Sudoku Players' Forum (clone)

m_b_metcalf wrote:

urhegyi wrote:It was rated with sukakuexplainer(1to9only version).
Anyone can confirm this result?

This puzzle is my Xtreme_gold! The true result is 12.0/12.0/8.1. You rated it with version 1.2.1 which contains a known bug corrected in 1.14.1 (see upthread). I used 1.17.1, which took two hours.

Regards,

Mike

I see now it's your highest rated x-sudoku. I got it from here: https://sigh.github.io/Interactive-Sudoku-Solver/ and choose example x-sudoku.
Thank you for mentioning the bug in version 1.2.1. I wasn't aware of.

Peter

m_b_metcalf wrote:

urhegyi wrote:It was rated with sukakuexplainer(1to9only version).
Anyone can confirm this result?

This puzzle is my Xtreme_gold! The true result is 12.0/12.0/8.1. You rated it with version 1.2.1 which contains a known bug corrected in 1.14.1 (see upthread). I used 1.17.1, which took two hours.

SudokuMonster's SukakuExplainer does rate this as noted above: ED=12.0/12.0/8.1.

The 1.2.1 version is an old version of my SukakuExplainer!! The last released SE displays the release date as version: 2021.4.7. My own most recent build (not released) is 2021.6.17 rates the sudoku-x as ED=11.9/11.9/8.1.

1to9only wrote: The last released SE displays the release date as version: 2021.4.7. My own most recent build (not released) is 2021.6.17 rates the sudoku-x as ED=11.9/11.9/8.1.

I've generated a number of X-windokus that rate as below, although SE claims the puzzle is valid. Could I ask your advice?

Thanks,

Mike

Code: Select all

000000010000020003000000000400003000000050006000010000000007800000060000000032000   ED=20.0/9.4/2.9

m_b_metcalf wrote:

1to9only wrote: The last released SE displays the release date as version: 2021.4.7. My own most recent build (not released) is 2021.6.17 rates the sudoku-x as ED=11.9/11.9/8.1.

I've generated a number of X-windokus that rate as below, although SE claims the puzzle is valid. Could I ask your advice?

Thanks,

Mike

Code: Select all

000000010000020003000000000400003000000050006000010000000007800000060000000032000   ED=20.0/9.4/2.9

With version 2021.4.7 I get this:

Code: Select all

000000010000020003000000000400003000000050006000010000000007800000060000000032000 ED=9.6/9.4/2.6

m_b_metcalf wrote:I've generated a number of X-windokus that rate as below, although SE claims the puzzle is valid. Could I ask your advice?

Code: Select all

000000010000020003000000000400003000000050006000010000000007800000060000000032000   ED=20.0/9.4/2.9

SudokuMonster's SukakuExplainer (private build, version 1.17.11) solves this X-Windoku incorrectly, and chokes up...

Code: Select all

.......1.....2...3.........4....3.......5...6....1.........78......6........32... ED=20.0/9.4/2.9
23954671851792864368437125942568397197125438686371942535.19786279246513.146832597


1to9only's SukakuExplainer rating (private build, version 2021.6.17):

Code: Select all

.......1.....2...3.........4....3.......5...6....1.........78......6........32... ED=9.6/9.4/2.6
932674518678521943541398267469283751217459386853716429326147895794865132185932674


When time permits, I'll check this out a little bit more...
I've raised this as an issue on SukakuExplainer's GitHub.

1to9only wrote:SudokuMonster's SukakuExplainer (private build, version 1.17.11) solves this X-Windoku incorrectly, and chokes up...

Code: Select all

.......1.....2...3.........4....3.......5...6....1.........78......6........32... ED=20.0/9.4/2.9
23954671851792864368437125942568397197125438686371942535.19786279246513.146832597


I've looked at SukakuExplainer' solution path, and it goes wrong at step 64:

Code: Select all

4.3, Grouped 2 Strong links 101: Cell r9c7,r7c9,r7c5,r1c5 on value 4: r1c6<>4


It removes candidate 4 from r1c6, when it is the cell solution.

