dailysudoku.com :: View topic - Vanhegan Fiendish 5/4/12

dailysudoku.com
Discussion of Daily Sudoku puzzles

FAQ

Memberlist

Usergroups

Profile

Vanhegan Fiendish 5/4/12

dailysudoku.com Forum Index -> Other puzzles

View previous topic :: View next topic

Author

Message

arkietech

Joined: 31 Jul 2008
Posts: 1834
Location: Northwest Arkansas USA

Posted: Fri May 04, 2012 7:13 am Post subject: Vanhegan Fiendish 5/4/12

Code:

*-----------*
|.41|2..|...|
|9..|..1|...|
|.8.|..5|..1|
|---+---+---|
|.95|.24|.7.|
|...|1.7|...|
|.3.|98.|45.|
|---+---+---|
|5..|8..|.4.|
|...|4..|..5|
|...|..9|26.|
*-----------*

Play/Print online

aran

Joined: 19 Apr 2010
Posts: 70

Posted: Fri May 04, 2012 10:33 am Post subject:

Vanhegan 4.5.12 after ssts

Code:

*-----------------------------------------------------------*
| 67 4 1 | 2 9 8 | 5 3 67 |
| 9 5 26 | 67 3 1 | 678 28 4 |
| 237 8 236 | 67 4 5 | 679 29 1 |
|-------------------+-------------------+-------------------|
| 68 9 5 | 3 2 4 | 1 7 68 |
| 248 26 2468 | 1 5 7 | 3689 89 389 |
| 1 3 7 | 9 8 6 | 4 5 2 |
|-------------------+-------------------+-------------------|
| 5 126 2369 | 8 16 23 | 379 4 379 |
| 23 67 89 | 4 67 23 | 89 1 5 |
| 348 17 348 | 5 17 9 | 2 6 38 |
*-----------------------------------------------------------*

In illustratation of notations suggested recently :

8r2c7=AUR67r23c47=9r3c7-(9=8)r8c7 : =><8>r5c7
6r5c7=XYwing(389 : r5c7/r8c7/r9c9)=3r5c7 : =><9>r5c7
sstste

arkietech

Joined: 31 Jul 2008
Posts: 1834
Location: Northwest Arkansas USA

Posted: Fri May 04, 2012 11:39 am Post subject:

aran wrote:

8r2c7=AUR67r23c47=9r3c7-(9=8)r8c7 : =><8>r5c7
6r5c7=XYwing(389 : r5c7/r8c7/r9c9)=3r5c7 : =><9>r5c7
sstste

We both had our eyes on r5c7

Code:

*--------------------------------------------------------------------*
| 67 4 1 | 2 9 8 | 5 3 67 |
| 9 5 26 | 67 3 1 | 678 28 4 |
| 2367 8 236 | 67 4 5 | 679 29 1 |
|----------------------+----------------------+----------------------|
| 68 9 5 | 3 2 4 | 1 7 a68 |
| 2468 26 2468 | 1 5 7 |d389-6 89 c3689 |
| 1 3 7 | 9 8 6 | 4 5 2 |
|----------------------+----------------------+----------------------|
| 5 1267 2369 | 8 167 23 | 379 4 379 |
| 2368 267 23689 | 4 67 23 | 3789 1 5 |
| 348 17 348 | 5 17 9 | 2 6 b38 |
*--------------------------------------------------------------------*

(6=8)r4c9-(8=3)r9c9-r5c9=(3)r5c7 => r5c7<>6; lcls

I like your notation. I'm still studying it. Mine is a wing ( I believe) does it have a name?

tlanglet

Joined: 17 Oct 2007
Posts: 2468
Location: Northern California Foothills

Posted: Fri May 04, 2012 11:59 am Post subject:

Code after basics:

Code:

*--------------------------------------------------------------------*
| 67 4 1 | 2 9 8 | 5 3 67 |
| 9 5 26 | 67 3 1 | 678 28 4 |
| 2367 8 236 | 67 4 5 | 679 29 1 |
|----------------------+----------------------+----------------------|
| 68 9 5 | 3 2 4 | 1 7 68 |
| 2468 26 2468 | 1 5 7 | 3689 89 3689 |
| 1 3 7 | 9 8 6 | 4 5 2 |
|----------------------+----------------------+----------------------|
| 5 1267 2369 | 8 167 23 | 379 4 379 |
| 2368 267 23689 | 4 67 23 | 3789 1 5 |
| 348 17 348 | 5 17 9 | 2 6 38 |
*--------------------------------------------------------------------*

I never got past the AUR(67)

AUR(67)r23c47 SIS (8)r2c7, (9)r3c7 => r1c9=7
(8)r2c7-r2c8=r5c8-(8=6=7)r41c9;
||
(9)r3c7-r78c7=(9-7)r7c9=r1c9;

I also appreciate the notational suggestions and am attempting to realign my thinking to employ them. I have previously posted solutions using the following technique where the SIS are explicitly stated.

