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 

Oct 16 puzzle question

 
Post new topic   Reply to topic    dailysudoku.com Forum Index -> Daily Sudoku puzzles
View previous topic :: View next topic  
Author Message
Louise56



Joined: 21 Sep 2005
Posts: 94
Location: El Cajon, California USA

PostPosted: Sun Oct 16, 2005 3:25 am    Post subject: Oct 16 puzzle question Reply with quote

I'm trying to solve this one and have a question for one of you experts. These are my bottom three, 3x3 boxes

-/-/- 359/2/1 389/-/6
9/6/- 35/7/358 -/-/4
-/-/1 4/6/3589 389/7/238

My question is with the 9's. In row 7 I have a 9 in r7c4 and r7c7. In row 9 I have a 9 in r9c6 and r9c7. These are the only 9's in rows 6 and 9. It looks like an X-wing set up, but in the middle box the 9's are not in the same column. But since they are in the same 3x3 box, it seems like I could treat this like an x-wing, but I don't know if I can eliminate the remaining 9's in column 4 or column 6. Does this setup have a name? Thanks!
Back to top
View user's profile Send private message
someone_somewhere



Joined: 07 Aug 2005
Posts: 275
Location: Munich

PostPosted: Sun Oct 16, 2005 5:30 am    Post subject: Reply with quote

Hi Louise,

That was real a nice question!
I was working a lot on X-wing.
You made a nice discovery. The X-wing that has the same number at intersection of 2 rows and 2 columns (the name cames from the airplains flying in this formation) and having this same number ONLY 2 times in the rows or in the columns - permits us to eliminate the rest of this number from the 2 columns respective 2 rows (if there are any additional such ones).

I think that one of the "generalization" of this X-wing is the technique called "intersectioon of four groups". It is the combination of 4 intersection rows, columns, 3x3 blocks.

* If we have 2 rows and 2 columns - then we speak about the "X-wing".

The other cases are:

* 2 rows, 1 column & 1 block

* 1 row, 2 columns & 1 block

* 2 rows, 2 blocks

* 2 columns, 2 blocks

Let's take an example:

X-X
6=6======
X-X
=== the sign "=" means no "6"
==6
6==
X-X
X-X
X-X

We have here number 6:
- in row 2 exact twice
- in column 1 exact twice
- in column 3 exact twice
- in 3x3 block (4,1) - (6,3) exact twice
We have the same number 6 at the intersection of four groups:
one row, 2 columns and 1 3X3 block.

We can quickly deduce that from all places marked with "X" the "6" can be eliminated!

For the rest of the cases, I let you draw your own diagrams.

see u,
Back to top
View user's profile Send private message
zigzagpython



Joined: 27 Sep 2005
Posts: 4

PostPosted: Sun Oct 16, 2005 8:02 am    Post subject: Reply with quote

still not understand!!!!
I spent nearly an hour on the sheet
only found 1 in r6c2, 9 in r1c9 and most easily 6 in r8c2
then move no step

help me Mater
Back to top
View user's profile Send private message
someone_somewhere



Joined: 07 Aug 2005
Posts: 275
Location: Munich

PostPosted: Sun Oct 16, 2005 8:10 am    Post subject: Reply with quote

Hi,

Please post me the initial position.

see u,
Back to top
View user's profile Send private message
someone_somewhere



Joined: 07 Aug 2005
Posts: 275
Location: Munich

PostPosted: Sun Oct 16, 2005 8:18 am    Post subject: Reply with quote

Hi,

Sorry, now I see which puzzle you have problem with.

