The New Sudoku Players' Forum

[UKGuy]

I have been doing Su Doku for a while and whilst not the fastest kid on the block I can usually solve them. However last Thursdays Times Fiendish Su Doku has got me beat. Can anyone give me the next step? I have checked with the solution and so far I have not made any mistakes.

The puzzle was:

***87***5
13****487
*****5***
*5****174
*********
893****2*
***7*****
614****98
5***98***

I have got to here:

**287***5
135***487
**8**5***
256*8*174
**1***839
893**7526
3897***5*
614***798
527*98*4*

I am sure there must be some technique I don't know about!
Any help will be greatly appreciated.

[Guest]

[Edited by Pappocom]


thanks

[ronk]

Can anyone give me the next step?

Some people don't read and give the solution instead.

Angus Johnson's Simple Sudoku program has a pleasant graphic interface and can provide hints on a step-by-step basis as you need them.

However, if you want a specific hint from someone in a forum such as this, you really need to provide the candidate list at your sticking point. For the puzzle you asked about, if your candidate list were ...

Code: Select all

 *-----------------------------------------------------------------------------*
 | 49      46      2       | 8       7       13469   | 369     16      5       |
 | 1       3       5       | 269     26      269     | 4       8       7       |
 | 479     467     8       | 123469  12346   5       | 2369    16      123     |
 |-------------------------+-------------------------+-------------------------|
 | 2       5       6       | 39      8       39      | 1       7       4       |
 | 47      47      1       | 2456    2456    246     | 8       3       9       |
 | 8       9       3       | 14      14      7       | 5       2       6       |
 |-------------------------+-------------------------+-------------------------|
 | 3       8       9       | 7       1246    1246    | 26      5       12      |
 | 6       1       4       | 235     235     23      | 7       9       8       |
 | 5       2       7       | 136     9       8       | 36      4       13      |
 *-----------------------------------------------------------------------------*

... they might point to the naked pair 16 in box 3,
or naked pair 14 in box 5,
or the lock by box 3 on candidate 2 in row 3,
or the lock by box 9 on candidate 2 in row 7,
etc. etc.

[UKGuy]

Thanks Ronk the program you sent me to showed me immediately where the gap in my logic was!

[SammyJ]

I programmed that one in just now and my site worked it out. Also gives the steps it uses, and does it online, so no downloads.

Yours is the latest one entered so will come up automatically

http://sudoku.sampercy.com/index.php

I know you've solved it now, but can you tell me if this is helpful? I'm still working on it when I can.

Sam.

[Pi]

I don't really understand the annotation on that site, could somebody give direction from here, my program is under construction and so i am doing it manualy



|49 :46 : 2 | 8 : 7 134|3 9:1 6: 5 |
| 1 : 3 : 5 |269:26:269| 4 : 8 : 7 |
|479:467: 8 |134:134: 5 |239: 16: 23|
| 2 : 5 : 6 | 39: 8 : 39| 1 : 7 : 4 |
| 47: 47: 1 |256:256: 26| 8 : 3 : 9 |
| 8 : 9 : 3 | 14: 14: 7 | 5 : 2 : 6 |
| 3 | 8 | 9 | 7 :146:146| 26: 5 : 12|
| 6 : 1 : 4 |235:235: 23| 7 : 9 : 8 |
| 5 : 2 : 7 | 16 : 9 : 8 | 36: 4 13|

[ronk]

I don't really understand the annotation on that site

The parenthesized locations are (x, y) => (col, row) which is reversed from the more usual row-col notation such as r4c3. Also ...
groups 1 to 9 are rows 1 to 9,
groups 10 to 18 are columns 1 to 9, and
groups 19 to 27 are boxes 1 to 9.

I have no idea why the user is asked to perform such mental gymnastics.

could somebody give direction from here

Look for a naked quad in column 6

[masb]

Simple Sudoku looks good. An alternative Sudoku grid with visual hints is http://www.axcis.com.au/numbler.

[foxface]

I have the solution for this puzzle, can you give me your email so i can send it to you.

[Cec]

" ...I have no idea why the user is asked to perform such mental gymnastics..."

I share your same opinion Ronk. Whilst commending anyone for expanding their programming skills, I too find SammyJ's notation of using 'x' (for column) and 'y' (for row) is less "user-friendly" than this forum's established practice of defining a cell, say r8c4, where the "r" obviously refers to the row number and "c" the column number.

Similarly, dividing the grid into nine 3X3 boxes (numbered 1 to 9) is far preferable to Sammyj's notation of defining these boxes as groups 19 to 27 - for example, 'box 7' is mentally much easier to focus in your mind compared to locating where 'group 25' is.

Cec

[Lardarse]

cecbevwr and ronk:

It makes sense to define it this way, with each group as a collection of cells, because it's then easy to add additional groups, like diagonals, and the whole idea can then be extended further to multiple grids.

[ronk]

with each group as a collection of cells, because it's then easy to add additional groups, like diagonals, and the whole idea can then be extended further to multiple grids.

Most people think in terms of rows, columns, and boxes. Should users pay a penalty to make it easier for the programmer? Or make extension easier to puzzle styles that very people care about? IMO the answers are clearly ... "no" and "no".

[David B]

I have completed every puzzle since 472 apart from yesterday's Fiendish 512 but looking at this forum I see 472 is a classic.

I reached the grid submitted by Ronk and got a little further tidying up box 2 but have not completed another cell yet. I use a Mac so the applications mentioned are no use to me. Can someone point me to the next step, I'm not so much interested in the final solution, I have a program which will display that, but I wonder if there is a logical technique I am not using.

Many thanks.

[Pi]

If you have got to the stage where i posted..

In collum 6 there is a hidden pair of 1 and 4

Therefore you can eliminate some more candidates.

[David B]

Thanks, I hadn't spotted that, now I'll see where it leads.


Minutes later ------ it fell out from there, many thanks.