(7)r1c9=(7-9)r7c9=r78c7-AUR(67)r23c47[(9)r3c7=(8)r2c7]-r2c8=r5c8-(8=6=7)r41c9 => r1c9=9

[Edited one time to include alternate notation.}

Ted

tlanglet

Joined: 17 Oct 2007
Posts: 2468
Location: Northern California Foothills

Posted: Fri May 04, 2012 12:25 pm Post subject:

aran wrote:

6r5c7=XYwing(389 : r5c7/r8c7/r9c9)=3r5c7 : =><9>r5c7

Aran,

When I first looked at this statement, I did not fully understand it. Initially, the XYwing component caused me to expect to see a deletion, not an assignment statement. After viewing the grid it become obvious that the XYwing deletes the 3 in r5c9 that then forces a 3 in r5c7.

I am still thinking about this missing step in the notation.

Ted

arkietech

Joined: 31 Jul 2008
Posts: 1834
Location: Northwest Arkansas USA

Posted: Fri May 04, 2012 1:16 pm Post subject:

(6=8)r4c9-(8=3)r9c9-r5c9=(3)r5c7 => r5c7<>6 Is this right? sis(6)r4c9,(3)r9c9=(3)r5c7 => r5c7<>6

Marty R.

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

Posted: Fri May 04, 2012 10:10 pm Post subject:

Not particularly proud of this, but r1c1=7 proves two 8s in box 6; r1c1<>7.

daj95376

Joined: 23 Aug 2008
Posts: 3854

Posted: Fri May 04, 2012 10:41 pm Post subject:

arkietech wrote:

Mine is a wing ( I believe) does it have a name?

Code:

H3-Wing: (X=Y)a - (Y=Z)b - (Z)c = (Z)d "a" and "d" in same unit; a<>Z, d<>X

arkietech

Joined: 31 Jul 2008
Posts: 1834
Location: Northwest Arkansas USA

Posted: Fri May 04, 2012 11:31 pm Post subject:

daj95376 wrote:

arkietech wrote:

Mine is a wing ( I believe) does it have a name?

Code:

H3-Wing: (X=Y)a - (Y=Z)b - (Z)c = (Z)d "a" and "d" in same unit; a<>Z, d<>X

Thanks Danny Very Happy

I have been calling: 3-SIS using bivalue(XY), bilocation(Y), and bilocation(Z) an h=wing

Is that right?

ronk

Joined: 07 May 2006
Posts: 398

Posted: Sat May 05, 2012 1:12 am Post subject:

arkietech wrote:

I have been calling: 3-SIS using bivalue(XY), bilocation(Y), and bilocation(Z) an h=wing

Is that right?

I often wonder how, in this case for example, VLL3-wing would have been accepted. IOW sequentially list 'V' and 'L' for each bivalue and bilocal strong link, respectively, and add '2' or '3' depending upon whether two or three candidate values were involved.

Then no need to remember the correct pattern applying to m-, w-, s-, h2-, h3-, L2-, and L3-wings.

arkietech

Joined: 31 Jul 2008
Posts: 1834
Location: Northwest Arkansas USA

Posted: Sat May 05, 2012 1:55 am Post subject:

ronk wrote:

Then no need to remember the correct pattern applying to m-, w-, s-, h2-, h3-, L2-, and L3-wings.

I like it h2 and L2 are the xy's and h3 and L3 are the xyz's Very Happy

ronk

Joined: 07 May 2006
Posts: 398

Posted: Sat May 05, 2012 3:20 am Post subject:

arkietech wrote:

ronk wrote:

Then no need to remember the correct pattern applying to m-, w-, s-, h2-, h3-, L2-, and L3-wings.

I like it h2 and L2 are the xy's and h3 and L3 are the xyz's Very Happy

Unfortunately, IIRC both the h2- and h3-wings are "xyz."

arkietech

Joined: 31 Jul 2008
Posts: 1834
Location: Northwest Arkansas USA

Posted: Sat May 05, 2012 5:43 am Post subject:

ronk wrote:

arkietech wrote:

ronk wrote:

Then no need to remember the correct pattern applying to m-, w-, s-, h2-, h3-, L2-, and L3-wings.

