A Sudoku with 2 solutions?

Juice Box

Diamond Member
Nov 7, 2003
9,615
1
0
Link removed...im an idiot

Got down to the last 4 squares, and correct me if im wrong, but you could put those 3's and 9's wherever (as long as they are opposite eachother) and theoretically come up with 2 seperate solutions to this puzzle. I thought that sudokus were famous for only having one solution for each puzzle? Did i just prove this wrong?
 

CPA

Elite Member
Nov 19, 2001
30,322
4
0
Originally posted by: Juice Box
http://pics.bbzzdd.com/users/digitalsnare/sudoku.JPG

Got down to the last 4 squares, and correct me if im wrong, but you could put those 3's and 9's wherever (as long as they are opposite eachother) and theoretically come up with 2 seperate solutions to this puzzle. I thought that sudokus were famous for only having one solution for each puzzle? Did i just prove this wrong?

Dude you have two 4s in the second column from the left.
 

thawolfman

Lifer
Dec 9, 2001
11,107
0
76
Originally posted by: CPA
Originally posted by: Juice Box
http://pics.bbzzdd.com/users/digitalsnare/sudoku.JPG

Got down to the last 4 squares, and correct me if im wrong, but you could put those 3's and 9's wherever (as long as they are opposite eachother) and theoretically come up with 2 seperate solutions to this puzzle. I thought that sudokus were famous for only having one solution for each puzzle? Did i just prove this wrong?

Dude you have two 4s in the second column from the left.

Yeah that too :D
 

CPA

Elite Member
Nov 19, 2001
30,322
4
0
btw, anyone else getting bored with Sudoku. Even the very hard/5 star ones are getting rather easy.
 

spidey07

No Lifer
Aug 4, 2000
65,469
5
76
Originally posted by: CPA
Originally posted by: Juice Box
http://pics.bbzzdd.com/users/digitalsnare/sudoku.JPG

Got down to the last 4 squares, and correct me if im wrong, but you could put those 3's and 9's wherever (as long as they are opposite eachother) and theoretically come up with 2 seperate solutions to this puzzle. I thought that sudokus were famous for only having one solution for each puzzle? Did i just prove this wrong?

Dude you have two 4s in the second column from the left.

why am I not surprised that a guy who stares at numbers all day spotted that so quickly.
;)

nice catch!
 

Juice Box

Diamond Member
Nov 7, 2003
9,615
1
0
haha holy crap, I didnt even notice that :eek:

well, I guess nevermind then

damn, thought I did well on this "very hard" one :(
 

paulney

Diamond Member
Sep 24, 2003
6,909
1
0
Yes, a Sudoku puzzle with 2 solutions is possible, but rare
There are open-source programs that use alternative alogirthms for solving it, and can come up with differnet solutions.
In fact, I wrote one myself :)
 
Oct 20, 2005
10,978
44
91
Originally posted by: paulney
Yes, a Sudoku puzzle with 2 solutions is possible, but rare
There are open-source programs that use alternative alogirthms for solving it, and can come up with differnet solutions.
In fact, I wrote one myself :)

Really? I thought all Sudoku puzzles only had 1 solution.

Give an example please!
 

paulney

Diamond Member
Sep 24, 2003
6,909
1
0
Originally posted by: Schfifty Five
Originally posted by: paulney
Yes, a Sudoku puzzle with 2 solutions is possible, but rare
There are open-source programs that use alternative alogirthms for solving it, and can come up with differnet solutions.
In fact, I wrote one myself :)

Really? I thought all Sudoku puzzles only had 1 solution.

Give an example please!

I don't have an example off the top of my head, unfortuantely (it's been a while since I dealt with this), but here's the page with the open-source project and description of various algorithms and how they can find different solutions if there are any.

http://act365.com/sudoku/
 

paulney

Diamond Member
Sep 24, 2003
6,909
1
0
Originally posted by: ViviTheMage
I wrote one too paulney, in C.

yeah, I wrote mine in C++
People who wrote it in Java sometimes can post better performance numbers due to Java's excellent native support of multidiemnsional arrays. I started out with STL vectors, and the program was hideously slow. Then I replaced everything with one-dimensional arrays, folding 2 and 3 dimensions into 1. That improved the performance significantly.

There are still areas to improve, but when I reached 0.01 - 0.05 times for the hardest puzzles on 1Ghz Athlon, I was satisfied and completed the project.