If you disable the 'Skyscraper, Kite, Turbot' solving technique, the X-Windoku solves OK, X-Windoku rating: ED=9.6/9.4/2.6.
From the command line, specify: --techs=1111111111010111011011111011100111100111111111

Thanks. Mike

A few x-sudokus generated with the pattern from the last pattern game:

Code: Select all

..8...6...4...3...3...9...5.....7.5...5.6.7...7.5.....7...8...2...2...9...4...1.. ED=9.3/9.3/2.6
..8...1...4...3...3...9...5.....7.5...5.1.7...7.5.....7...8...2...2...9...4...6.. ED=9.3/9.3/2.6
..8...2...4...3...3...9...5.....7.5...3.5.7...7.1.....7...8...6...2...9...4...5.. ED=7.9/7.9/2.6
..8...2...4...3...3...9...5.....7.5...3.5.7...7.6.....7...8...1...2...9...4...5.. ED=7.9/7.9/2.6
..8...3...4...6...3...9...5.....7.5...1.5.7...7.6.....7...8...1...2...9...4...5.. ED=9.0/7.8/2.6
..8...3...4...6...3...9...5.....7.5...6.5.7...7.6.....7...8...1...2...9...4...5.. ED=9.1/6.5/2.6
..6...3...1...8...8...6...5.....5.3...5.7.8...8.9.....9...1...3...6...4...4...2.. ED=9.0/6.6/2.6
..3...8...7...6...2...1...4.....1.8...6.9.4...8.6.....9...3...5...9...2...7...9.. ED=9.0/6.5/2.6
..2...3...4...3...1...4...8.....8.9...8.1.6...1.6.....7...5...3...2...7...1...5.. ED=9.2/9.2/2.6
..2...3...4...3...1...4...8.....8.9...8.1.7...1.7.....6...5...3...2...6...1...5.. ED=9.2/9.2/2.6
..9...5...1...6...6...9...3.....5.7...7.3.6...6.8.....4...1...7...4...2...8...4.. ED=9.1/6.6/2.6
..3...9...1...3...6...9...3.....5.7...7.3.1...6.8.....4...2...7...4...2...8...4.. ED=9.2/9.2/2.6
..3...9...1...3...6...9...3.....5.7...7.3.5...6.8.....2...4...7...5...2...8...1.. ED=9.1/6.5/2.6
..3...2...1...3...6...9...3.....5.7...7.3.4...6.8.....1...5...7...7...2...8...9.. ED=9.1/6.6/2.8
..4...5...1...8...6...9...3.....5.7...7.4.8...6.8.....2...8...7...9...2...9...3.. ED=9.0/8.9/2.6
..7...4...9...2...3...7...1.....4.1...5.6.7...8.2.....8...4...6...6...5...6...1.. ED=9.0/7.8/2.6
..7...4...9...2...3...7...1.....4.3...5.6.2...8.2.....8...2...6...6...5...6...1.. ED=9.1/9.1/2.6
..2...9...9...2...3...7...1.....4.1...5.6.4...8.7.....8...2...6...6...5...6...1.. ED=9.2/9.2/2.6
..2...9...9...2...3...7...1.....4.1...5.6.8...8.7.....8...2...6...6...5...6...1.. ED=8.9/8.9/2.6
..8...2...9...2...3...7...1.....4.1...5.6.4...8.2.....8...4...6...6...5...6...1.. ED=9.2/6.6/2.6
..8...2...9...2...3...7...1.....4.1...5.6.9...8.2.....8...4...6...6...5...6...1.. ED=9.2/6.6/2.6
..6...8...2...6...3...7...6.....3.8...3.6.1...1.9.....4...5...1...3...4...7...2.. ED=9.3/9.3/2.8
..6...9...2...6...5...3...6.....3.9...3.7.1...1.9.....4...5...1...3...4...7...2.. ED=9.4/7.7/2.6
..6...5...2...6...3...7...6.....3.5...3.6.1...1.9.....4...8...1...3...4...7...2.. ED=9.3/9.3/2.8
..6...9...2...6...8...3...6.....3.9...3.7.1...1.9.....4...8...1...3...4...7...2.. ED=9.4/7.7/2.6
..9...6...8...6...3...7...5.....2.1...1.3.4...7.6.....8...6...1...2...5...7...3.. ED=9.2/7.8/2.6
..9...6...4...6...3...7...5.....2.1...1.3.4...7.6.....8...6...1...2...5...7...3.. ED=8.9/7.8/2.6
..2...4...8...2...9...3...1.....5.1...5.9.7...9.1.....6...8...3...2...7...9...6.. ED=7.2/6.6/2.6
..3...4...8...2...9...3...1.....4.1...5.9.2...9.1.....6...8...3...2...5...9...6.. ED=9.0/6.6/2.6
..3...7...8...2...9...3...1.....7.1...5.9.2...9.1.....6...8...3...2...5...9...6.. ED=9.0/6.6/2.6


1to9only wrote:

1to9only wrote:SudokuMonster's SukakuExplainer (private build, version 1.17.11) solves this X-Windoku incorrectly, and chokes up...

