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 

Dec 21 VH
Goto page Previous  1, 2
 
Post new topic   Reply to topic    dailysudoku.com Forum Index -> Daily Sudoku puzzles
View previous topic :: View next topic  
Author Message
Asellus



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

PostPosted: Sat Dec 22, 2007 11:20 pm    Post subject: Reply with quote

Craig,

I believe there is a little bit of working at cross purposes going on here. Keith's interest seems to be in finding useful patterns involving chains of repeating identical bivalue cells (remote naked pairs, extended XY Wings, etc.). And that's all well and good.

However, if your interest is in pincer coloring, then I suggest it is better to forget about bivalue chains and focus on the coloring. Coloring from a pincer is just a matter of using any sequences of end-to-end strong links that are available. It doesn't matter if the cells along the way are bivalues.

So, not only does that {48} cell with the transported pincer in my example not need to be {28}, it doesn't even need to be a bivalue. It could be {24789} and still work, provided that <8> is strongly linked to the <8> at r1c5.

Another nice thing about thinking in alternating colors (red-green) along the "color chain" is that you don't have to remember if it is odd numbers or even numbers of links, nodes, whatever. Just keep track of the colors.

Pincers can also be transported by means of one or more ERs, if you are comfortable with that idea.

You should only transport/color extend one pincer (of a pincer pair) and perform the eliminations against the other pincer. (While it is possible to color from both pincers at the same time, the coloring rules are a bit more involved. So, I wouldn't recommend it unless you are confident you understand the rules for such things.)

By the way... I wouldn't abandon the ideas of chains so completely. They can be useful and aren't really so difficult if, in my opinion, you learn the general rules for chains rather than trying to memorize set chain patterns. But, that's for another day, I'm sure.
Back to top
View user's profile Send private message Visit poster's website
alanr555



Joined: 01 Aug 2005
Posts: 198
Location: Bideford Devon EX39

PostPosted: Sun Dec 23, 2007 1:21 am    Post subject: Reply with quote

Asellus wrote:

Not only does that {48} cell not need to be {28}, it could be {24789}
and still work, provided that <8> is strongly linked to the <8> at r1c5.

By the way... I wouldn't abandon the ideas of chains so completely. They can be useful and aren't really so difficult if, in my opinion, you learn the general rules for chains rather than trying to memorize set patterns.


One of the advantages of using Mandatory Pairs is that ALL the links
resulting in M/P marks are ALWAYS STRONG in nature - in so far
as they relate to single digits.

In the example r1c5 and r3c6 are marked as a M/Pair on 8 and so
the task of checking the status of the link is obviated. Indeed, the
link draws attention to itself without having to check any patterns.

It is refreshing to read the second part of the quote - learning the
general rules rather than memorising set patterns. However, that
does not avoid the need for the skill of applying those same general
rules in order to SPOT the existence of such patterns.

Personally, I do not know the general rules. Is there a good place
at which to find out about them?

Circumstances in my life led to my leaving this forum in April 2006
and I am only just returning. My recollection is that "colouring" is
another way of viewing implication chains - using the same
objectives of contradiction (ie a cell must be both red AND green)
but maybe there is more to it than that. If the two are connected,
then my tip about triples and diagonals being polarity reversers
may be useful in "colouring" as well as in implication chains.

However, I believe that some solvers prefer to identify patterns
(various wings or types of fish) as preferable to chains. Certainly
they have a greater "neatness" about them and would appeal to
the logical, mathematical mind - but simple souls such as myself
who abandoned advanced mathematics decades ago may still
find chains based on simple logic principles far easier to use.
Back to top
View user's profile Send private message
Asellus



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

PostPosted: Sun Dec 23, 2007 2:32 am    Post subject: Reply with quote

Alan,

This thread doesn't seem the place for lots of theoretical discussion; but, I'll try to give a brief reply to your points as I see them.

First, I'm all in favor of pattern solutions such as the "fish" methods and XY-Wings, Kites, etc., exactly because there is nothing "mathematical" about them. Most of these techniques involve patterns that are easy to recognize and use even if one does not fully understand how they work. They are by far the most popular and widely used methods by ordinary folks who have no interest in learning any "theory."

My response to Craig about patterns applies only to implication chains. In that case, I believe it is easier to learn and apply the basic rules for AICs than it is to try to memorize various patterns for chains (such as "xz-xy-yz=yz=yz").

