| View previous topic :: View next topic |
| Author |
Message |
nataraj
Joined: 03 Aug 2007
Posts: 1048
Location: near Vienna, Austria
|
 Posted: Sat Jan 19, 2008 2:19 pm Post subject: Jan 19 vh |
 |
|
A simple UR (56) makes r6c7=9
a skyscraper (3:-r1c2=r8c2-r8c5=r2c5-) removes 3 from r1c4, r2c3
2 xwings (4: c14, 6: c16)
and a nice BUG+1
| Code: |
+--------------------------+--------------------------+--------------------------+
| 7 36 39 | 68 2 5 | 1 89 4 |
| 5 2 49 | 1 38 34 | 7 89 6 |
| 46 1 8 | 49 7 69 | 2 3 5 |
+--------------------------+--------------------------+--------------------------+
| 8 9 2 | 5 6 7 | 4 1 3 |
| 1 7 5 | 39 4 39 | 6 2 8 |
| 3 4 6 | 2 1 8 | 9 5 7 |
+--------------------------+--------------------------+--------------------------+
| 46 5 13 | 34 9 236 | 8 7 12 |
| 2 36 7 | 68 38 1 | 5 4 9 |
| 9 8 14 | 7 5 24 | 3 6 12 |
+--------------------------+--------------------------+--------------------------+ |
makes r7c6=3 (the "odd one out")
unconventional. nice. |
|
| Back to top |
|
 |
nataraj
Joined: 03 Aug 2007
Posts: 1048
Location: near Vienna, Austria
|
| Posted: Sat Jan 19, 2008 2:33 pm Post subject: |
|
|
O.K., O.K.
there IS the conventional path.
After basics
| Code: |
+--------------------------+--------------------------+--------------------------+
| 7 36 39 | 3689 2 5 | 1 89 4 |
| 45 2 349 | 1 38 349 | 7 589 6 |
| 456 1 8 | 469 7 469 | 2 3 59 |
+--------------------------+--------------------------+--------------------------+
| 8 9 2 | 5 6 7 | 4 1 3 |
| 1 7 56 | 39 4 39 | 56 2 8 |
| 3 4 56 | 2 1 8 | 569 59 7 |
+--------------------------+--------------------------+--------------------------+
| 46 5 134 | 346 9 2346 | 8 7 12 |
| 2 36 7 | 368 38 1 | 59 4 59 |
| 9 8 14 | 7 5 24 | 3 6 12 |
+--------------------------+--------------------------+--------------------------+
|
The x-wing (6) c16 gets rid of 6 in r7c4,r3c4
Then later on
| Code: | +--------------------------+--------------------------+--------------------------+
| 7 36 39 | 68 2 5 | 1 89 4 |
| 45 2 49 | 1 38 349 | 7 589 6 |
| 456 1 8 | 49 7 469 | 2 3 59 |
+--------------------------+--------------------------+--------------------------+
| 8 9 2 | 5 6 7 | 4 1 3 |
| 1 7 56 | 39 4 39 | 56 2 8 |
| 3 4 56 | 2 1 8 | 569 59 7 |
+--------------------------+--------------------------+--------------------------+
| 46 5 134 | 34 9 2346 | 8 7 12 |
| 2 36 7 | 68 38 1 | 59 4 59 |
| 9 8 14 | 7 5 24 | 3 6 12 |
+--------------------------+--------------------------+--------------------------+ |
xy(z)-wings galore:
xy: 34-46-36 (pivot r7c1) => r8c5 <>3
xyz: {1 3 4} ; r7c1<>4
xyz: {3 4 9} ; r3c6<>9 |
|
| Back to top |
|
|
cgordon
Joined: 04 May 2007
Posts: 769
Location: ontario, canada
|
| Posted: Sat Jan 19, 2008 2:42 pm Post subject: |
|
|
I got the skyscraper. I had X-wings coming out my ears - for 4s, 6's and 9's in both columns and rows. I found the UR (56) leaving 79 in R6C7.
But I didn't reach any Bug+1 stage. (I had a Bug+2 - if there is such a thing). I finished with the xy-wing <346> with pivot R7C1.
