| View previous topic :: View next topic |
| Author |
Message |
daj95376
Joined: 23 Aug 2008
Posts: 3854
|
 Posted: Sat Apr 17, 2010 6:01 pm Post subject: Puzzle 10/04/17 ___ BBDB |
 |
|
| Code: | +-----------------------+
| 8 . 7 | 1 4 . | 9 . . |
| . 2 . | 9 . . | . 1 . |
| 1 . 9 | . 8 5 | . . . |
|-------+-------+-------|
| 4 8 . | 7 . . | 3 5 . |
| 9 . 2 | . . 3 | . . . |
| . . 3 | . 1 . | 2 9 . |
|-------+-------+-------|
| 3 . . | 8 . 4 | . . . |
| . 7 . | 5 . 1 | . 8 . |
| . . . | . . . | . . . |
+-----------------------+
|
Play this puzzle online at the Daily Sudoku site
| Hint wrote: | Potpourri and XYZ-Wing possible.
|
|
|
| Back to top |
|
 |
tlanglet
Joined: 17 Oct 2007
Posts: 2468
Location: Northern California Foothills
|
| Posted: Sat Apr 17, 2010 10:43 pm Post subject: |
|
|
Finned xy-wing 15-6 with vertex 15 in r7c3 and fin 6 in r7c3
If xy-wing is true: r8c3<>6
If fin is true: r79c2|r8c3<6>6 to complete the puzzle.
Ted |
|
| Back to top |
|
|
storm_norm
Joined: 18 Oct 2007
Posts: 1741
|
| Posted: Sat Apr 17, 2010 11:53 pm Post subject: |
|
|
| tlanglet wrote: | Finned xy-wing 15-6 with vertex 15 in r7c3 and fin 6 in r7c3
If xy-wing is true: r8c3<>6
If fin is true: r79c2|r8c3<6>6 to complete the puzzle.
Ted |
very nice |
|
| Back to top |
|
|
daj95376
Joined: 23 Aug 2008
Posts: 3854
|
| Posted: Sun Apr 18, 2010 12:52 am Post subject: |
|
|
| (abbreviated) tlanglet wrote: | Finned xy-wing 15-6 with vertex 15 in r7c3 and fin 6 in r7c3
If xy-wing is true: r8c3<>6
If fin is true: r8c3<>6 to complete the puzzle.
|
Ted: I get the feeling you found ...
| Code: | <15+6> XYZ-Wing r7c3/r4c3+r9c1 <> 6 r8c3
+-----------------------------------------------------------------------------------------+
| 8 36 7 | 1 4 26 | 9 236 5 |
| 56 2 456 | 9 367 67 | 468 1 3468 |
| 1 346 9 | 23 8 5 | 467 23467 23467 |
|-----------------------------+-----------------------------+-----------------------------|
| 4 8 16 | 7 29 29 | 3 5 16 |
| 9 16 2 | 46 5 3 | 14678 467 14678 |
| 7 5 3 | 46 1 8 | 2 9 46 |
|-----------------------------+-----------------------------+-----------------------------|
| 3 169 156 | 8 2679 4 | 1567 267 12679 |
| 2 7 46 | 5 39 1 | 46 8 39 |
| 56 1469 8 | 23 23679 2679 | 14567 23467 1234679 |
+-----------------------------------------------------------------------------------------+
# 92 eliminations remain
|
... which bypasses all of the other VH/VH+ steps in my solver's solution.
I need to pay closer attention to all of the steps present after basics!  |
|
| Back to top |
|
|
Luke451
Joined: 20 Apr 2008
Posts: 310
Location: Southern Northern California
|
| Posted: Sun Apr 18, 2010 2:36 am Post subject: |
|
|
| tlanglet wrote: | Finned xy-wing 15-6 with vertex 15 in r7c3 and fin 6 in r7c3
If xy-wing is true: r8c3<>6
If fin is true: r79c2|r8c3<6>6 to complete the puzzle.
Ted |
That's a creative way of looking at an als-xz. |
|
| Back to top |
|
|
Mogulmeister
Joined: 03 May 2007
Posts: 1151
|
| Posted: Sun Apr 18, 2010 4:51 am Post subject: |
|
|
Very nice Ted.