Code: Select all

.......1.....2...3.........4....3.......5...6....1.........78......6........32... ED=20.0/9.4/2.9
23954671851792864368437125942568397197125438686371942535.19786279246513.146832597


I've looked at SukakuExplainer' solution path, and it goes wrong at step 64:

Code: Select all

4.3, Grouped 2 Strong links 101: Cell r9c7,r7c9,r7c5,r1c5 on value 4: r1c6<>4


It removes candidate 4 from r1c6, when it is the cell solution.

If you disable the 'Skyscraper, Kite, Turbot' solving technique, the X-Windoku solves OK, X-Windoku rating: ED=9.6/9.4/2.6.
From the command line, specify: --techs=1111111111010111011011111011100111100111111111

This bug has been logged to look into soon. From preliminary analysis it appears that the bug may involve any grouped strong link in a box or disjoint group where the cells of the grouped strong link include what I term "heart cell" which is the intersection cell of the mini column and mini row in a box or disjoint group. This will appear more commonly in variants with windoku and will not appear in vanilla. The work around until this is sorted would be as 1to9only kindly suggested.

I have updated SE files on the repository to incorporate some typo errors and weak link improvement code in some variants but haven't made a new release until the above bug is sorted but you can still build it from the repository to get the latest version (1.18.0)

SE repository now at version 1.18.1 with issue fixed

For the record, this is my latest list of the best, PG style, X Pearls & Diamonds:


Mike

I confirm ! The last X-sudoku is the hardest I've found. Rated EXTREME-45 with my solver.
Very impressive X-sudokus !
phil

perhaps Hajime's puzzles belong on the list?



3.4/3.4/2.6

...958.....5...6...3.....5.2.......63.......77.......1.7.....9...4...8.....291...




3.6/3.6/2.6

...356.....5...6...3.....5.2.......93.......75.......1.7.....9...4...8.....291...

Pat wrote:perhaps Hajime's puzzles belong on the list?

Done.

M

A part from your list are no X Pearls or Diamonds, are they?

m_b_metcalf wrote:

Code: Select all

...123.....4...3...5.....2.6.......21.......57.......4.7...4.8...6...7.....759.1.   ED=4.2/2.9/2.9 
...123.....1...4...5........6........2.347....13...7...8....9....6...2.....962...   ED=5.7/4.3/2.9   
.............1...2.345.67....2......4..8.72.9......8....86.234.3...8.............   ED=6.4/4.1/4.0 
......1....23....44........5.3....4..6......5..1..........73.....5.8.....7...1.8.   ED=7.6/4.3/2.9   
.............1...2..34.56....2......7.89.62.3......9....95.387.8...6.............   ED=7.8/4.3/4.3   
...123.....1...4...5.....6..7.....3........8.......7.......4.......6.....9423785.   ED=8.1/4.0/2.9
.............1...2.3.2.45....3......6.78.53.9......8....84.297.7...5.............   ED=8.3/4.3/4.3   
.............1...2.3.2.45....6......7.38.56.9......8....84.296.9...5.............   ED=8.4/7.9/4.3 
..1...2....34....56........7.4....6..2......7..5..........84.....7.9.....8...5.9.   ED=8.8/4.3/2.9 

In stead I have found in the Xoup Q topic:

Code: Select all

...913.....7...8...1.....6.5.......21.......69.......8.3...4.8...5...6.....732.5.   ED=4.2/4.2/2.6
...213.....4...3...1.....2.5.......21.......67.......4.7...4.8...5...6.....786.5.   ED=7.6/7.6/2.6
...163.....7...3...1.....2.5.......21.......57.......4.7...4.8...5...6.....786.5.   ED=7.8/7.8/2.6
...913.....7...3...3.....6.7.......91.......65.......8.7...5.1...5...6.....287.4.   ED=8.1/8.1/2.6
...163.....4...3...1.....6.5.......21.......67.......4.7...4.8...5...6.....986.5.   ED=8.3/8.3/2.6
...123.....4...3...1.....2.5.......21.......67.......4.7...4.8...5...6.....968.5.   ED=8.4/8.4/2.6
...413.....4...3...1.....6.7.......24.......65.......8.7...5.8...5...6.....239.5.   ED=8.8/8.8/2.6

No ED=5.7/4.3/2.9 and ED=6.4/4.1/4.0 alternatives unfortunately...