(Also after learning about APE's from last VH - I couldn't find any here). |
|
| Back to top |
|
|
Johan
Joined: 25 Jun 2007
Posts: 206
Location: Bornem Belgium
|
| Posted: Sat Jan 19, 2008 3:23 pm Post subject: |
|
|
There is a contradiction in R7, collapsing the grid in one move.
When R7C6=6 => R8C45=38 => R7C4=4 => R8C2=6 => R7C1=4 => R7C6 ≠ 6, solving the puzzle.
| Code: |
+--------------------------+--------------------------+--------------------------+
| 7 36 39 | 3689 2 5 | 1 89 4 |
| 45 2 349 | 1 38 349 | 7 589 6 |
| 456 1 8 | 469 7 469 | 2 3 59 |
+--------------------------+--------------------------+--------------------------+
| 8 9 2 | 5 6 7 | 4 1 3 |
| 1 7 56 | 39 4 39 | 56 2 8 |
| 3 4 56 | 2 1 8 | 569 59 7 |
+--------------------------+--------------------------+--------------------------+
| 46 =4 5 134 | 346 =4 9 6= 2346 | 8 7 12 |
| 2 36 =6 7 | 3-[6]8 38 1 | 59 4 59 |
| 9 8 14 | 7 5 24 | 3 6 12 |
+--------------------------+--------------------------+--------------------------+ |
|
|
| Back to top |
|
|
tlanglet
Joined: 17 Oct 2007
Posts: 2468
Location: Northern California Foothills
|
| Posted: Sat Jan 19, 2008 4:24 pm Post subject: |
|
|
| Johan wrote: | There is a contradiction in R7, collapsing the grid in one move.
When R7C6=6 => R8C45=38 => R7C4=4 => R8C2=6 => R7C1=4 => R7C6 ≠ 6, solving the puzzle.
|
Johan, I follow the logic of the contradiction, but I fail to see how eliminating the 6 in r7c6 directly solves the puzzle.
Thanks .....
Ted |
|
| Back to top |
|
|
Marty R.
Joined: 12 Feb 2006
Posts: 5770
Location: Rochester, NY, USA
|
| Posted: Sat Jan 19, 2008 4:27 pm Post subject: |
|
|
I decided to use this for Medusa practice, after which an XY-Wing finished it off.
| Quote: | | (I had a Bug+2 - if there is such a thing) |
Craig, there certainly is. There's also a BUG+3. But the +3 is the highest I've ever seen discussed. |
|
| Back to top |
|
|
Johan
Joined: 25 Jun 2007
Posts: 206
Location: Bornem Belgium
|
| Posted: Sat Jan 19, 2008 4:33 pm Post subject: |
|
|
| Quote: | | Johan, I follow the logic of the contradiction, but I fail to see how eliminating the 6 in r7c6 directly solves the puzzle |
Ted,
When <6> is eliminated in R7C6, then <6> must be placed in R3C6,
which eliminates <6> in R3C1 |
|
| Back to top |
|
|
cgordon
Joined: 04 May 2007
Posts: 769
Location: ontario, canada
|
| Posted: Sat Jan 19, 2008 5:40 pm Post subject: |
|
|
Marty: This is as far as I got. Didn't reach Nataraj's Bug+1, so I assume I missed something obvious. (Actually- maybe I didn't - I shouldn't be so self-critical - maybe he did something he shouldn't have). But regardless!! Is there one of them Bug+2 s here that you alluded to. I can't find an explanation for these.
Cheers
| Code: |
+-----------+------------+---------+
| 7 36 39 | 68 2 5 | 1 89 4 |
| 5 2 49 | 1 38 34 | 7 89 6 |
| 46 1 8 | 49 7 69 | 2 3 5 |
+-----------+------------+---------+
| 8 9 2 | 5 6 7 | 4 1 3 |
| 1 7 5 | 39 4 39 | 6 2 8 |
| 3 4 6 | 2 1 8 | 9 5 7 |
+-----------+------------+---------+
| 46 5 134 | 34 9 236 | 8 7 12 |
| 2 36 7 | 368 38 1 | 5 4 9 |
| 9 8 14 | 7 5 24 | 3 6 12 |
+-----------+------------+---------+
|
|
|
| Back to top |
|
|
nataraj
Joined: 03 Aug 2007
Posts: 1048
Location: near Vienna, Austria
|
| Posted: Sat Jan 19, 2008 6:04 pm Post subject: |
|
|
Craig,
the x-wing (4) takes out 4 in r7c3,
the triple (349) in c4 removes 3 from r8c4
the rest is BUG+1
| Quote: | | maybe he did something he shouldn't have |
That is quite possible  |
|
| Back to top |
|
|
Marty R.
