View previous topic :: View next topic |
Author |
Message |
guest Guest
|
Posted: Tue Aug 23, 2005 5:15 pm Post subject: unsolved sudoku in our local newspaper |
|
|
Hi there,
I try to solve a sudoku in our local newspaper but have problems. I know right now these numbers:
9 - 3 1 - 5 7 5 6
- - - - - 4 1 - 5
1 - - 7 - 6 - 8 3
2 1 8 - 4 3 - - 7
5 7 - - - 8 3 1 4
3 - - 5 7 1 8 - 2
8 3 - 4 1 2 - - 9
4 - - 8 - 7 - - 1
- - 1 - - 9 4 - 8
Any idea how to continue? |
|
Back to top |
|
|
tushar Guest
|
Posted: Tue Aug 23, 2005 5:42 pm Post subject: |
|
|
first row cannot have two 5 |
|
Back to top |
|
|
Guest
|
Posted: Wed Aug 24, 2005 7:21 am Post subject: |
|
|
tushar wrote: | first row cannot have two 5 |
Of course not. Sorry. That's the right one:
9 0 3 1 0 5 7 4 6
0 0 0 0 0 4 1 0 5
1 0 0 7 0 6 0 8 3
2 1 8 0 4 3 0 0 7
5 7 0 0 0 8 3 1 4
3 0 0 5 7 1 8 0 2
8 3 0 4 1 2 0 0 9
4 0 0 8 0 7 0 0 1
0 0 1 0 0 9 4 0 8 |
|
Back to top |
|
|
someone_somewhere
Joined: 07 Aug 2005 Posts: 275 Location: Munich
|
Posted: Wed Aug 24, 2005 8:38 am Post subject: |
|
|
Hi,
This one (after correction) looks very, very hard.
I could not find anything with the standard methods.
Has anyone tried Swordfish, XY-wing, Coloring, etc on it?
see u, |
|
Back to top |
|
|
modulated Guest
|
Posted: Wed Aug 24, 2005 9:41 am Post subject: can't be solved. |
|
|
The Drawer reports that this is an invalid puzzle (i.e., no unique solution solvable by logic -- requires guessing or trial and error). |
|
Back to top |
|
|
someone_somewhere
Joined: 07 Aug 2005 Posts: 275 Location: Munich
|
Posted: Wed Aug 24, 2005 9:50 am Post subject: |
|
|
Hi,
Correction the Draw does NOT report that it is IMPOSSIBLE.
It reports that it COULD be impossible.
At least this is how I read the very long message it is issuing!
see u, |
|
Back to top |
|
|
someone_somewhere
Joined: 07 Aug 2005 Posts: 275 Location: Munich
|
Posted: Wed Aug 24, 2005 9:51 am Post subject: |
|
|
In other words, the DRAWER is as smart as we are for that position.
We can only to try to be better.
see u, |
|
Back to top |
|
|
guest Guest
|
Posted: Wed Aug 24, 2005 11:47 am Post subject: I know the solution, but not the way |
|
|
Hi there,
now there was the solution of this sudoku in our newspaper. But I do not know how to get to the solution from the point showed you above.
9 2 3 1 8 5 7 4 6
6 8 7 2 3 4 1 9 5
1 5 4 7 9 6 2 8 3
2 1 8 6 4 3 9 5 7
5 7 6 9 2 8 3 1 4
3 4 9 5 7 1 8 6 2
8 3 5 4 1 2 6 7 9
4 9 2 8 6 7 5 3 1
7 6 1 3 5 9 4 2 8 |
|
Back to top |
|
|
someone_somewhere
Joined: 07 Aug 2005 Posts: 275 Location: Munich
|
Posted: Wed Aug 24, 2005 11:58 am Post subject: |
|
|
Hi,
You could imporve your technique from:
http://www.simes.clara.co.uk/programs/sudokutechniques.htm
But, . . . the for the one you have posted, you will have to try Swordfish, Coloring, XY-wing, Forcing Chains, Nishio, or (let's hope not Trial and Error) ...
not even X-wing will help you!
Have fun,
see u |
|
Back to top |
|
|
Kiwi Guest
|
Posted: Wed Aug 24, 2005 1:10 pm Post subject: simple way if all else fails ...... |
|
|
1. Pencil in possibilities
2. Take a pair (best choose a pair where it occurs in several places) eg ( 6,9) in r6,col 8
3. Now assume the 6 to be the correct one.
4 Quickly follow thru the whole puzzle using this assumption until you reach a dead end or it comes out.
5 If you reach a dead end it obviously must be the other one (9) and follow that thru instead.
6 Hey presto - prob solved!!
cheers from down under |
|
Back to top |
|
|
someone_somewhere
Joined: 07 Aug 2005 Posts: 275 Location: Munich
|
Posted: Wed Aug 24, 2005 1:25 pm Post subject: |
|
|
Hi,
What you are describing, is the "Trail and Error" and it should be applied when anything else did not work!!!
