| View previous topic :: View next topic |
| Author |
Message |
arkietech
Joined: 31 Jul 2008
Posts: 1834
Location: Northwest Arkansas USA
|
 Posted: Sun Sep 30, 2012 5:01 am Post subject: Vanhegan Fiendish September 30, 2012 |
 |
|
| Code: |
*-----------*
|..1|...|9..|
|...|318|...|
|3..|9.2|..8|
|---+---+---|
|.7.|253|.6.|
|..3|.7.|4..|
|.2.|486|.3.|
|---+---+---|
|6..|8.4|..7|
|...|531|...|
|..8|...|2..|
*-----------*
|
Play/Print this puzzle online |
|
| Back to top |
|
 |
Clement
Joined: 24 Apr 2006
Posts: 1111
Location: Dar es Salaam Tanzania
|
| Posted: Sun Sep 30, 2012 4:22 pm Post subject: Vanhagen Fiendish September 30, 2012 |
|
|
Code after Basics | Code: |
+-------------+-------+------------+
| 24 8 1 | 7 6 5 | 9 24 3 |
| 2479 49 267 | 3 1 8 | 5 247 246 |
| 3 5 67 | 9 4 2 | 16 17 8 |
+-------------+-------+------------+
| 189 7 4 | 2 5 3 | C18 6 19 |
| A58 6 3 | 1 7 9 | 4 B258 25 |
| 159 2 59 | 4 8 6 | 7 3 159 |
+-------------+-------+------------+
| 6 1 59 | 8 2 4 | 3 59 7 |
| 27 49 27 | 5 3 1 | D68 489 E46 |
| 4-5 3 8 | 6 9 7 | 2 145 F145 |
+-------------+-------+------------+
|
Dual Cell Forcing Chain ABCDEF
(5=8)r5c1-(25-8)r5c8-(1=8)r4c7-(8=6)r8c7-(6=4)r8c9=15r9c89; r9c1<>5 solves it. |
|
| Back to top |
|
|
arkietech
Joined: 31 Jul 2008
Posts: 1834
Location: Northwest Arkansas USA
|
| Posted: Sun Sep 30, 2012 6:11 pm Post subject: |
|
|
Same target; different arrow. 
| Code: |
*-----------------------------------------------------------*
| 24 8 1 | 7 6 5 | 9 b24 3 |
| 2479 49 267 | 3 1 8 | 5 b247 246 |
| 3 5 67 | 9 4 2 | 16 b17 8 |
|-------------------+-------------------+-------------------|
| 189 7 4 | 2 5 3 | 18 6 19 |
|a58 6 3 | 1 7 9 | 4 b258 25 |
| 159 2 9-5 | 4 8 6 | 7 3 159 |
|-------------------+-------------------+-------------------|
| 6 1 c59 | 8 2 4 | 3 b59 7 |
| 27 49 27 | 5 3 1 | 68 489 46 |
| 4-5 3 8 | 6 9 7 | 2 b145 145 |
*-----------------------------------------------------------*
pseudo xy-wing
(5=8)r5c1-(8=9)als:r123579c8-(9=5)r7c3 => -5r6c3,r9c1; stte
|
Play/Print this puzzle online |
|
| Back to top |
|
|
ronk
Joined: 07 May 2006
Posts: 398
|
| Posted: Sun Sep 30, 2012 11:01 pm Post subject: |
|
|
| arkietech wrote: |
pseudo xy-wing
(5=8)r5c1-(8=9)als:r123579c8-(9=5)r7c3 => -5r6c3,r9c1; stte |
I just remembered that bennys named this an ALS xy wing way back in 2005. See here. It was presented in "set form" rather than as an AIC, but it's the same structure. Although I suggested the pseudo term a few days ago, in retrospect the ALS xy-wing term now seems more appropriate. |
|
| Back to top |
|
|
Marty R.