Joined: 12 Feb 2006
Posts: 5770
Location: Rochester, NY, USA
|
| Posted: Sat Jan 19, 2008 6:28 pm Post subject: |
|
|
| cgordon wrote: | Marty: This is as far as I got. Didn't reach Nataraj's Bug+1, so I assume I missed something obvious. (Actually- maybe I didn't - I shouldn't be so self-critical - maybe he did something he shouldn't have). But regardless!! Is there one of them Bug+2 s here that you alluded to. I can't find an explanation for these.
Cheers
| Code: |
+-----------+------------+---------+
| 7 36 39 | 68 2 5 | 1 89 4 |
| 5 2 49 | 1 38 34 | 7 89 6 |
| 46 1 8 | 49 7 69 | 2 3 5 |
+-----------+------------+---------+
| 8 9 2 | 5 6 7 | 4 1 3 |
| 1 7 5 | 39 4 39 | 6 2 8 |
| 3 4 6 | 2 1 8 | 9 5 7 |
+-----------+------------+---------+
| 46 5 134 | 34 9 236 | 8 7 12 |
| 2 36 7 | 368 38 1 | 5 4 9 |
| 9 8 14 | 7 5 24 | 3 6 12 |
+-----------+------------+---------+
|
|
Craig,
I believe there is a valid BUG+3 pattern here. R7c3 must = 4 or r7c6 must = 3 or r8c4 = 3 to break up the Deadly Pattern. I haven't tried it, but at this point you could then try a triple-implication chain with those three numbers. This, I believe, is the Deadly Pattern:
| Code: | +-----------+------------+---------+
| 7 36 39 | 68 2 5 | 1 89 4 |
| 5 2 49 | 1 38 34 | 7 89 6 |
| 46 1 8 | 49 7 69 | 2 3 5 |
+-----------+------------+---------+
| 8 9 2 | 5 6 7 | 4 1 3 |
| 1 7 5 | 39 4 39 | 6 2 8 |
| 3 4 6 | 2 1 8 | 9 5 7 |
+-----------+------------+---------+
| 46 5 13 | 34 9 26 | 8 7 12 |
| 2 36 7 | 68 38 1 | 5 4 9 |
| 9 8 14 | 7 5 24 | 3 6 12 |
+-----------+------------+---------+ |
|
|
| Back to top |
|
|
George Woods
Joined: 28 Mar 2006
Posts: 304
Location: Dorset UK
|
| Posted: Sat Jan 19, 2008 6:49 pm Post subject: bugs and XY wings |
|
|
| As I understand the Bug +1 logic ,, it says that wrong value in the 3candidate cell leaves nothing but doublets (going around in perpetuity - and no solution) and that just leaves one valid candidate there BUT what if the doublets have an XY wing included - as they clearly do here - is the BUG+1 logic valid? What is shown above as a "Deadly Pattern" is suceptible to the XY wing! |
|
| Back to top |
|
|
Clement
Joined: 24 Apr 2006
Posts: 1111
Location: Dar es Salaam Tanzania
|
| Posted: Sat Jan 19, 2008 7:03 pm Post subject: Daily Sudoku: Sat 19-Jan-2008 VH |
|
|
| Interesting. It begins with a UR of 5,6 in Blocks 4 and 6 leaving 9 in r6c7. This leads to two X-Wings of 4,s in columns c1&4 and 6,s in c1&6 respectively, removing the 4,s in r7c3 and r7c6 and 6 in r7c4.After these reductions an XY-Wing appears with {1,3}r7c3 as a pivot and {3,4}r7c4; {1,4}r9c3 as pincers eliminating 4 in r7c1. This solves the puzzle. |
|
| Back to top |
|
|
Marty R.