080016700
600030005
100890000
320000500
006000200
008000097
000021006
900070004
001460070

6 in r8c2 - Unique Horizontal
4 not in r4c6, it is in r4c5 or r5c5 or r6c5 (Column on 3x3 Block interaction)
4 not in r5c6, it is in r4c5 or r5c5 or r6c5 (Column on 3x3 Block interaction)
4 not in r6c6, it is in r4c5 or r5c5 or r6c5 (Column on 3x3 Block interaction)
5 not in r5c4, it is in r5c5 or r6c5 (Column on 3x3 Block interaction)
5 not in r5c6, it is in r5c5 or r6c5 (Column on 3x3 Block interaction)
5 not in r6c4, it is in r5c5 or r6c5 (Column on 3x3 Block interaction)
5 not in r6c6, it is in r5c5 or r6c5 (Column on 3x3 Block interaction)
8 not in r4c6, it is in r4c5 or r5c5 (Column on 3x3 Block interaction)
8 not in r5c6, it is in r4c5 or r5c5 (Column on 3x3 Block interaction)
1 not in r4c8, it is in r4c9 or r5c9 (Column on 3x3 Block interaction)
1 not in r5c8, it is in r4c9 or r5c9 (Column on 3x3 Block interaction)
1 not in r6c7, it is in r4c9 or r5c9 (Column on 3x3 Block interaction)
2 not in r2c8, Hidden Pair 1 8 in r2c7 and r2c8 (in Row)
4 not in r2c7, Hidden Pair 1 8 in r2c7 and r2c8 (in Row)
4 not in r2c8, Hidden Pair 1 8 in r2c7 and r2c8 (in Row)
9 not in r2c7, Hidden Pair 1 8 in r2c7 and r2c8 (in Row)
4 not in r6c2, Nacked Pair 4 5 in r6c1 and r6c5 (same Row)
5 not in r6c2, Nacked Pair 4 5 in r6c1 and r6c5 (same Row)
4 not in r6c7, Nacked Pair 4 5 in r6c1 and r6c5 (same Row)
1 in r6c2 - Sole Candidate

and from here on, you can swim alone...

see u,
Back to top
View user's profile Send private message
zigzagpython



Joined: 27 Sep 2005
Posts: 4

PostPosted: Sun Oct 16, 2005 9:44 am    Post subject: Reply with quote

thx for yr tips
'5' in either r4c5 or r5c5
'1' in eihter r4c9 or r5c9
they are important tips and main numbers blocking my going

need more training la
but next week will be very busy.
loads work and have a test from my part time studying in the coming weeks:cry:
got little bit injury from yesterday's rugby match.

see you master
Back to top
View user's profile Send private message
David Bryant



Joined: 29 Jul 2005
Posts: 559
Location: Denver, Colorado

PostPosted: Sun Oct 16, 2005 2:52 pm    Post subject: Here's another tip Reply with quote

zigzagpython wrote:
thx for yr tips
'5' in either r4c5 or r5c5
'1' in eihter r4c9 or r5c9
they are important tips and main numbers blocking my going


This is a _very_ tough puzzle. Here's what I found that let me solve it.

In column 9, "1" can only appear in r4c9 or r5c9. So we can eliminate "1" from r4c8, r5c8, & r6c7.

Because of the {4, 5} pair in row 6, the "4" in column 8 must lie in the middle right 3x3 box -- that is, in r4c8 or r5c8. Now there is an "X-Wing" on "4" -- in r2c6, r3c6, r2c7, & r3c7. So we can eliminate other possible "4"s in rows 2 & 3.

That's probably enough of a hint for now. dcb Smile
Back to top
View user's profile Send private message Send e-mail Visit poster's website
junior
Guest





PostPosted: Sun Oct 16, 2005 4:59 pm    Post subject: Re: Here's another tip Reply with quote

David Bryant wrote:
zigzagpython wrote:
thx for yr tips
'5' in either r4c5 or r5c5
'1' in eihter r4c9 or r5c9
they are important tips and main numbers blocking my going


This is a _very_ tough puzzle. Here's what I found that let me solve it.

In column 9, "1" can only appear in r4c9 or r5c9. So we can eliminate "1" from r4c8, r5c8, & r6c7.