As well as a finned xy isn't the 56-156-16 a normal XYZ wing ? |
|
| Back to top |
|
|
Mogulmeister
Joined: 03 May 2007
Posts: 1151
|
| Posted: Sun Apr 18, 2010 7:12 am Post subject: |
|
|
| Oops sorry Danny - forgot to look at all the output from your solver. |
|
| Back to top |
|
|
tlanglet
Joined: 17 Oct 2007
Posts: 2468
Location: Northern California Foothills
|
| Posted: Sun Apr 18, 2010 2:11 pm Post subject: |
|
|
About a year ago, Asellus made several posts about finned xy-wings and even completed one somewhat messy puzzle using only that pattern. Ever since, I have had a fondness for them.
As usual, after basics for this puzzle I started looking for xy-wings and quickly spotted this pattern and reacted "finned xy-wing"; I never paused to consider a normal old xyz-wing. Also, I have recently been looking for more unusual pattern and this case fit that mode of thinking nicely.
Danny, I was very surprised when the puzzle fell apart with one move. but you have no reason to be "red faced". In fact, maybe I am the one who should be embarrassed, putting a fancy name on a standard pattern but I am not; that's life!
Luke, I did see the ALS aspect; (6)r47c3 = (5)r7c3 but choose to post the finned pattern. Even this view did not translate into a normal xyz-wing.
Ted |
|
| Back to top |
|
|
Marty R.
Joined: 12 Feb 2006
Posts: 5770
Location: Rochester, NY, USA
|
| Posted: Mon Apr 19, 2010 1:10 am Post subject: |
|
|
I used probably some unnecessary moves before finding the one that did the job. I used an XY-Wing (263) plus pincer transport, Finned X-Wing (4), XY-Wing (392) before noticing the XYZ (156).
Danny, I wouldn't bother altering your solver for Finned XY-Wings. No offense to anyone, nothing personal, apologies to Ted, don't want to ruffle any feathers, but I view Finned XY-Wings, along with Finned skyscrapers, pairs and triples, as trial and error, basically one step up from a forcing chain. |
|
| Back to top |
|
|
daj95376
Joined: 23 Aug 2008
Posts: 3854
|
| Posted: Mon Apr 19, 2010 2:45 am Post subject: |
|
|
Marty: A lot has been discussed about trial & error, and semantics has always gotten in the way IMHO.
FWIW, consider the Naked Pair below.
| Code: | +-----------------------------------+
| . . . | . . . | . . . |
| . . . | . . . | . . . |
| . . . | . . . | . . . |
|-----------+-----------+-----------|
| . . . | . . . | . . . |
| * 12 * | * * * | * 12 * |
| . . . | . . . | . . . |
|-----------+-----------+-----------|
| . . . | . . . | . . . |
| . . . | . . . | . . . |
| . . . | . . . | . . . |
+-----------------------------------+
stream #1) r5c2=1, r5c8=2 => all (*) cells <>12
stream #2) r5c2=2, r5c8=1 => all (*) cells <>12
|
A forcing chain "network" is the basis for every Naked Pair elimination you've ever performed.
Regards, Danny |
|
| Back to top |
|
|
Marty R.
Joined: 12 Feb 2006
Posts: 5770
Location: Rochester, NY, USA
|
| Posted: Mon Apr 19, 2010 3:48 am Post subject: |
|
|
| daj95376 wrote: | Marty: A lot has been discussed about trial & error, and semantics has always gotten in the way IMHO.
FWIW, consider the Naked Pair below.
| Code: | +-----------------------------------+
| . . . | . . . | . . . |
| . . . | . . . | . . . |
| . . . | . . . | . . . |
|-----------+-----------+-----------|
| . . . | . . . | . . . |
| * 12 * | * * * | * 12 * |
| . . . | . . . | . . . |
|-----------+-----------+-----------|
| . . . | . . . | . . . |
| . . . | . . . | . . . |
| . . . | . . . | . . . |
+-----------------------------------+
stream #1) r5c2=1, r5c8=2 => all (*) cells <12> all (*) cells <>12
|
A forcing chain "network" is the basis for every Naked Pair elimination you've ever performed.