Joined: 12 Feb 2006
Posts: 5770
Location: Rochester, NY, USA
|
| Posted: Sat Jan 19, 2008 7:10 pm Post subject: Re: bugs and XY wings |
|
|
| George Woods wrote: | | As I understand the Bug +1 logic ,, it says that wrong value in the 3candidate cell leaves nothing but doublets (going around in perpetuity - and no solution) and that just leaves one valid candidate there BUT what if the doublets have an XY wing included - as they clearly do here - is the BUG+1 logic valid? |
Yes it is. All that means is that--as is very often the case--there are multiple techniques that are available.
| Quote: | | What is shown above as a "Deadly Pattern" is suceptible to the XY wing! |
If I understand you correctly, you are saying that the Deadly Pattern grid contains XY-Wings. But you would never get to that grid because it is indeed the Deadly Pattern and thus no longer has a unique solution. |
|
| Back to top |
|
|
cgordon
Joined: 04 May 2007
Posts: 769
Location: ontario, canada
|
| Posted: Sat Jan 19, 2008 7:23 pm Post subject: |
|
|
Nataraj: You're absolutely right. There's a triple in C4 that leaves a Bug+1.
Marty: I don't really follow the Bug+3 thing. I think that a Bug+1 is when you just have one triple left - which forms an xyz wing with an xy and an xz. For some reason, which I've long forgotten, you make the triple equal x. But what you are implying here is that I can't do the same for the three xyz wings available. I have to do some work !! |
|
| Back to top |
|
|
storm_norm
Joined: 18 Oct 2007
Posts: 1741
|
| Posted: Sat Jan 19, 2008 7:32 pm Post subject: |
|
|
UR {56},
x-wing on 6,
then after a naked triple in col 4 which takes out the 3 that the skyscraper takes out
an x-wing on 4
then xy-wing on {3,4,9} pincers in r5c4 and r3c6 |
|
| Back to top |
|
|
keith
Joined: 19 Sep 2005
Posts: 3355
Location: near Detroit, Michigan, USA
|
|
| Back to top |
|
|
storm_norm
Joined: 18 Oct 2007
Posts: 1741
|
| Posted: Sat Jan 19, 2008 9:12 pm Post subject: |
|
|
| other than Johan's one step approach. this was a very nice "VH" to solve, considering that the puzzles have had a history of one step moves lately. |
|
| Back to top |
|
|
cgordon
Joined: 04 May 2007
Posts: 769
Location: ontario, canada
|
| Posted: Sat Jan 19, 2008 11:09 pm Post subject: |
|
|
| I sussed out Keith’s thread for Medusa colouring and BUG+3’s. Wasn’t Medusa the one who had snakes growing out of her head. She was probably OK until she took up the colouring thing. The BUG+3s would have probably killed her. Definitely too much work for this loafer. |
|
| Back to top |
|
|
keith
Joined: 19 Sep 2005
Posts: 3355
Location: near Detroit, Michigan, USA
|
| Posted: Sun Jan 20, 2008 12:10 am Post subject: |
|
|
cgordon,
Medusa coloring was invented / discovered by Bob Hanson, a Chemistry Professor in Minnesota.
He has an interesting 3D way of looking at Sudoku puzzles. Imagine a 9x9x9 cube. Each layer contains information about each candidate. Layer 1 is <1>, layer 2 is <2>, etc.
So, if you go coloring on <1>, it is all in the first layer. But, when he saw that you could also color cells that have only two candidates, his 3D visualization reminded him of the intertwined snakes in Medusa's hair. Hence, the name.
Go here:
http://www.stolaf.edu/people/hansonr/sudoku/index.htm
and click on 3D view it the top right of the page. Wait for it to load. Then, drag on the image with your mouse. How cool is that??
Keith |
|
| Back to top |
|
|
|
FAQ
Search
Memberlist
Usergroups
Register
Profile
Log in to check your private messages
Log in