Because of the {4, 5} pair in row 6, the "4" in column 8 must lie in the middle right 3x3 box -- that is, in r4c8 or r5c8. Now there is an "X-Wing" on "4" -- in r2c6, r3c6, r2c7, & r3c7. So we can eliminate other possible "4"s in rows 2 & 3.

That's probably enough of a hint for now. dcb Smile

I can't see the X-wing on "4". I only see the naked couple [1,8] in r2c6 and r2c7 and the naked couple [4,5] in r6c1 and r6c5, so 4 becomes the sole candidate for r3c7. Am I wrong?
Back to top
David Bryant



Joined: 29 Jul 2005
Posts: 559
Location: Denver, Colorado

PostPosted: Sun Oct 16, 2005 5:28 pm    Post subject: Re: Here's another tip Reply with quote

Junior wrote:
I can't see the X-wing on "4". I only see the naked couple [1,8] in r2c6 and r2c7 and the naked couple [4,5] in r6c1 and r6c5, so 4 becomes the sole candidate for r3c7. Am I wrong?

No, you're not wrong. "4" in r3c7 is correct. But I think you may be forgetting the {4, 5, 8} triplet that must appear in column 5, in the middle center 3x3 box. Since there's a "4" at r9c4, that triplet tells you that the "4" in column 6 must go either at r2c6 or at r3c6 -- hence at r2c6, since you already have the "4" at r3c7, somehow. dcb
Back to top
View user's profile Send private message Send e-mail Visit poster's website
Louise56



Joined: 21 Sep 2005
Posts: 94
Location: El Cajon, California USA

PostPosted: Sun Oct 16, 2005 6:18 pm    Post subject: Reply with quote

Thanks Someone for an excellent explanation. This was my favorite puzzle to date! I'm always looking for patterns and shortcuts.
Back to top
View user's profile Send private message
scorpio56xxl



Joined: 16 Oct 2005
Posts: 7
Location: Bonn, Germany

PostPosted: Sun Oct 16, 2005 7:14 pm    Post subject: Reply with quote

Hi Louise,
I have the same problem as you but I come to another conclusion. Since
in r3c5 there is a nine so there could not be one in r4c5 - r6c5. So the nines in the middle box goes in r4c4/r5c4 and r4c6/r5c6. Box number 8 and 9 the last one contains possible nines as you described. So there couldn't be a nine in r9c9. Since there is a nine in box 7 which intersects r9, box 6 which intersects c9 and an intersection between the boxes 8 and 9 the nine for the last column must be placed in r1c9. But I could't see that you can eleminate the nines in the middle box. I 'm looking forward to getting your answer. Thanks in advance.
Back to top
View user's profile Send private message Send e-mail
Louise56



Joined: 21 Sep 2005
Posts: 94
Location: El Cajon, California USA

PostPosted: Mon Oct 17, 2005 12:22 am    Post subject: Reply with quote

Hi Scorpio,

I don't remember all the steps I took to solve the puzzle. I do remember finding a 7/9 pair in r4c6 and r5c6 and that helped me to elimate the 9 in r9c6. That put a 9 in r7c4 and a 9 in r9c7. I hope that helps!
Back to top
View user's profile Send private message
David Bryant



Joined: 29 Jul 2005
Posts: 559
Location: Denver, Colorado

PostPosted: Mon Oct 17, 2005 7:29 pm    Post subject: "Nishio" worked on this puzzle Reply with quote

Louise56 wrote:
Thanks Someone for an excellent explanation. This was my favorite puzzle to date! I'm always looking for patterns and shortcuts.

I thought this was an excellent puzzle. Interestingly, I spotted a "Nishio" pattern at a very early stage, and I'm wondering if anyone else noticed it.