Joined: 12 Feb 2006
Posts: 5770
Location: Rochester, NY, USA
|
| Posted: Sun Sep 30, 2012 11:21 pm Post subject: |
|
|
| Code: |
+-------------+-------+------------+
| 24 8 1 | 7 6 5 | 9 24 3 |
| 2479 49 267 | 3 1 8 | 5 247 246 |
| 3 5 67 | 9 4 2 | 16 17 8 |
+-------------+-------+------------+
| 189 7 4 | 2 5 3 | 18 6 19 |
| 58 6 3 | 1 7 9 | 4 258 25 |
| 159 2 59 | 4 8 6 | 7 3 159 |
+-------------+-------+------------+
| 6 1 59 | 8 2 4 | 3 59 7 |
| 27 49 27 | 5 3 1 | 68 489 46 |
| 45 3 8 | 6 9 7 | 2 145 145 |
+-------------+-------+------------+
|
Play this puzzle online at the Daily Sudoku site
I used the potential DP 19 in boxes 46. Either r4c1=8 or one of r6c19=5. Common outcome; r6c3=9. |
|
| Back to top |
|
|
arkietech
Joined: 31 Jul 2008
Posts: 1834
Location: Northwest Arkansas USA
|
| Posted: Sun Sep 30, 2012 11:47 pm Post subject: |
|
|
| Marty R. wrote: | | Code: |
+-------------+-------+------------+
| 24 8 1 | 7 6 5 | 9 24 3 |
| 2479 49 267 | 3 1 8 | 5 247 246 |
| 3 5 67 | 9 4 2 | 16 17 8 |
+-------------+-------+------------+
| 189 7 4 | 2 5 3 | 18 6 19 |
| 58 6 3 | 1 7 9 | 4 258 25 |
| 159 2 59 | 4 8 6 | 7 3 159 |
+-------------+-------+------------+
| 6 1 59 | 8 2 4 | 3 59 7 |
| 27 49 27 | 5 3 1 | 68 489 46 |
| 45 3 8 | 6 9 7 | 2 145 145 |
+-------------+-------+------------+
|
Play this puzzle online at the Daily Sudoku site
I used the potential DP 19 in boxes 46. Either r4c1=8 or one of r6c19=5. Common outcome; r6c3=9. |
Very Nice! |
|
| Back to top |
|
|
tlanglet
Joined: 17 Oct 2007
Posts: 2468
Location: Northern California Foothills
|
| Posted: Mon Oct 01, 2012 12:42 pm Post subject: |
|
|
I also spotted the AUR(19). Initially I viewed the external strong inferences as:
AUR(19)r46c19[(9)r6c3=(1)r4c7]-(1=9)r4c9; => -9r4c1,r6c9 which is not very useful.
Then I tried another variation of the external strong inferences:
AUR(19)r46c19[(9)r6c3=(1)r4c7]-r4c1=r6c1; => -9r6c1 which is not useful by itself, but combined with the prior approach gives r6c3=9
Finally I used the internal strong inferences:
AUR(19)r46c19[(5)r6c19=(8)r4c1]-(8=5)r5c1; => -5r6c3 which completes the puzzle.
For me the lesson is to "keep working a pattern" to explore all possibilities.
Ted |
|
| Back to top |
|
|
arkietech
Joined: 31 Jul 2008
Posts: 1834
Location: Northwest Arkansas USA
|
| Posted: Mon Oct 01, 2012 1:35 pm Post subject: |
|
|
| tlanglet wrote: | Finally I used the internal strong inferences:
AUR(19)r46c19[(5)r6c19=(8)r4c1]-(8=5)r5c1; => -5r6c3 which completes the puzzle.
|
I have been playing with some new notation. Is this valid?
