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 

What else is here?
Goto page Previous  1, 2
 
Post new topic   Reply to topic    dailysudoku.com Forum Index -> Other puzzles
View previous topic :: View next topic  
Author Message
nataraj



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

PostPosted: Fri Jul 18, 2008 6:40 am    Post subject: Reply with quote

keith wrote:

The M-wing is so named because, if you do Medusa coloring, it WILL show up as a (minimum) four-cell chain making a Medusa elimination.

So, my question is: If you do Medusa coloring on this grid, will Nataraj' chain show up, making the elimination he identifies? (I have not tried this.)

If yes, I agree it is an M-wing, albeit harder to recognize.

If no, maybe it has another name?


Ah ... I was not aware that Medusa had anything to do with it ...

Now that you mention it, Keith, I see the connection (start with a bi-value, use strong inferences in both candidates, etc.) Yep, makes sense.

And no, I don't think that my elimination has anything to do with Medusa. First, there are too many weak links and cells with more than two candidates involved. Second, I Medusa'd the given grid, and as Marty said, it quickly turns into a wrap that does much much more than just the one elimination I found.

Therefore: no "M-wing" Crying or Very sad

If it needs a new name indeed, so be it. Why not call it a "left wing" because obviously we left the orthodox definition of M-wing Laughing ?
If you guys think a one-letter wing would be more appropriate, how about r-wing then?
I keep saying "O.K., if it's not an m-wing, it's an - err ..."
(opens and closes mouth)
... "wing"
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: Fri Jul 18, 2008 1:28 pm    Post subject: Reply with quote

Quote:
This is easily solved with XY-Chains or simple Medusa.


For me, that's like saying it's easily solved by quantum theory or relativity. But since I did once try to learn XY-Chains - where's the XY Chain here? I see three linked 79s but they don't seem to go anywhere.

Cheers
Back to top
View user's profile Send private message
Steve R



Joined: 24 Oct 2005
Posts: 289
Location: Birmingham, England

PostPosted: Fri Jul 18, 2008 4:12 pm    Post subject: Reply with quote

In Marty’s original grid you could try:

r4c7 = 9 => r9c7 = 7 => r9c1 = 8 => r2c1 = 3 => r4c1 =9 => r4c7 ≠ 9

Steve
Back to top
View user's profile Send private message
Marty R.



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

PostPosted: Fri Jul 18, 2008 4:50 pm    Post subject: Reply with quote

Code:
+-------------+---------+----------+----------
| 2    4    7     | 6   3   8  | 1   9   5   |
| 38   358  358   | 1   7   9  | 4   2   6   |
| 1    6    9     | 2   4   5  | 78  78  3   |
+-------------+---------+----------+----------
| 39   1    4     | 7   5   2  | 39  6   8   |
| 6    38g  2     | 4   89  1  | 5   37f 79e |
| -789 578i 58h   | 3   89  6  | 2   4   1   |
+-------------+---------+----------+----------
| 4    9    1     | 5   6  7   | 38  38  2   |
| 5    3-78 38    | 89  2  4   | 6   1   79d |
| 78a  2    6     | 89b 1  3   | 79c 5   4   |
+-------------+---------+----------+---------- 

Craig,

One XY-Chain I saw started with setting r9c1=8 and moved alphabetically from a to i. Then r6c2(i) must be7 and that acts as a pincer with r9c1. There might be some purists who believe it's not an XY-Chain because r6c2 is not a bivalue.
Back to top
View user's profile Send private message
cgordon



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

PostPosted: Fri Jul 18, 2008 5:16 pm    Post subject: Reply with quote

I must be a purist because I once wrote down a posted explanation of extended xy wings (from Keith I think). It went: xy - xy = xy = xy - yz. Nothing like the above. Very confusing to a young lad trying to find his way in the big league !
Back to top
View user's profile Send private message
Asellus



Joined: 05 Jun 2007
Posts: 865
Location: Sonoma County, CA, USA

PostPosted: Fri Jul 18, 2008 7:42 pm    Post subject: Reply with quote

nataraj's "M-Wing" is only a very slight variant of an M-Wing, and one which has been discussed before on this board. As Marty notes, the basic M-Wing involves a pair of matching bivalues in remote cells. Plus, the links involved are all conjugate. That allows the M-Wing to be colored with basic Medusa.