After entering a "6" at r8c2 and a "1" at r6c2 the puzzle looked like this to me:
Code:

  .    8    .    .    1    6    7    .    9
  6    .    .    .    3    4?   4?   .    5
  1    .    .    8    9    4?   4?   6   1/8
  3    2    .    .   4/8  7/9   5    .    .
  .    .    6    .    .    .    2    .    .
 4/5   1    8    .   4/5  2/3  3/6   9    7
  .    .    .    .    2    1    .    .    6
  9    6    .   3/5   7    .    .    .    4
  .    .    1    4    6    .    .    7    .

I'm not very good at spotting "hidden pairs." But because of the {4, 5, 8} triplet in r4c5, r5c5, & r6c5, plus the {3, 6} pair at r6c7, it was apparent that the "4"s in columns 6 and 7 had to appear in the little square marked 4? above. This meant I could eliminate the possibility of a "4" in r2c2, r2c3, r2c4, r3c2, & r3c3. After doing that I found a triplet {2, 7, 9} in row 2, and that uncovered the "hidden pair" in the top right 3x3 box. It also let me enter a few more numbers in the top right region of the puzzle. Now my grid looked like this:
Code:

 2/4/5    8      .      .      1      6      7     2/3     9
   6     7/9   2/7/9   2/7     3      4     1/8    1/8     5
   1      .      .      8      9      .      4      6     2/3
   3      2      .      .     4/8    7/9     5      .     1/8
   .      .      6      .    4/5/8  3/7/9    2      .    1/3/8
  4/5     1      8      .     4/5    2/3    3/6     9      7
   .      .      .      .      2      1      .      .      6
   9      6      2?    3/5     7      .      .      2?     4
 2/5/8    .      1      4      6      .      .      7    2/3/8

This was when I spotted the "Nishio" pattern. Clearly there are only two places to put a "2" in column 1, and in column 9. You have to look more closely, but it's also clear that there are only two places to put a "2" in the top right, the bottom left, and the bottom right 3x3 boxes. The chain of connected "2"s makes the "Nishio" pattern.

Consider what would happen if r1c1 = 2. Then r9c1 is not 2, and r8c3 = 2. But then r8c8 is not 2, and r1c8 = 2. But this is impossible -- we can't have two "2"s in the first row. So we must place "2" at r9c1, r8c8, and r3c9.

I've entered another puzzle that uses "Nishio" in the other puzzles forum, in case you're interested. dcb
Back to top
View user's profile Send private message Send e-mail Visit poster's website
alanr555



Joined: 01 Aug 2005
Posts: 194
Location: Bideford Devon EX39

PostPosted: Mon Oct 17, 2005 9:14 pm    Post subject: Re: "Nishio" worked on this puzzle Reply with quote

Code:

> Interestingly, I spotted a "Nishio" pattern at a very early stage, and I'm
> wondering if anyone else noticed it.

 2/4/5    8      .      .      1      6      7     2/3     9
   6     7/9   2/7/9   2/7     3      4     1/8    1/8     5
   1      .      .      8      9      .      4      6     2/3
   3      2      .      .     4/8    7/9     5      .     1/8
   .      .      6      .    4/5/8  3/7/9    2      .    1/3/8
  4/5     1      8      .     4/5    2/3    3/6     9      7
   .      .      .      .      2      1      .      .      6
   9      6      2?    3/5     7      .      .      2?     4
 2/5/8    .      1      4      6      .      .      7    2/3/8

> This was when I spotted the "Nishio" pattern.

Thank you for this explanation of "Nishio".
It would seem to add significance to my promotion of Mandatory Pairs
if I have understood the technique correctly.

Essentially I understand a "Nishio" to be a CLOSED chain of Mandatory
Pairs where the pairs can be in ANY region or line (row/column). The
values MUST toggle along the chain being alternately the value and not
the value. Thus the determination of ANY value within the chain will
immediately resolve half the members of the chain - BRILLIANT!