(5=8)aur:r46c19-(8=5)r5c1 => -5r6c3
I find it easier to understand.
dan |
|
| Back to top |
|
|
tlanglet
Joined: 17 Oct 2007
Posts: 2468
Location: Northern California Foothills
|
| Posted: Mon Oct 01, 2012 2:55 pm Post subject: |
|
|
| arkietech wrote: | | tlanglet wrote: | Finally I used the internal strong inferences:
AUR(19)r46c19[(5)r6c19=(8)r4c1]-(8=5)r5c1; => -5r6c3 which completes the puzzle.
|
I have been playing with some new notation. Is this valid?
(5=8)aur:r46c19-(8=5)r5c1 => -5r6c3
I find it easier to understand.
dan |
Dan,
I am definitely the wrong person to be commenting on notation but here is my opinion.......
1) I tend to be verbose and show detail so that others can easily flow the logic, especially those individuals trying to learn the technique.
2) Your notation is very curt but is fine when using internal strong inferences.
3) I think it is important/required to specify the location of any external strong inferences. How would you suggest to do that?
Ted |
|
| Back to top |
|
|
arkietech
Joined: 31 Jul 2008
Posts: 1834
Location: Northwest Arkansas USA
|
| Posted: Mon Oct 01, 2012 3:24 pm Post subject: |
|
|
| tlanglet wrote: |
3) I think it is important/required to specify the location of any external strong inferences. How would you suggest to do that? |
I do not understand "external". I assume you mean "Local" (Line,Column or Box) rather than within a cell. 
I would show any necessary strong (and weak) inferences in the normal manner. eg. =*rncn or -*rncn. ( *=inference rncn=location)
Do you see any important/necessary inferences not shown in this solution?
(5=8)aur:r46c19-(8=5)r5c1 => -5r6c3
I guess it is a personal thing. To me too many details tend to confuse the issue rather than help.
dan |
|
| Back to top |
|
|
arkietech
Joined: 31 Jul 2008
Posts: 1834
Location: Northwest Arkansas USA
|
| Posted: Mon Oct 01, 2012 4:48 pm Post subject: |
|
|
| ronk wrote: | | arkietech wrote: |
pseudo xy-wing
(5=8)r5c1-(8=9)als:r123579c8-(9=5)r7c3 => -5r6c3,r9c1; stte |
I just remembered that bennys named this an ALS xy wing way back in 2005. See here. It was presented in "set form" rather than as an AIC, but it's the same structure. Although I suggested the pseudo term a few days ago, in retrospect the ALS xy-wing term now seems more appropriate. |
I got so excited about Marty's aur I almost missed your post!
Sometimes an almost naked pair (anp) or triple (ant) seems to fit better than als. Could there be such a thing as an anp xy-wing?
for example:
| Code: | *-----------------------------------------------------------*
| 1 234 2349 | 6 29 8 | 235 2459 7 |
| 8 7 29 | 3 4 5 | 6 29 1 |
| 2346 2346 5 | 1 29 7 | 23 2489 48 |
|-------------------+-------------------+-------------------|
|b34 1 6 | 8 5 9 | 7 34 2 |
| 245 8 24 | 7 3 6 | 9 1 45 |
| 9 a35 7 | 4 1 2 | 8 35 6 |
|-------------------+-------------------+-------------------|
| 56 9 1 | 2 68 3 | 4 7 58 |
|c245 24-5 8 | 9 7 1 |c25 6 3 |
| 7 236 23 | 5 68 4 | 1 28 9 |
*-----------------------------------------------------------*
als? xy-wing
(5=3)r6c2-(3=4)r4c1-(4=5)anp:r8c17 => -5r8c2; stte
|
|
|
| Back to top |
|
|
tlanglet
Joined: 17 Oct 2007
Posts: 2468
Location: Northern California Foothills
|
| Posted: Mon Oct 01, 2012 10:03 pm Post subject: |
|
|
| arkietech wrote: | | tlanglet wrote: |
3) I think it is important/required to specify the location of any external strong inferences. How would you suggest to do that? |
I do not understand "external". I assume you mean "Local" (Line,Column or Box) rather than within a cell.
