am Post subject: |
|
|
Either I made a lucky error or else this didn't require a VH + AE, but it was a one -stepper.
|
|
| Back to top |
|
|
keith
Joined: 19 Sep 2005
Posts: 3355
Location: near Detroit, Michigan, USA
|
| Posted: Tue Jan 20, 2009 11:23 pm Post subject: |
|
|
After basics: | Code: | +-------------+-------------+-------------+
| 3 9 7 | 2 4 6 | 5 8 1 |
| 6 5 4 | 8 1 79 | 29 279 3 |
| 8 2 1 | 5 3 79 | 6 4 79 |
+-------------+-------------+-------------+
| 7 4 2 | 9 6 3 | 1 5 8 |
| 5 1 3 | 7 8 2 | 4 69 69a |
| 9 68 68 | 4 5 1 | 7 3 2 |
+-------------+-------------+-------------+
| 4 68d 9 | 1 2 58 | 3 67 5-67 |
| 2 3 58c | 6 7 4 | 89 1 59b |
| 1 7 568 | 3 9 58 | 28 26 4 |
+-------------+-------------+-------------+ |
Marty is correct: A W-wing takes out <5> in R9C3, solving the puzzle.
Also, abcd is a 4-cell chain with pincers <6>. This leads to another 4-cell chain that takes out <5> in R8C9. Or, it leads to a BUG+1.
Keith |
|
| Back to top |
|
|
daj95376
Joined: 23 Aug 2008
Posts: 3854
|
| Posted: Wed Jan 21, 2009 12:32 am Post subject: Re: Set Keith_1 Puzzle 2 |
|
|
| daj95376 wrote: | My first attempt to purposely generate VH+AE puzzles (as suggested by Keith) has produced some interesting failures. Here's one.
|
After basics, there are:
2x 5-cell XY-Chain that crack the puzzle
2x W-Wing that crack the puzzle
2x M-Wing that crack the puzzle
It also appears to be a BUG+3.
I thought it was interesting that so few remaining candidates were this difficult to resolve. |
|
| Back to top |
|
|
tlanglet
Joined: 17 Oct 2007
Posts: 2468
Location: Northern California Foothills
|
| Posted: Wed Jan 21, 2009 1:44 pm Post subject: |
|
|
Another one-step solution is a "funny" version of Keiths' 4-cell extended xy-wing.
| Code: |
+-------------+-------------+-------------+
| 3 9 7 | 2 4 6 | 5 8 1 |
| 6 5 4 | 8 1 79 | 29 279 3 |
| 8 2 1 | 5 3 79 | 6 4 79 |
+-------------+-------------+-------------+
| 7 4 2 | 9 6 3 | 1 5 8 |
| 5 1 3 | 7 8 2 | 4 69 69a |
| 9 68 68 | 4 5 1 | 7 3 2 |
+-------------+-------------+-------------+
| 4 68a 9 | 1 2 58 | 3 67d 567c|
| 2 3 58b | 6 7 4 | 89 1 -59e|
| 1 7 568 | 3 9 58 | 28 26 4 |
+-------------+-------------+-------------+ |
Looking at the xy-wing a, b, and pseudo-cell c-d,
If pivot a = <8>, then b = <5>
If pivot a = <6>, then c-d = <5>
This deletes the <5> at e to solve the puzzle
Ted |
|
| Back to top |
|
|
cgordon
Joined: 04 May 2007
Posts: 769
Location: ontario, canada
|
| Posted: Wed Jan 21, 2009 7:05 pm Post subject: |
|
|
| Quote: | | Marty is correct: A W-wing takes out <5> in R9C3, solving the puzzle. |
I am an expert on w-wings (have been for a fortnight now). I agree with Marty that there is an w-wing on <58> but surely this removes the <8> from R7C2. |
|
| Back to top |
|
|
daj95376
Joined: 23 Aug 2008
Posts: 3854
|
| Posted: Wed Jan 21, 2009 9:47 pm Post subject: |
|
|
This puzzle is a good example for why short explanations don't always get the job done. Ignore the abbreviated notation from my solver and just examine the results. All of these crack the puzzle open after basics.
| Code: | -5r8c3 8r8c3 9r8c7 2r2c7 8r9c7 5r9c6 [XY-Chain] <> 5 [r9c3]
-6r7c2 8r7c2 5r7c6 8r9c6 2r9c7 6r9c8 [XY-Chain] <> 6 [r7c89],[r9c3]
-5r8c3 8r8c3 8r7c6 5r7c9 [M-Wing ] <> 5 [r8c9]
-8r9c6 5r9c6 5r8c3 8r8c7 [M-Wing ] <> 8 [r9c7]
-5r8c3 8r8c3 8r7c6 5r9c6 [W-Wing ] <> 5 [r9c3]
-8r7c6 5r7c6 5r8c9 8r8c3 [W-Wing ] <> 8 [r7c2]
|
|
|
| Back to top |
|
|
wapati
Joined: 10 Jun 2008
Posts: 472
Location: Brampton, Ontario, Canada.
|
| Posted: Wed Jan 21, 2009 10:01 pm Post subject: |
|
|
| cgordon wrote: | | Quote: | | Marty is correct: A W-wing takes out <5> in R9C3, solving the puzzle. |
I am an expert on w-wings (have been for a fortnight now). I agree with Marty that there is an w-wing on <58> but surely this removes the <8> from R7C2. |
I have tried to agree. I cannot. What is your W-wing? |
|
| Back to top |
|
|
cgordon
Joined: 04 May 2007
Posts: 769
Location: ontario, canada
|
| Posted: Wed Jan 21, 2009 10:19 pm Post subject: |
|
|
| Quote: | | I have tried to agree. I cannot. What is your W-wing? |
Cut me some slack here because I am only a recent expert. Looking at Ted's grid there are only two <5>s in C9. So there has to be an <8> in one of the <58> pairs in R78. So any <8> that sees both pairs has to go. Isn't that a w-wing - if not I want it called a CG wing!
| Code: |
+-------+-------+--------+
| . . . | . . . | . . . |
| . . . | . . . | . . . |
| . . . | . . . | . . . |
+-------+-------+--------+
| . . . | . . . | . . . |
| . . . | . . . | . . . |
| . . . | . . . | . . . |
+-------+-------+--------+
| .68 . | . . 58| . . 567|
| . . 58| . . . | . . 59 |
| . . . | . . . | . . . |
+-------+-------+--------+
|
|
|
| Back to top |
|
|
|
FAQ
Search
Memberlist
Usergroups
Register
Profile
Log in to check your private messages
Log in