I solved this puzzle in just under an hour elapsed time after I retired to
bed about 00.30 local time on Monday morning. Often I fall asleep over
them but managed to complete this one. It was not amenable to a full
solution using Mandatory Pairs (cf my submission on that method) and
so I switched to candidate profiles after squeezing the information out
of the Mandatory Pairs data. However, I left such data present and found
that having done so was very useful later. I did not need any advanced
techniques other than finding congruent groups. (eg row 2 turned into
1/8 and 2/4/7/9 and row 6 into 1/4/5 and 2/3/6 and column 6 into 5/8
and 2/3/4/7/9). Success then depends upon finding an "entry" into the
congruent group to isolate one or more entries. Counting within a line
or a region is, of course, one form of entry. In this regard the quality
of "mutual reception" in Mandatory pairs makes "counting" much easier.
The rule about eliminating stragglers when all the occurrences of a
digit within a line are in one region (and vice versa) does not need any
searching with Mandatory Pairs as the pairs generally highlight such
anyway. This I find that M/P is an excellent prelude to the more detail
process of candidate profiling - making the latter MUCH easier in terms
of the tedious compilation involved.

Now that I have a comprehension of the Nishio, I can add it to the M/P
techniques and, hopefully, reduce solution times slightly. So thank you,
again, for clarifying the technique.

Alan Rayner BS23 2QT
[/code]
Back to top
View user's profile Send private message
Blackfoot



Joined: 18 Oct 2005
Posts: 1

PostPosted: Tue Oct 18, 2005 2:49 pm    Post subject: Re: "Nishio" worked on this puzzle Reply with quote

David Bryant wrote:

Code:

  .    8    .    .    1    6    7    .    9
  6    .    .    .    3    4?   4?   .    5
  1    .    .    8    9    4?   4?   6   1/8
  3    2    .    .   4/8  7/9   5    .    .
  .    .    6    .    .    .    2    .    .
 4/5   1    8    .   4/5  2/3  3/6   9    7
  .    .    .    .    2    1    .    .    6
  9    6    .   3/5   7    .    .    .    4
  .    .    1    4    6    .    .    7    .



I'm new to all this so forgive me.

I just don't see how you get the 1 in R6C2.

I see a 1 at R4C9 or R5C9
I see a 1 at R5C2 or R6C2
But I also see a 1 at R4C4,R5C4,R6C4.

The only this I can think is that I've missed something in R6C4.

Thanks in advance
Back to top
View user's profile Send private message
David Bryant



Joined: 29 Jul 2005
Posts: 559
Location: Denver, Colorado

PostPosted: Tue Oct 18, 2005 3:28 pm    Post subject: How to place the "1" at r6c2 Reply with quote

There's no need to apologize -- all of us were new to this at some point. Smile

Concentrate on row 6. With the {4, 5} pair placed in r6c1 and r6c5, there are only four numbers left -- {1, 2, 3, 6} -- to fill in the four open positions in the row (that is, the values {1, 2, 3, 6} must occupy the cells r6c2, r6c4, r6c6, & r6c7 in some order).

Now look at r6c2. It has to be a "1", a "2", a "3", or a "6." It clearly can't be either a "2" or a "6" because those values already appear in column 2. And it can't be a "3", either, because there's already a "3" in the middle left 3x3 box. So it has to be a "1." dcb

PS I have some trouble with this idea still -- it's easy to eliminate possibilities by looking up and down columns, or left and right across rows. It's a bit harder to remember to look all around the 3x3 box for additional numbers that can safely be eliminated from a list of possibilities. One set of eye motions comes naturally -- I have to work at the other one. Maybe crossed eyes would help!
Back to top
View user's profile Send private message Send e-mail Visit poster's website
someone_somewhere



Joined: 07 Aug 2005
Posts: 275
Location: Munich

PostPosted: Tue Oct 18, 2005 3:53 pm    Post subject: Reply with quote

Hi,

just take a look again at the steps that I
Posted: Sun Oct 16, 2005 9:18 am

that should answer your question about placing the "1".

see u,
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
Page 1 of 1

 
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