I would show any necessary strong (and weak) inferences in the normal manner. eg. =*rncn or -*rncn. ( *=inference rncn=location)
Do you see any important/necessary inferences not shown in this solution?
(5=8)aur:r46c19-(8=5)r5c1 => -5r6c3
I guess it is a personal thing. To me too many details tend to confuse the issue rather than help.
dan |
Dan,
Here is a reference to the term "external inferences".
In an earlier post, I used an external inference:
AUR(19)r46c19[(9)r6c3=(1)r4c7]-(1=9)r4c9; => -9r4c1,r6c9
How would you notate the AUR using those conditions?
Ted[/b] |
|
| Back to top |
|
|
arkietech
Joined: 31 Jul 2008
Posts: 1834
Location: Northwest Arkansas USA
|
| Posted: Tue Oct 02, 2012 12:10 am Post subject: |
|
|
| tlanglet wrote: | | How would you notate the AUR using those conditions? |
I would do it this way.
| Code: | *-----------------------------------------------------------*
| 24 8 1 | 7 6 5 | 9 24 3 |
| 2479 49 267 | 3 1 8 | 5 247 246 |
| 3 5 67 | 9 4 2 | 16 17 8 |
|-------------------+-------------------+-------------------|
| 189 7 4 | 2 5 3 |b18 6 c19 |
| 58 6 3 | 1 7 9 | 4 258 25 |
| 159 2 a59 | 4 8 6 | 7 3 159 |
|-------------------+-------------------+-------------------|
| 6 1 59 | 8 2 4 | 3 59 7 |
| 27 49 27 | 5 3 1 | 68 489 46 |
| 45 3 8 | 6 9 7 | 2 145 145 |
*-----------------------------------------------------------*
(9=5)r6c3-(5=8)aur:r46c19-(8=1)r4c7-(1=9)r4c9 => -9r4c1,r6c9 |
|
|
| Back to top |
|
|
aran
Joined: 19 Apr 2010
Posts: 70
|
| Posted: Tue Oct 02, 2012 10:15 am Post subject: |
|
|
On notation in an AUR setting :
tlanglet writes :
AUR(19)r46c19[(5)r6c19=(8)r4c1]-(8=5)r5c1; => -5r6c3
the logic here is :
[UR19r46c19]=[5r6c19=8r4c1-(8=5)r5c1] : <5>r6c3
Since UR19r46c19 cannot exist, the first term plays no part, so square brackets could be dropped from it giving
UR19r46c19=[5r6c19=8r4c1-(8=5)r5c1] : <5>r6c3
or even more daringly
UR19r46c19=5r6c19=8r4c1-(8=5)r5c1 : <5>r6c3
arkietech writes
(9=5)r6c3-(5=8)aur:r46c19-(8=1)r4c7-(1=9)r4c9 => -9r4c1,r6c9
which could be expressed
(9=5)r6c3-5r6c19=(AUR-UR)r46c19=1r4c7-(1=9)r4c9
in which the AUR appears in its own right and not so to speak as an explanatory note |
|
| Back to top |
|
|
tlanglet
Joined: 17 Oct 2007
Posts: 2468
Location: Northern California Foothills
|
| Posted: Tue Oct 02, 2012 7:02 pm Post subject: |
|
|
| aran wrote: | On notation in an AUR setting :
tlanglet writes :
AUR(19)r46c19[(5)r6c19=(8)r4c1]-(8=5)r5c1; => -5r6c3
the logic here is :
[UR19r46c19]=[5r6c19=8r4c1-(8=5)r5c1] : <5>r6c3
Since UR19r46c19 cannot exist, the first term plays no part, so square brackets could be dropped from it giving
UR19r46c19=[5r6c19=8r4c1-(8=5)r5c1] : <5>r6c3
or even more daringly
UR19r46c19=5r6c19=8r4c1-(8=5)r5c1 : <5>r6c3
arkietech writes
(9=5)r6c3-(5=8)aur:r46c19-(8=1)r4c7-(1=9)r4c9 => -9r4c1,r6c9
which could be expressed
(9=5)r6c3-5r6c19=(AUR-UR)r46c19=1r4c7-(1=9)r4c9
in which the AUR appears in its own right and not so to speak as an explanatory note |
Aran,
Thanks for the feedback; it is always appreciated.