Regards, Danny |
Yes, I've been in some T&E discussions and, in the final analysis, it's like beauty, in the eye of the beholder. As far as a Forcing Chain being the basis for a Naked Pair, I'm no theoretician and maybe it is, but whenever we see a naked pair (or triple, quad or quint) we need to test nothing, we know from the pattern that eliminations can be made without further ado. In a T&E situation, we sort of go on a fishing expedition, we see what happens if we do this and see what happens if we do that. We don't have to do that when we spot patterns like wings, URs, coloring and the like.
Don't get me wrong, I have to resort to T&E more than I'd like. It's either that or failure to complete the puzzle, but I don't get satisfaction from using it. |
|
| Back to top |
|
|
Mogulmeister
Joined: 03 May 2007
Posts: 1151
|
| Posted: Mon Apr 19, 2010 6:01 am Post subject: |
|
|
I completely agree about the various arguments etc but
Are all finned x-wings etc trial and error ? Normally a simple finned structure works because it eliminates a digit in common with the regular structure it is attached to. Generally there are less eliminations made at the end.
What you say usually is, "whether the fin is true or the xwing is true they both cause digit(s) x to be eliminated". This is directly analagous to the pincers in an XY chain or XY Wing where we don't know what values they have but that one of them being true is sufficiency to eliminate.
I expect I may have missed something.  |
|
| Back to top |
|
|
Luke451
Joined: 20 Apr 2008
Posts: 310
Location: Southern Northern California
|
| Posted: Mon Apr 19, 2010 6:41 am Post subject: |
|
|
| Marty wrote: | | I view Finned XY-Wings, along with Finned skyscrapers, pairs and triples, as trial and error, basically one step up from a forcing chain. |
Marty, I may be misunderstanding your point of view, because I'm no theorist either!
I think the term "finned" here is being used interchangeably with the term "almost." Almost patterns are logical patterns and certainly do not belong in the scrapheap of trial-and error, especially when puzzles get more difficult.
Here's what we are looking at:
| Code: | *--------------------------------------------------------------------------------------*
| 8 36 7 | 1 4 26 | 9 236 5 |
| 56 2 456 | 9 367 67 | 468 1 3468 |
| 1 346 9 | 23 8 5 | 467 23467 23467 |
|----------------------------+----------------------------+----------------------------|
| 4 8 *16 | 7 29 29 | 3 5 16 |
| 9 16 2 | 46 5 3 | 14678 467 14678 |
| 7 5 3 | 46 1 8 | 2 9 46 |
|----------------------------+----------------------------+----------------------------|
| 3 169 *15+6 | 8 2679 4 | 1567 267 12679 |
| 2 7 46 | 5 39 1 | 46 8 39 |
|*56 1469 8 | 23 23679 2679 | 14567 23467 1234679 |
*--------------------------------------------------------------------------------------* |
Ted chose to see this as an almost or "finned" xy-wing. In doing so, he made a statement. Either the xy-wing is true or 6(r7c3) is true. Where is the T&E in a logical statement?
| Marty wrote: | | In a T&E situation, we sort of go on a fishing expedition, we see what happens if we do this and see what happens if we do that. We don't have to do that when we spot patterns like wings, URs, coloring and the like. |
Patterns that one can easily recognize do not exist in a vacuum. They are powered by underlying logical statements. Let's say this puzzle had just the xy-wing.
| Code: | *--------------------------------------------------------------------------------------*
| 8 36 7 | 1 4 26 | 9 236 5 |
| 56 2 456 | 9 367 67 | 468 1 3468 |
| 1 346 9 | 23 8 5 | 467 23467 23467 |
|----------------------------+----------------------------+----------------------------|
| 4 8 *16 | 7 29 29 | 3 5 16 |
| 9 16 2 | 46 5 3 | 14678 467 14678 |
| 7 5 3 | 46 1 8 | 2 9 46 |
|----------------------------+----------------------------+----------------------------|
| 3 169 *15 | 8 2679 4 | 1567 267 12679 |
| 2 7 46 | 5 39 1 | 46 8 39 |
|*56 1469 8 | 23 23679 2679 | 14567 23467 1234679 |
*--------------------------------------------------------------------------------------* |
We all would see that pattern and immediately eliminate (6)r8c3. Is it the pattern that forced the elimination? Or is it the underlying chain of logic that we recognize as a pattern that forced the elimination? (6=1)r4c3-(1=5)r7c3-(5=6)r9c1 =>r8c3<>6.
