## Help with Fiendish puzzle

Advanced methods and approaches for solving Sudoku puzzles

### Help with Fiendish puzzle

Can anyone give me a few pointers on this puzzle? I can't seem to find any X-wings. Help!?!

..2/974/...
.../.../.57
.../5../...

..4/85./7.2
..9/417/6..
8.. /23./4..

.../.4./57.
13./795/...
.../682/9..
derekw

Posts: 3
Joined: 13 August 2005

I don't think you strictly need an X-Wing (though I'm sure this can be rephrased in terms of one):

The 3 in box 4 has to go in either r4c1 or r5c1. Either way, it's in column 1. So, look at row 1. The 3 can't go in r1c1 or r1c2, as there are already 3s in those columns. So the only remaining places are r1c7, r1c8 and r1c9 (not3 these are all in box 3).

Now looking at column 7, the 3 can't be in r8c7, as there's aready a 3 in the row. So it has to be in r1c7, r2c7, or r7c6 (again, all in box 3).

The only way the 3 can be in both of these sets of cells is if it's in r1c7.
PaulIQ164

Posts: 533
Joined: 16 July 2005

Can't follow what moves you're describing Paul, your grid references are confusing.
derekw

Posts: 3
Joined: 13 August 2005

Whenever I post something without using the Standard Approved Grid Referencing Technique, I get messages from people going "Agh! I can't understand these crazy non-standard grid references! What do you mean, 'the box in the middle'? If you haven't numbered ot, I can't tell which one you mean!".

So it seems I can't win. anyhow, there's a thread in one of the forums that explains the standard terminology.

Unless of course, I've just explained it badly or wrongly, and that's what's confusing you, in which case there's not much I can do about it...
PaulIQ164

Posts: 533
Joined: 16 July 2005

Funny one, this one. I used simes's solver to have a look, got to within about 12 moves of completion and suddenly - nowhere to go. Simes's device had to guess.

Any thoughts?

(And I understand your references Paul. Some folk must be a bit dim. "Let's see now, what could 'r' and 'c' mean?...")
Karyobin

Posts: 396
Joined: 18 June 2005

The solver at http://sudoku.sourceforge.net gets it all. After a bit of fiddleing to get that 3 and a 1 in r1c2, it all seems to be simple "cell x can only be number y" or "the only place for number x in unit y is cell z" stuff.
PaulIQ164

Posts: 533
Joined: 16 July 2005

Paul and Karyobin

How do you explain r7c6 being in box 3?

r7c6 is in box 9
derekw

Posts: 3
Joined: 13 August 2005

Excellent point! It should say r3c7.
PaulIQ164

Posts: 533
Joined: 16 July 2005

Karyobin wrote:Funny one, this one. I used simes's solver to have a look, got to within about 12 moves of completion and suddenly - nowhere to go. Simes's device had to guess.
That's strange, I just tried it myself, and it solved it - it didn't even need anything particularly tricky, and certainly no guessing.

Simes
simes

Posts: 324
Joined: 11 March 2005
Location: UK

Perhaps a transcription error?
PaulIQ164

Posts: 533
Joined: 16 July 2005

PaulIQ164 wrote:Perhaps a transcription error?
Perhaps, but karyobin uses my solver because it's possible to cut'n'paste from the forum message instead of typing it in. However, it didn't, until two minutes ago, allow "/" as a separator, so some manual jiggery-pokery will have been necessary.

Simes
simes

Posts: 324
Joined: 11 March 2005
Location: UK

I managed to solve this by spotting a couple of naked pairs, the first giving a second one in row 7, then one in column 3 giving another one in row 2 and also one in column 6.

This giving a single value in r2c6, and from there not very hard.
Anette

Posts: 55
Joined: 09 June 2005

Quite possibly was a transcription error. As Simes says, his solver doesn't like '/'s'.

My head hurts. Ate more than a chicken yesterday. Head spinning with morris-dancers.

I don't mean to compare the amount I eat with the amount a chicken eats, but I can't be bothered to clarify anything. You'll just have to struggle.

S** *ff.
Karyobin

Posts: 396
Joined: 18 June 2005

simes wrote:
That's strange, I just tried it myself, and it solved it - it didn't even need anything particularly tricky, and certainly no guessing.

Ditto my excel solver. One iteration and completely sorted. No fishy things or anything. Just straight exclusion.

stuartn