About 3-4 years ago some mentor on this website suggested the notation form I have been using for AURs; for me, it provides the details I find useful/important. It explicitly shows the external or internal strong inferences derived from the UR, and appears very similar to your daring form.
I find the alternate form you offered for arkietech's notation to be less intuitive and requires greater effort to understand, especially what SIS was used.
To each his/her own preferences.........
Ted |
|
| Back to top |
|
|
daj95376
Joined: 23 Aug 2008
Posts: 3854
|
| Posted: Tue Oct 02, 2012 8:42 pm Post subject: |
|
|
Strong links on <1> & <9>, and bivalue in r4c9, force these two eliminations.
| Code: | after basics
+--------------------------------------------------------------+
| 24 8 1 | 7 6 5 | 9 24 3 |
| 2479 49 267 | 3 1 8 | 5 247 246 |
| 3 5 67 | 9 4 2 | 16 17 8 |
|--------------------+--------------------+--------------------|
| 189 7 4 | 2 5 3 | 18 6 19 |
| 58 6 3 | 1 7 9 | 4 258 25 |
| 159 2 59 | 4 8 6 | 7 3 159 |
|--------------------+--------------------+--------------------|
| 6 1 59 | 8 2 4 | 3 59 7 |
| 27 49 27 | 5 3 1 | 68 489 46 |
| 45 3 8 | 6 9 7 | 2 145 145 |
+--------------------------------------------------------------+
# 42 eliminations remain
r46c19 <19> UR via s-link <> 9 r4c1
r46c19 <19> UR via s-link <> 9 r6c1
|
|
|
| Back to top |
|
|
ronk
Joined: 07 May 2006
Posts: 398
|
| Posted: Wed Oct 03, 2012 2:14 pm Post subject: |
|
|
| arkietech wrote: | Sometimes an almost naked pair (anp) or triple (ant) seems to fit better than als. Could there be such a thing as an anp xy-wing?
...
als? xy-wing
(5=3)r6c2-(3=4)r4c1-(4=5)anp:r8c17 => -5r8c2 |
As you know, each of anp, ant, and anq is an als, so it would be correct. However, what if the ALS xy-wing were comprised of one anp and one ant? Then your approach in the example above would be a good compromise IMO. |
|
| Back to top |
|
|
aran
Joined: 19 Apr 2010
Posts: 70
|
| Posted: Wed Oct 03, 2012 4:35 pm Post subject: |
|
|
| tlanglet wrote: | | aran wrote: |
arkietech writes
(9=5)r6c3-(5=8)aur:r46c19-(8=1)r4c7-(1=9)r4c9 => -9r4c1,r6c9
which could be expressed
(9=5)r6c3-5r6c19=(AUR-UR)r46c19=1r4c7-(1=9)r4c9
in which the AUR appears in its own right and not so to speak as an explanatory note |
Aran,
[...]
I find the alternate form you offered for arkietech's notation to be less intuitive and requires greater effort to understand, especially what SIS was used.
To each his/her own preferences.........
Ted |
Ted
Just to point out that there is no SIS in that chain above. It is a just a series of alternating links.
The trick, if it is one, being use of AUR-UR.
"-UR" is a truth in a sudoku with a unique solution, and as such isn't a consequence of AUR; but on the other hand, being a truth, can be inserted anywhere, so here just after AUR. |
|
| Back to top |
|
|
|
FAQ
Search
Memberlist
Usergroups
Register
Profile
Log in to check your private messages
Log in
