## Question

Everything about Sudoku that doesn't fit in one of the other sections

### Question

Can someone find a puzzle plus candidates list that we cant make any progress
using the following way
1)I can use all the elementary technics in any time
2)I can guess and continue using elementary technics if I get contradiction or I solve thats great I got a progress
3)if I get stuck I don't remove the candidate but move foreword to the next candidate or the next cell if I finished the candidates for a cell.
bennys

Posts: 156
Joined: 28 September 2005

No, any sudoku with a unique solution can be solved using that method. In fact, if you're going to use trial and error, then technically you don' even have to use elementary techniques. You could just make a huge series of nested guesses until you find a solution that works - this is how a simple computer program would solve them, though obviously it takes too much number-crunching to be practicable for a human to do it.
PaulIQ164

Posts: 533
Joined: 16 July 2005

Maybe I was not clear but I am not allowing nested guesses here
And that the main point.
bennys

Posts: 156
Joined: 28 September 2005

All sudoku puzzles with only one puzzle can be solved using logical techniques, however you can get to a point where you need to use a technique such as colouring which is sort of like guessing
Pi

Posts: 389
Joined: 27 May 2005

You are not answering the question
bennys

Posts: 156
Joined: 28 September 2005

Im still not sure what you mean, maybe the following?
You use elementary techniques until you get stuck.
Then you can try each remaining candidate - using only elementary techniques again - until you either get a contradiction (then you eliminate it) or solve it (then you are done) or get stuck again (then proceed with the next candidate).
If you mean this, there is no known puzzle you could not solve this way, as far as i know (at least this is true for the "hardest" puzzles posted here)
Wolfgang

Posts: 208
Joined: 22 June 2005

In this post I am not asking about totally solving I am talking just on one step (which is cell placement or removal of candidate)
I am looking for example that the way that I described will NOT work.
I assume there is example but I didn’t see any.
bennys

Posts: 156
Joined: 28 September 2005

If a move cannot be made based upon logic then the puzzle has more than one solution

You may not be able to use elementery tactics but if you look hard enough there is always a move
Pi

Posts: 389
Joined: 27 May 2005

I am not talking about logic I am talking about specific way
bennys

Posts: 156
Joined: 28 September 2005

There are lots of specific ways of removing candidates, the most common being:

Naked Pair
Nidden Pair
Naked tripple
Hidden Tripple
Locked candidate
X-Wing
Swordfish
Chain
You will find puzzles in which you reach a point where you can't use any of these.
Pi

Posts: 389
Joined: 27 May 2005

It looks to me that you dont understand the question please read it again
bennys

Posts: 156
Joined: 28 September 2005

Pi

Posts: 389
Joined: 27 May 2005

bennys wrote:It looks to me that you dont understand the question please read it again

Perhaps you should learn how to ask a question?

Point 1 of your original post: unclear + dubious
Point 2: a bit dubious but making more sense. You are talking about Trial and Error.
Point 3: completly irrelevant/redundant.

You never defuined something as 'elementary technics'. What do you understand on those two words? ...?

If your question is: does a puzzle exists where I need to use trial and error and then I can continue without guessing and solve it completly (without using Trial and Error) then the answer is both yes and no.

(The no-part: because there is always another way to find the move. (Forcing chains to name one.)
Animator

Posts: 469
Joined: 08 April 2005

I think the question he means is:

Can any puzzle be solved using only 'elementary tactics' (whichever set of tactics he considers them to be, my guess is either just singles or just up to Pappocom-Hard tactics) and single guesses, that is, never having to make a guess within a 'branch' of a guess?
PaulIQ164

Posts: 533
Joined: 16 July 2005

Basically the way that PaulIQ164 put it is equivalent to my question.
Just to make it simpler In a sense I want to get the puzzle in a specific stage where I will not be able to make any progress using that method.
Yes Pappocom-Hard tactics sound OK.
bennys

Posts: 156
Joined: 28 September 2005

Next