I like it h2 and L2 are the xy's and h3 and L3 are the xyz's Very Happy

Unfortunately, IIRC both the h2- and h3-wings are "xyz."

pjb

Joined: 05 May 2012
Posts: 1
Location: Sydney

Posted: Sat May 05, 2012 6:43 am Post subject:

Here's a chain-free way of solving it: After basic moves, an X-wing of 6 at r1c19, r4c19 gives r358c1; r5c9 <> 6. A finned X-wing of 8's at r4c19; ; r9c139 gives r8c1 <> 8. This produces a naked pair of 23 in row 8 giving r8c2 <>2, r8c3 <>23 and r8c7 <>3, and a second naked pair of 67 in row 8 giving r8c3 <>6 and r8c7 <> 7. As 7's at r7c79 are only ones in box 9, this gives r7c2 and r7c5 <> 7. After this the state is:

Code:

67 4 1 | 2 9 8 | 5 3 67
9 5 26 | 67 3 1 | 678 28 4
237 8 236 | 67 4 5 | 679 29 1
---------------------+----------------------+---------------------
68 9 5 | 3 2 4 | 1 7 68
248 26 2468 | 1 5 7 | 3689 89 389
1 3 7 | 9 8 6 | 4 5 2
---------------------+----------------------+---------------------
5 126 2369 | 8 16 23 | 379 4 379
23 67 89 | 4 67 23 | 89 1 5
348 17 348 | 5 17 9 | 2 6 38

Now there is a lovely type 1 BUG-Lite at r789c2 and r789c5 giving r7c2 = 2, after which the puzzle solves easily.

pjb

aran

Joined: 19 Apr 2010
Posts: 70

Posted: Sat May 05, 2012 11:35 am Post subject:

tlanglet wrote:

aran wrote:

6r5c7=XYwing(389 : r5c7/r8c7/r9c9)=3r5c7 : =><9>r5c7

tlanglet
The idea is to simplify presentation, and in particular to remove any unnecessary brackets
which btw is one reason why I don't use brackets to surround candidates eg 7r1c9 rather than (7)r1c9 unless required for clarity eg 7r1c9=(7-9)r7c9
but that's not the real subject here
which is : in a chain supposedly of alternating inferences can it make sense to have a sequence of =x= (as opposed to =x-) ?
well for a start there are already non-alternating inferences any time brackets are used to represent possibilities
For example imagine this :
xr4c9=...-yr1c1=(zr9c1=abr19c1-(ab=x)r4c1)
which establishes that if r9c1 is not z, then x can be eliminated from r4c2345678 subject to examination of the alternative possibility : r9c1 is z. If it turned out that zr9c1=>xr5c5, then the overall result would be <x>r4c456

When however the alternative is an illegal or impossible structure such as a UR, there is nothing to be examined.
So the brackets can reasonably be removed, or so I contend.
The interpretation in such a case is then this :
non-alternating inferences =X=, and no brackets =>X is an impossible structure, and what immediately follows is necessary to avoid that illegal structure.

Take now your presentation
(7)r1c9=(7-9)r7c9=r78c7-AUR(67)r23c47[(9)r3c7=(8)r2c7]-r2c8=r5c8-(8=6=7)r41c9 => r1c9=9
This would become :
7r1c9=(7-9)r7c9=r78c7-9r3c7=UR67r23c47=8r2c7-r2c8=r5c8-(8=67)r41c9 => r1c9=7
here X is the impossible structure UR67r23c47, and what follows =8r2c7 is necessary to avoid it.
The A of AUR is also surplus to requirements since it is the UR which is the invalid structure.

On the other chain :
6r5c7=XYwing(389 : r5c7/r8c7/r9c9)=3r5c7 : =><9>r5c7
The illegal structure is the sum of the XY wing and the z-candidate which would produce an empty cell at its heart, and what follows =3r5c7 is necessary to avoid that.
The question is : how succinctly can that be written ? Is it necessary to note the XY wing, and the z-candidate, or sufficient to show XY wing alone ? Certainly the latter is preferable stylistically, and the avoidance mechanism clearly points to what the z-candidate must be. Under this approach then, z remains a silent partner....but then he shouldn't be there in the first place !

keith

Joined: 19 Sep 2005
Posts: 3355
Location: near Detroit, Michigan, USA

Posted: Sat May 05, 2012 1:03 pm Post subject:

pjb wrote:

Code:

Now there is a lovely type 1 BUG-Lite at r789c2 and r789c5 giving r7c2 = 2, after which the puzzle solves easily.

pjb

pjb,

Welcome to the forum!

Inspired by your post, I found:

Type 4 UR 17
X-wing 6
XYZ-wing 3-89

(Although I suppose the point of this thread is the notation, not the puzzle solution.)

Keith

Display posts from previous:

	dailysudoku.com Forum Index -> Other puzzles	All times are GMT
Page 1 of 1

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