dailysudoku.com Forum Index dailysudoku.com
Discussion of Daily Sudoku puzzles
 
 FAQFAQ   SearchSearch   MemberlistMemberlist   UsergroupsUsergroups   RegisterRegister 
 ProfileProfile   Log in to check your private messagesLog in to check your private messages   Log inLog in 

Aug 20 VH
Goto page 1, 2  Next
 
Post new topic   Reply to topic    dailysudoku.com Forum Index -> Daily Sudoku puzzles
View previous topic :: View next topic  
Author Message
Johan



Joined: 25 Jun 2007
Posts: 206
Location: Bornem Belgium

PostPosted: Mon Aug 20, 2007 12:34 am    Post subject: Aug 20 VH Reply with quote

This VH has a one step move that succumbs the puzzle completely.
Back to top
View user's profile Send private message Send e-mail
Marty R.



Joined: 12 Feb 2006
Posts: 5770
Location: Rochester, NY, USA

PostPosted: Mon Aug 20, 2007 5:08 am    Post subject: Re: Aug 20 VH Reply with quote

Johan wrote:
This VH has a one step move that succumbs the puzzle completely.

Three steps for me: two XY-Wings and a Type 2 rectangle.
Back to top
View user's profile Send private message
cgordon



Joined: 04 May 2007
Posts: 769
Location: ontario, canada

PostPosted: Mon Aug 20, 2007 1:39 pm    Post subject: Reply with quote

I found an xy wing 48, 38, 34(pivot) that got rid of 8s in R9, Box 8 and opened the way. (Also some subsequent x wings on 4s - but I'm not sure whether x wings fall in the VH category).

Actually I've spent a few months trying to learn about URs, ERs and W-wings but can never spot them - certainly can't see any here. I am even wondering if they are of academic interest only - particularly since I have yet to find a VH that I couldn't solve without xy or xyz wings.

Is that controversial??
Back to top
View user's profile Send private message
TKiel



Joined: 22 Feb 2006
Posts: 292
Location: Kalamazoo, MI

PostPosted: Mon Aug 20, 2007 2:25 pm    Post subject: Reply with quote

I also used the same XY-wing as cgordon but there seemed to be almost as many basic moves after it as before it, before everything turned to singles.

cgordon wrote:
Actually I've spent a few months trying to learn about URs, ERs and W-wings but can never spot them - certainly can't see any here. I am even wondering if they are of academic interest only - particularly since I have yet to find a VH that I couldn't solve without xy or xyz wings.


That's probably partly because UR's, ER's and W-wings are not programmed in the puzzle generator/solver and thus not used in the grading, while XY & XYZ-wings are. Finding UR's & W-wings is generally not a problem, it's finding useful ones that is.
Back to top
View user's profile Send private message
nataraj



Joined: 03 Aug 2007
Posts: 1048
Location: near Vienna, Austria

PostPosted: Mon Aug 20, 2007 3:40 pm    Post subject: Reply with quote

Same xy-wing as cgordon,
then xy-wing 89-39-38 (pivot r9c8) making r8c6=4.
x-wing on 4s r3,7 clears c2,3
only singles from there
Back to top
View user's profile Send private message Send e-mail Visit poster's website
TKiel



Joined: 22 Feb 2006
Posts: 292
Location: Kalamazoo, MI

PostPosted: Mon Aug 20, 2007 4:38 pm    Post subject: Reply with quote

It appears most reached the 'same' point, but with much different looking grids.
Back to top
View user's profile Send private message
nataraj



Joined: 03 Aug 2007
Posts: 1048
Location: near Vienna, Austria

PostPosted: Mon Aug 20, 2007 5:59 pm    Post subject: Reply with quote

Tracy you are right - the "key move" is the same, the other moves probably depend on the preferred order of methods one applies.

When trying to recreate the "same" point, I arrived at this position:
Code:

+--------------------------+--------------------------+--------------------------+
| 7       124     9        | 468     5       2468     | 3       16      246      |
| 2346    5       12346    | 46      39      7        | 1269    8       2469     |
| 8       234     2346     | 1       39      246      | 269     5       7        |
+--------------------------+--------------------------+--------------------------+
| 26      9       2568     | 3       4       1        | 2568    7       268      |
| 34      7       48       | 2       6       5        | 189     139     389      |
| 236     123     12356    | 7       8       9        | 256     4       236      |
+--------------------------+--------------------------+--------------------------+
| 1       248     24       | 4689    7       3        | 689     69      5        |
| 9       6       34       | 5       1       48       | 7       2       38       |
| 5       38      7        | 689     2       68       | 4       369     1        |
+--------------------------+--------------------------+--------------------------+

and the xy-wing 38-34-48 pivot r8c3 solves it all without any more x wings xy wings or anything . I must have done something different this time or maybe just saw a basic move I didn't see first time. For one thing I am sure that first time I missed the AIC (call it kite or skyscraper or whatever) on '1' ( 62=12-18=58 ) which removes 1 from r5c6 and r6c7 (both cells see both ends of the chain)
Back to top
View user's profile Send private message Send e-mail Visit poster's website
Johan



Joined: 25 Jun 2007
Posts: 206
Location: Bornem Belgium

PostPosted: Mon Aug 20, 2007 8:07 pm    Post subject: key-move Reply with quote

The xy-wing with pivot in R8C3 is the key-move, after that it's just basic eliminations.