As for the use of implication chains for solving puzzles, I believe it makes sense to try to exhaust the pattern techniques first, using chains as a "last resort" for the most difficult puzzles. This also means that learning the pattern techniques is a higher priority.

"Coloring" is a general term for a variety of related methods. I'd say that your statement is backwards: it seems more accurate to me to say that implication chains are another way to view coloring. All coloring can be expressed as implication chains (as can virtually all of the sudoku techniques) but a great many implication chains cannot be captured using coloring.

Pencil solvers tend to avoid coloring because of the problem of making color marks. But many simple cases can be done in ones head without marking. Also, using blank grids to mark the locations of all cells for a single candidate digit is useful both for coloring and other pattern methods. Multi-digit "Medusa" coloring poses the biggest challenges to pencil solvers. But, it is also the most powerful of the coloring techniques.

I don't have any specific links to provide. However, searching sudopedia.org or the solving guide at sudocue.net are two possible places to begin looking.
Back to top
View user's profile Send private message Visit poster's website
Marty R.



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

PostPosted: Sun Dec 23, 2007 4:38 am    Post subject: Reply with quote

Quote:
Pencil solvers tend to avoid coloring because of the problem of making color marks.

Not necessarily true. All it takes is a few blank grids for testing coloring, chains, Medusa and the like.
Back to top
View user's profile Send private message
mjchamorro



Joined: 29 Jan 2008
Posts: 8

PostPosted: Tue Jan 29, 2008 1:36 pm    Post subject: sorry but 21-dec-2007 sudoku HEEEEELP!!! Reply with quote

Hello Everyone:

This is where I got stuck in daily sudoku from 21-Dec-2007.

3 248 1|45 258 6|45 9 7
6 24 7|1459 259 124|345 345 8
5 48 9|3 7 48|6 2 1
-------------------------------------
9 135 35|7 26 24|8 14 56
4 6 2|8 1 5|9 7 3
7 15 8|469 69 3|2 14 56
------------------------------------
1 9 356|56 4 78|357 3568 2
2 35 4|156 3568 178|357 3568 9
8 7 356|2 356 9|1 356 4

The thing is that when I ask for a hint in the draw/play section on dailysudoku, I'm given this hint:
r4c5=6

My question is WHY is that the next logical step?

thank you for any help or advise.

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



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

PostPosted: Tue Jan 29, 2008 2:32 pm    Post subject: Reply with quote

Hi, and welcome

Code:
3 248 1  | 45   258  6   | 45  9    7
6 24  7  | 1459 259  124 | 345 345  8
5 48  9  | 3    7    48  | 6   2    1
-------------------------------------
9 135 35 | 7    26   24  | 8   14   56
4 6   2  | 8    1    5   | 9   7 3
7 15  8  | 469  69   3   | 2   14   56
--------------------------------------
1 9  356 | 56   4    78  | 357 3568 2
2 35 4   | 156  3568 178 | 357 3568 9
8 7  356 | 2    356  9   | 1   356  4


The first step is to tidy the top row to allow for the pairs (45) and (28):

Code:
3 28  1  | 45   28   6   | 45  9    7
6 24  7  | 1459 259  124 | 345 345  8
5 48  9  | 3    7    48  | 6   2    1
-------------------------------------
9 135 35 | 7    26   24  | 8   14   56
4 6   2  | 8    1    5   | 9   7 3
7 15  8  | 469  69   3   | 2   14   56
--------------------------------------
1 9  356 | 56   4    78  | 357 3568 2
2 35 4   | 156  3568 178 | 357 3568 9
8 7  356 | 2    356  9   | 1   356  4


Now there is an xy-wing with pivot r3c6 and pincers r1c5 and r4c6.

If the pivot, r3c6, contains 8, r1c5 contains 2; if it contains 4, r4c6 contains 2. Whichever applies, 2 may be eliminated from r4c5. (In fact it may be eliminated from r2c6 as well).

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



Joined: 29 Jan 2008
Posts: 8

PostPosted: Tue Jan 29, 2008 2:43 pm    Post subject: sorry but 21-dec-2007 sudoku HEEEEELP!!! Reply with quote

Thank you steve, now i just have to find out what a xy-wing is. I knew about the x-wing and the swordfish, but not xy-wing...

anyway thank you very much, really, now there is light at the end of the tunner Wink
Back to top
View user's profile Send private message
Steve R



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

PostPosted: Tue Jan 29, 2008 3:06 pm    Post subject: Reply with quote

Glad to be of help.

You may find SadMan’s explanation of the xy-wing helpful, as I did.

Steve
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 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