What makes one T&E and the other legitimate? |
|
| Back to top |
|
|
daj95376
Joined: 23 Aug 2008
Posts: 3854
|
| Posted: Mon Apr 19, 2010 6:57 am Post subject: |
|
|
I am terrible at spotting patterns w/o the help of Simple Sudoku to highlight information for me. There are others who can spot a finned X-Wing from their chicken scratches on a piece of paper. Still others can "see it in their head" without ever writing anything down.
Bottom line: Until you look for something, you'll never find it. So searching does not make it T&E. If you recognize a pattern while scanning the contents of a grid, then it's not T&E -- no matter how complicated the pattern.
If you see a strong link in one row and then check other rows for a strong link in the same columns, then you are searching for an X-Wing pattern and not on a T&E hunt. If you're good, then you'll also learn to spot a finned/Sashimi X-Wing pattern while looking for the X-Wing. It's a simple step from there to Skyscrapers. |
|
| Back to top |
|
|
Mogulmeister
Joined: 03 May 2007
Posts: 1151
|
| Posted: Mon Apr 19, 2010 7:29 am Post subject: |
|
|
| It's funny how we're all configured differently - I find that skyscrapers jump out of a puzzle at me. On the other hand, spotting ALS style eliminations is still the hardest thing for me. |
|
| Back to top |
|
|
Marty R.
Joined: 12 Feb 2006
Posts: 5770
Location: Rochester, NY, USA
|
| Posted: Mon Apr 19, 2010 3:30 pm Post subject: |
|
|
| Quote: | | Are all finned x-wings etc trial and error ? Normally a simple finned structure works because it eliminates a digit in common with the regular structure it is attached to. Generally there are less eliminations made at the end. |
I don't think Finned X-Wings are trial and error. As soon as you spot one you know immediately what eliminations can be made without testing or going fishing.
| Quote: | | Ted chose to see this as an almost or "finned" xy-wing. In doing so, he made a statement. Either the xy-wing is true or 6(r7c3) is true. Where is the T&E in a logical statement? |
The T&E comes into play because once you make that statement, you have to try it out and see what happens. It may or may not yield an elimination. Where is the T&E in looking at r8c5 and saying either the 3 is true or the 9 is true? And so on and so forth. R9c9 either is 1 or is not 1. You can find a million of these "either/or" situations in every puzzle and try them out.
| Quote: | | We all would see that pattern and immediately eliminate (6)r8c3. Is it the pattern that forced the elimination? Or is it the underlying chain of logic that we recognize as a pattern that forced the elimination? (6=1)r4c3-(1=5)r7c3-(5=6)r9c1 =>r8c3<>6. |
Referring to the XY-Wing you mentioned, the pattern forced the elimination as a result of Sudoku's founding fathers having done those chains of logic and proving that the same type of eliminations can be made every time just by virtue of the pattern's existence.
As I said in my original post, it's like beauty, it's in the eye of the beholder. The only thing I know that's 100% factually true is that I don't feel satisfied if I have to play a move that I perceive as T&E.
By the way, just a few short years ago I was arguing about the logic of moves that I today view as T&E. |
|
| Back to top |
|
|
Mogulmeister
Joined: 03 May 2007
Posts: 1151
|
| Posted: Mon Apr 19, 2010 5:12 pm Post subject: |
|
|
Apologies Marty - I misread you - I thought you had said finned x wings when you actually said finned xy wings.
Suitably inspired I used one of these finned xy wings to crack daj100419c ! |
|
| Back to top |
|
|
Luke451
Joined: 20 Apr 2008
Posts: 310
Location: Southern Northern California
|
| Posted: Mon Apr 19, 2010 5:36 pm Post subject: |
|
|
*Argumentative post deleted*
Let's just have a good time the way we all see fit . |
|
| Back to top |
|
|
|
FAQ
Search
Memberlist
Usergroups
Register
Profile
Log in to check your private messages
Log in