Johan,
Back to top
View user's profile Send private message Send e-mail
TKiel



Joined: 22 Feb 2006
Posts: 292
Location: Kalamazoo, MI

PostPosted: Mon Aug 20, 2007 9:04 pm    Post subject: Reply with quote

Johan wrote:
The xy-wing with pivot in R8C3 is the key-move, after that it's just basic eliminations.


You're right about the key move, but it seems like most of the time with the VH's once the key move is done, nothing remains but singles. This puzzle had a box-line interaction (which I know is basic, though not always easy to spot) after the key move, before it devolved to singles.

Marty,

I'm curious as to which UR you used.
Back to top
View user's profile Send private message
Marty R.



Joined: 12 Feb 2006
Posts: 5770
Location: Rochester, NY, USA

PostPosted: Mon Aug 20, 2007 9:57 pm    Post subject: Reply with quote

Quote:
Marty,

I'm curious as to which UR you used.


Tracy,

I'm a paper-and-pencil guy so I don't have any intermediate positions available. However, I had only a 26 in r4c9 and one of them was a strong link in row 6, so it was a Type 2 with the roof cells both being 236. This came after the two XY-Wings, so it was the rectangle that busted open the puzzle.

But from the posted position here, the 348 XY-Wing solves the puzzle, so I don't really know exactly what I did.
Back to top
View user's profile Send private message
bette



Joined: 22 Jan 2006
Posts: 6
Location: Fredericton, NB Canada

PostPosted: Tue Aug 21, 2007 12:28 am    Post subject: Reply with quote

Usually I can solve these VH or at least understand how you more experienced folks solved them but I've been staring at this one for quite some time and I just can't see how the 1 and 3 were eliminated from R5C3. Can someone help, please?
Back to top
View user's profile Send private message
cgordon



Joined: 04 May 2007
Posts: 769
Location: ontario, canada

PostPosted: Tue Aug 21, 2007 12:52 am    Post subject: Reply with quote

Quote:
Marty, I'm curious as to which UR you used


So am I. To me a UR is a rectangle with a 12 in three corners and a 123 in the fourth --- which leaves the 3 as the right number. Dead simple - except I can never find them. Still can't see them here!!
Back to top
View user's profile Send private message
Marty R.



Joined: 12 Feb 2006
Posts: 5770
Location: Rochester, NY, USA

PostPosted: Tue Aug 21, 2007 3:53 am    Post subject: Reply with quote

Craig, see my above post responding to Tracy. If you're unfamiliar with UR theory, you can read Keith's primer at:

http://www.sudoku.com/forums/viewtopic.php?p=29105#29105

Bette, I don't know about the 3, but as for the 1, look at box 6. The only possibilities for a 1 in that box are in row 5, so a 1 can't exist anywhere else in row 5.
Back to top
View user's profile Send private message
nataraj



Joined: 03 Aug 2007
Posts: 1048
Location: near Vienna, Austria

PostPosted: Tue Aug 21, 2007 7:04 am    Post subject: Reply with quote

Bette, in my previous post I mentioned an AIC that gets rid of the '1' in r5c3. It works as follows:

notice the chain of cells containing '1': 62=12-18=58.
The notation A=B means: if A is 1 then B isn't, if A isn't 1, then B is (strong link).
Strong links are relatively easy to spot. They have exactly two cells with '1' in a row, column or box.

A-B means: if A is 1 then B is not. (weak link)

a chain like this 62=12-18=58 means:
if r6c2 is not 1 then r1c2 is (strong). Then r1c8 is not (weak). Then r5c8 is 1.

As a whole, any such chain with strong links at both ends and alternating weak and strong links in between (called an alternating inference chain, or AIC for short) means that one of the ends must contain '1'

So any cell that "sees" both ends cannot contain '1'.

A similar argument holds for '3'. This time the chain is 58=98-89=83
Back to top
View user's profile Send private message Send e-mail Visit poster's website
bette



Joined: 22 Jan 2006
Posts: 6
Location: Fredericton, NB Canada

PostPosted: Tue Aug 21, 2007 12:29 pm    Post subject: Reply with quote

Marty, thanks for the response. I guess I was assuming the "1" was eliminated from R5C3 first. Perhaps the "1" was eliminated from R6C7 first, in which case your explanation is logical. Now I've got a whole new problem.

Obviously there are some new techniques being applied here other than the ususal X wings, XY wings and XYZ wings which up to now have solved these puzzles for me.

