TNU

Advanced methods and approaches for solving Sudoku puzzles

TNU

Postby Neunmalneun » Tue Mar 31, 2009 5:30 pm

There is a very easy but powerful new technique (at least I hope it's new, please correct me if I am wrong). Every grid consists of 54 units of three numbers, 27 vertically and 27 horizontally. Look at the following grid, and you will see what I mean.



*-----------*
|532|168|794|
|497|352|186|
|186|974|532|
|---+---+---|
|975|216|843|
|843|795|621|
|261|483|975|
|---+---+---|
|359|821|467|
|628|547|319|
|714|639|258|
*-----------*




The first horizontal "three numbers units" (TNU's) are <532>, <168> and <794>, the first vertical TNU's are <541>, <982> and <367>. Since in no grid there are two identical TNU's you can easily deduct from the solved TNU's in your grid which of the candidates are not possible. Example:
Code: Select all



 
 *-----------------------------------------------------------*
 | 2     1     9     | 8     27    5     | 6     4     3     |
 | 8     7     6     | 1     4     3     | 2     9     5     |
 | 25    4     3     | 6     2     9     | 1     8     7     |
 |-------------------+-------------------+-------------------|
 | 4567  56    45    | 3     1     46    | 9     2     8     |
 | 456   3     2     | 45    9     468   | 7     1     46    |
 | 1     9     48    | 7     68    2     | 5     3     46    |
 |-------------------+-------------------+-------------------|
 | 459   25    145   | 2459  3     7     | 8     6     129   |
 | 3     256   7     | 259   568   68    | 4     5     129   |
 | 4569  8     45    | 2459  56    1     | 3     57    29    |
 *-----------------------------------------------------------*

As you see in C9R123 there is a TNU <357>, and in R8C123 you have a <3?7> combination. That means R8C2 <> 5.

Best wishes from Berlin

Neunmalneun
 
Neunmalneun
 
Posts: 52
Joined: 22 December 2005

Postby coloin » Tue Mar 31, 2009 5:52 pm

Not quite but perhaps has a place !

Code: Select all
+---+---+---+
|...|...|..1|
|...|...|..2|
|...|...|..3|
+---+---+---+
|...|...|..4|
|...|...|..5|
|...|...|..6|
+---+---+---+
|...|...|...|
|...|...|...|
|123|4..|...|
+---+---+---+



r9c56 - cant be 5 or 6.

there is a hidden quad in r78c78

C
coloin
 
Posts: 2385
Joined: 05 May 2005
Location: Devon

Postby 999_Springs » Tue Mar 31, 2009 6:09 pm

Uh, r8c2<>5 because r8c8=5.
999_Springs
 
Posts: 591
Joined: 27 January 2007
Location: In the toilet, flushing down springs, one by one.

Re: TNU

Postby eleven » Wed Apr 01, 2009 1:11 pm

Neunmalneun wrote:Since in no grid there are two identical TNU's ...
Your post was too early, should have been dated April 1st (which it was in Germany, i suppose):)
In your first grid there are some identical horizontal TNU's (like 532, 186, 843), in the second one you will have 219 both in r1c123 and r789c9 etc.

Added: Is this a server problem ?
The time now is Wed Apr 01, 2009 5:27 am, but i have Thu Apr 2 10:05:58 UTC 2009
eleven
 
Posts: 3097
Joined: 10 February 2008

Re: TNU

Postby gsf » Wed Apr 01, 2009 3:22 pm

eleven wrote:
Neunmalneun wrote:Since in no grid there are two identical TNU's ...
Your post was too early, should have been dated April 1st (which it was in Germany, i suppose):)
In your first grid there are some identical horizontal TNU's (like 532, 186, 843), in the second one you will have 219 both in r1c123 and r789c9 etc.

Added: Is this a server problem ?
The time now is Wed Apr 01, 2009 5:27 am, but i have Thu Apr 2 10:05:58 UTC 2009

the forum clock slips by a noticable amount each day
the patterns game referee keeps track of it
its currently about 28h31m slow
anyone care to try notifying the admin(s)?
I've given up
www.ntp.org should be in that notice
and, even though the forum says so, this is not an april fools post
gsf
2014 Supporter
 
Posts: 7306
Joined: 21 September 2005
Location: NJ USA

Postby 999_Springs » Wed Apr 01, 2009 5:10 pm

gsf wrote:its currently about 28h31m slow

Really? It only seems to me to be about 20h32m slow, after adjusting for daylight savings and all that rubbish...
999_Springs
 
Posts: 591
Joined: 27 January 2007
Location: In the toilet, flushing down springs, one by one.

slow

Postby Pat » Wed Apr 01, 2009 5:16 pm

gsf wrote:the forum clock slips by a noticable amount each day

it's currently about 28h31m slow


currently circa 20 hr 35 min slow
User avatar
Pat
 
Posts: 4056
Joined: 18 July 2005

Re: slow

Postby gsf » Wed Apr 01, 2009 5:31 pm

Pat wrote:
gsf wrote:the forum clock slips by a noticable amount each day

it's currently about 28h31m slow


currently circa 20 hr 35 min slow

sorry, the number I gave was the offset from the default forum tz (gmt-0800) to gmt, so it was 8 hours off
thanks
gsf
2014 Supporter
 
Posts: 7306
Joined: 21 September 2005
Location: NJ USA

Postby Pat » Wed Apr 01, 2009 5:39 pm

but 999_Springs beat me by 6 minutes
( and i was so distracted i didn't even notice my comment was no longer necessary )
User avatar
Pat
 
Posts: 4056
Joined: 18 July 2005

Postby Neunmalneun » Wed Apr 01, 2009 7:26 pm

Well I have to admit that my concept is complete nonsense. And I have also to admit that I was too stupid to post it on time (at least it was April 1 here then). Next year I will do better.
Neunmalneun
 
Posts: 52
Joined: 22 December 2005

Postby eleven » Wed Apr 01, 2009 9:05 pm

999_Springs wrote:... about 20h32m slow...
This shows that you cant trust such kind of "publishing date" on the internet:(
eleven
 
Posts: 3097
Joined: 10 February 2008


Return to Advanced solving techniques