Keep it mind that when you are applying this method to early, you will have to make a second assumtion, and maybe this will not be enough, and you will have to use Araine wire not to get lost in the labyrinth.
Have fun & see u, |
|
Back to top |
|
|
someone_somewhere
Joined: 07 Aug 2005 Posts: 275 Location: Munich
|
Posted: Wed Aug 24, 2005 1:28 pm Post subject: |
|
|
When you do it "not too deep" it is called also "Nishio".
This you can do without any pencil !!! |
|
Back to top |
|
|
Kiwi Guest
|
Posted: Wed Aug 24, 2005 1:44 pm Post subject: |
|
|
I did say when all else fails!!!! |
|
Back to top |
|
|
David Bryant
Joined: 29 Jul 2005 Posts: 559 Location: Denver, Colorado
|
Posted: Wed Aug 24, 2005 6:48 pm Post subject: |
|
|
Anonymous wrote: | tushar wrote: | first row cannot have two 5 |
Of course not. Sorry. That's the right one:
9 0 3 1 0 5 7 4 6
0 0 0 0 0 4 1 0 5
1 0 0 7 0 6 0 8 3
2 1 8 0 4 3 0 0 7
5 7 0 0 0 8 3 1 4
3 0 0 5 7 1 8 0 2
8 3 0 4 1 2 0 0 9
4 0 0 8 0 7 0 0 1
0 0 1 0 0 9 4 0 8 |
I think your next move is to place a "9" in row 2, col 8, and a "2" in row 3, col 7.
Consider what happens if one enters a "2" in r2c8. This forces a "9" in r3c7, which in turn forces a "2" in r3c5. (There are naked pairs {2, 9} in each of the three aforementioned cells.)
Now with "2" entered in both the second and third rows a "2" must appear in r1c2 -- there's nowhere else it can fit in the first block. But that forces a "2" in r8c3, which in turn forces a "2" in r9c8. Now there are two "2"s in the same column -- this contradiction proves that one must place a "9" in r2c8.
Incidentally, there are three naked pairs {6, 9} linked together in the middle part of the puzzle -- they're in r4c4, r5c3, and r7c8. One can reach the same conclusion by testing the alternatives there as well, but I think the chain I described above is shorter.
This puzzle is fun! dcb |
|
Back to top |
|
|
squirrel
Joined: 24 Aug 2005 Posts: 3
|
Posted: Wed Aug 24, 2005 9:03 pm Post subject: |
|
|
There are also hidden pairs {4,5} in r3c2 and r3c3.
After finding these I went looking for forced chains of 1, 2, etc. Fortunately I never got to "etc" because 2 had only one possible solution. After this, everything else fell into place pretty easily.
Got lucky with this one. |
|
Back to top |
|
|
David Bryant
Joined: 29 Jul 2005 Posts: 559 Location: Denver, Colorado
|
Posted: Wed Aug 24, 2005 11:47 pm Post subject: Nishio |
|
|
After reading some stuff on other web sites, I realize that the technique I described in my previous post is called Nishio, an eponymic honor to a renowned Japanese player.
Here's a good resource if you want to understand this technique: rubylips.
dcb |
|
Back to top |
|
|
Milan Guest
|
Posted: Sat Aug 27, 2005 3:05 pm Post subject: Trial and error |
|
|
When you run out of obvious moves, look for where there are a pair in the same row. Save your moves up to that point and take a guess and pick one number of the pair, leaving the other number in the other square. Play it out and if you run into a problem, then go back and reverse your two picks. it worked for me on your puzzle. |
|
Back to top |
|
|
Milan Guest
|
Posted: Sat Aug 27, 2005 3:09 pm Post subject: |
|
|
When you run out of obvious moves, look for where there are a "matching" pair in the same row. Sorry for omitting that in my first reply. |
|
Back to top |
|
|
Guest
|
Posted: Tue Aug 30, 2005 8:14 pm Post subject: |
|
|
This puzzle has multiple solution.
One other solution is
9 2 3 1 8 5 7 4 6
6 8 7 3 9 4 1 2 5
1 4 5 7 2 6 9 8 3
2 1 8 9 4 3 6 5 7
5 7 9 2 6 8 3 1 4
3 6 4 5 7 1 8 9 2
8 3 6 4 1 2 5 7 9
4 9 5 8 3 7 2 6 1
7 2 1 6 5 9 4 3 8
|
|
Back to top |
|
|
chobans
Joined: 21 Aug 2005 Posts: 39
|
Posted: Tue Aug 30, 2005 9:23 pm Post subject: |
|
|
Column 2 has two 2s on your solution. And column 3 has two 5s on your solution. So that's not correct. |
|
Back to top |
|
|
|