I'm going to have to study a bit more on nataraj's theory as its still fuzzy in my brain yet.
Back to top
View user's profile Send private message
TKiel



Joined: 22 Feb 2006
Posts: 292
Location: Kalamazoo, MI

PostPosted: Tue Aug 21, 2007 2:30 pm    Post subject: Reply with quote

bette,

Another way to look at what nataraj mentioned above.

This is the grid a few steps before the one shown in nataraj's post. Only cells that could contain <1> are shown.

Code:

 *--------------------------------------------------------------------*
 |        -124          |                      |        +16           |
 |               +12346 |                      | -1269                |
 |                      |                      |                      |
 |----------------------+----------------------+----------------------|
 |                      |                      |                      |
 |               1348   |                      | 189    -139          |
 |        +123   12356  |                      | 1256                 |
 |----------------------+----------------------+----------------------|
 |                      |                      |                      |
 |                      |                      |                      |
 |                      |                      |                      |
 *--------------------------------------------------------------------*


In C2, either r6c2 or r1c2 will end up being <1>. Whichever one is <1>, the other won't be. These cells are said to be strongly linked, which means "if r6c2 is <1>, then r1c2 is not AND if r1c2 is not <1>, then r6c2 is.

The same is true for R1. Either r1c2 or r1c8 will end up being <1> and the other won't. The same is true for any row, column or box that have only two cells that can contain a <1>.

When those strongly linked cells are connected, we can use them to form a chain. Starting with r6c2, I've marked all the strongly linked cells, with either + or -, to indicate they are opposites of each other in regard to the digit <1>. If r6c2 is <1>, r1c2 is not; if r1c2 is not, r2c3 is; if r2c3 is, r2c7 is not; if r2c7 is not, r1c8 is; if r1c8 is, r5c8 is not.

Constructing a chain like this allows us to see the relationship between the cells in the chain. All of the + cells either will end up being <1> or all of them will end up not being <1>, but they will all be the same. Therefore I can say, if r6c2 is <1>, so is r1c8. Or I can say if r5c8 is <1>, r6c2 is not.

So now what we want to look for is cells outside the chain that see cells marked with + & -. There are two, r6c7 & r5c3. Both of these cells can have the <1> excluded. Why?

If the cells in our chain with + end up being where <1> is, what cells can't have <1>? All the cells marked -, plus r5c3, r6c3 & r6c7.

If the cells in our chain with - end up being where <1> is, what cells can't have <1>? All the cells marked +, plus r5c3, r5c7 & r6c7.

So r5c3 & r6c7 can't be <1> no matter which set of cells in our chain ends up having <1>.

There is also a chain that does the same thing for the <3> in r5c3.
Back to top
View user's profile Send private message
cgordon



Joined: 04 May 2007
Posts: 769
Location: ontario, canada

PostPosted: Tue Aug 21, 2007 5:33 pm    Post subject: Reply with quote

Quote:
Bette, I don't know about the 3, but as for the 1, look at box 6. The only possibilities for a 1 in that box are in row 5, so a 1 can't exist anywhere else in row 5.


Bette: Alluding to Marty's comment: at my early stage of the puzzle there were 1's in both rows 5 and 6 of Box 6, so a straightforward elimination wasn't possible. I used a Skyscraper on 1's in cols 2 & 8 which eliminated the 1's in R5C3 and R6C7
Back to top
View user's profile Send private message
bette



Joined: 22 Jan 2006
Posts: 6
Location: Fredericton, NB Canada

PostPosted: Tue Aug 21, 2007 5:50 pm    Post subject: Reply with quote

Thanks, Tracy....your explanation helped me see the logic. I don't know if I could spot another one on my own but this certainly brought a whole new meaning to the "VH" puzzles. I'll be looking for them in the future.
Back to top
View user's profile Send private message
nataraj



Joined: 03 Aug 2007
Posts: 1048
Location: near Vienna, Austria

PostPosted: Tue Aug 21, 2007 6:02 pm    Post subject: Reply with quote

Men at work .... (just for fun)

if you are so inclined (and don't mind looking at muscles, sweat and blood) this is a photo of the (de-)construction site Wink



now that I lookat it, I realize I missed many strong links ( in 8, 2 in 4, in 9 ) - oh my, how do I ever solve any of these?
Back to top
View user's profile Send private message Send e-mail Visit poster's website
cgordon



Joined: 04 May 2007
Posts: 769
Location: ontario, canada

PostPosted: Tue Aug 21, 2007 6:38 pm    Post subject: Reply with quote

Marty: You are right - there was an Type 2 UR on 236. I printed out that UR website before but there were no diagrams for Type 2s. I gotta have pictures!!

I didn't need the UR to solve - but this does challenge my earlier statement that they are never around.
Back to top
View user's profile Send private message
Display posts from previous:   
Post new topic   Reply to topic    dailysudoku.com Forum Index -> Daily Sudoku puzzles All times are GMT
Goto page 1, 2  Next
Page 1 of 2

 
Jump to:  
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