impossible game generated by Sudoku

[watergeus]

Binary package hint: gnome-games

Very hard puzzle under Ubuntu 9.04 is impossible to solve
As below with 0 for initial blank squares

000008002
208300100
073010800
000001345
000935000
531700000
009060270
006007409
700200000

This isn't the first I've encountered, but it's the easiest to confirm. I could produce a screenshot for you if you want.

sudoku:
  Installed: (none)
  Candidate: 1.0.1-3
  Version table:
     1.0.1-3 0
        500 http://gb.archive.ubuntu.com jaunty/universe Packages

ProblemType: Bug
Architecture: i386
DistroRelease: Ubuntu 9.04
NonfreeKernelModules: wl
Package: gnome-games 1:2.26.1-0ubuntu2
ProcEnviron:
 LANG=en_GB.UTF-8
 SHELL=/bin/bash
SourcePackage: gnome-games
Uname: Linux 2.6.28-13-generic i686

[Sebastien Bacher (seb128)]

Thank you for your bug report, how do you know it's impossible to solve? a screenshot would be clearer than your text drawing

[watergeus (moschris)]

Agreed, screenshot attached.

All that has been done with the initial configuration is
1. Use the "Fill all squares" option.
2. Add to row 1 the list of all possible completions.
3. Expose hidden pairs.

>From here it will be seen there are several contradictions particularly
column 5 which must contain both 7 and 9.

OK?

Chris

On Fri, 2009-07-10 at 10:55 +0000, Sebastien Bacher wrote:
> Thank you for your bug report, how do you know it's impossible to solve?
> a screenshot would be clearer than your text drawing
>
> ** Changed in: gnome-games (Ubuntu)
> Importance: Undecided => Low
>
> ** Changed in: gnome-games (Ubuntu)
> Status: New => Incomplete
>
> ** Changed in: gnome-games (Ubuntu)
> Assignee: (unassigned) => Ubuntu Desktop Bugs (desktop-bugs)
>

[Sebastien Bacher (seb128)]

The issue is an upstream one and it would be nice if somebody having it could send the bug the to the people writting the software (https://wiki.ubuntu.com/Bugs/Upstream/GNOME)

There is no clear indication the game can't be solved

[watergeus (moschris)]

apologies, I'm quite new to Linux reporting systems.

Also I'm new to the "uncovered pairs" approach and I realise I made a
mistake in the screenshot I sent you. I'm trying to find a simple
representation of the problem but the only one I have so far is rather
more complicated.

Chris

On Fri, 2009-07-10 at 12:38 +0000, Sebastien Bacher wrote:
> The issue is an upstream one and it would be nice if somebody having it
> could send the bug the to the people writting the software
> (https://wiki.ubuntu.com/Bugs/Upstream/GNOME)
>
> There is no clear indication the game can't be solved
>
> ** Changed in: gnome-games (Ubuntu)
> Status: Incomplete => New
>

[Kim spence-Jones (ksj)]

OK. I have another impossible puzzle...

ab9546c3d
__8___5_7
25_______
_4_63___8
___1_8___
5___24_1_
______e26
8_5___3__
_3_2897__

Here's the proof it's impossible (numbering from 1,1 bottom left):

cell e must be 8, because rows 1 and 2 have an 8 in, and e is the only remaining free cell in bottom right-hand 3x3 group.
a and b can't be 8 because there is an 8 in the top left 3x3 group (at 3,8), and c can't be 8 because of e. d can't be 8 because of (9,6). So, there is no space for an 8 in the top row.

[uselessness (uselessness)]

I have this problem. First puzzle I ever tried, it's on Medium. Check out the screenshot (sorry it's not cropped or anything), pay attention to the center column, the 8s... there are other contradictions too but that one is obvious.

[Pablo Castellano (pablocastellano)]

Upstream : https://bugzilla.gnome.org/show_bug.cgi?id=619190

[Vish (vish)]

Thanks for sending upstream

[Tiffany Antopolski (mimico-tiffany)]

Only 2 of the above 3 reports of invalid puzzles appear to be valid.

See upstream comments on this bug: https://bugzilla.gnome.org/show_bug.cgi?id=619190