We could notate the M-Wing as:
YX=X=XY=Y
Or, showing all of the links:
(Y=X)=X=(X=Y)=Y
While all of these links are conjugate in the basic M-Wing, the inferences are strong (in red) and weak (in green).

ravel introduced a "half M-Wing" with an actual weak (only) link in the second position:
YX-X=XY=Y

nataraj's elimination is similar except that the actual weak (only) link is now in the fourth position inside what would otherwise be the second bivalue cell:
YX=X=(X-Y)=Y

If r6c1 were a 79 bivalue instead of a 789 triple, the M-Wing would be just fine and easily recognized. But, that non-pincer bivalue doesn't actually need to be a bivalue and the wing still works.

The "classic" all-conjugate M-Wing is a simple case of basic Medusa. These two forms of "half M-Wings" are simple cases of Medusa multi-coloring. In ravel's case, with the Aa and Bb coloring indicated below the XY notation:
YX-X=XY=Y
Aa-b=Bb=B
The weak link bridges the two clusters with "ab" making "AB" the strong color pair. In nataraj's case:
YX=X=(X-Y)=Y
Aa=A=(a-b)=B

So Medusa coloring IS involved, just multi-coloring instead of basic coloring. For the record, we could have a "half half M-wing" with both weak (only) links present. But, this requires 3-cluster Medusa multi-coloring:
YX-X=(X-Y)=Y
Aa-b=(B-C)=c
The "AB" strong pair and "bc" strong pair induce the "Ac" strong pair of the pincers. But, this getting a bit far afield from that all conjugate M-Wing. Still, it is all related to Medusa and there is a common structure. That common structure is just the underlying AIC:
(Y=X)-(X)=(X-Y)=(Y)
Back to top
View user's profile Send private message Visit poster's website
nataraj



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

PostPosted: Fri Jul 18, 2008 8:58 pm    Post subject: Reply with quote

Asellus wrote:
ravel introduced a "half M-Wing" with an actual weak (only) link in the second position...

Actually, that's what inspired my broader interpretation of m-wing - and thus my suggestion to call it "r-wing" (wouldn't want to call it half-half-m-wing:) ), plus of course the fact that the letter "m", when split in half, and then again split in half, looks like an "r":


____

P.S. isn't it funny that what is in realitiy an extension of the m-wing principle should be called "half" m-wing?

edit: just a thought - if we'd follow that logoc and instead of splitting the m in half, we'd double it, and then tack on a transport ...




... wouldn't that give the ULTIMATE WEAPON ?



"My name is Bond Wing. James Bond Wing" ... Laughing Laughing Laughing
Back to top
View user's profile Send private message Send e-mail Visit poster's website
keith



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

PostPosted: Fri Jul 18, 2008 10:17 pm    Post subject: Reply with quote

Quote:
... isn't it funny that what is in realitiy an extension of the m-wing principle should be called "half" m-wing?


As I recall, ravel was calling these things "semi remote pairs" because there is a pair that makes an elimination on only one of the candidates. I called it a "half M-wing" because you can add the Medusa link only to one of the pairs.

Following on Assellus' post and Nataraj' comments, I suggest we call all of Asellus' examples "extended M-wings". This is spirit of what we mean with the term "extended XY-wings".

By the way, I really like the use of Medusa to explain how these chains work. It gives me a logical framework as to how to look for them in actual puzzles.

Keith

PS: For those who are thinking "What?" Question here is more explanation:

http://www.dailysudoku.com/sudoku/forums/viewtopic.php?t=2143
Back to top
View user's profile Send private message
Display posts from previous:   
Post new topic   Reply to topic    dailysudoku.com Forum Index -> Other puzzles All times are GMT
Goto page Previous  1, 2
Page 2 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