The New Sudoku Players' Forum

[m_b_metcalf]

As One millisecond was not very accurate, and as I am working in parallel on a revised version of the "Zhou" process, I pushed the test to 100000 calls.

fss2 solves the easy puzzle 100 000 times in 77 milliseconds
and the current code on which I am working in 102 milliseconds

In both cases, the average clock time for the 8 puzzles at the top of the file of potential hardest is less than 10 seconds

[snip]
EDIT This also means that I did not pick the right puzzle in the former post. The first puzzle was not the easy one, but a "non valid" puzzle.

Gérard,
Thanks for the new information, but the EDIT leaves me slightly confused. Do you mean this puzzle:

Code: Select all

.1.....2.23.....45...2.5.....67324.....8.6.....79546.....3.8...85.....36.4.....9.;1.20;1.00;1.00


because it is very easy, has one solution and is minimal? So in what way is it "non valid"? Or did you mean a different one?

Anyway, I understand that the new timing, at 3.6 GHz, becomes roughly 1 microsecond per call for the easy puzzle, and an average of roughly 100 microseconds for each of the eight very hard ones (could you please post them here?).

Bon dimanche

Mike

[champagne]

Hi Mike,

Here after the full results for my tiny sample file.

In the first tests for the brute force I work with the following file

Code: Select all

...12........35......96....2.48.6.9....7.3....1.4.26.8....41......57........89...test t=0 1
...12........35......96....2.48.6.5....7.3....5.4.26.8....41......57........89...test t=59 2
...12........35......96....2.48.6.1....7.3....9.4.26.8....41......57........89...test t=47 3
...12........35......96....3.74.2.9....7.3....1.8.63.7....41......57........89...test t=55 4

.1.....2.23.....45...2.5.....67324.....8.6.....79546.....3.8...85.....36.4.....9.;1.20;1.00;1.00 5   
.1.....2.23.....45...2.5.....67324.....8.6.....79546.....3.8...85.....36.4.....9.;1.20;1.00;1.00 6
6....87......4..5....5....6....1..6...34.78...1..2....7....1....2..6......83....5;1.50;1.50;1.50 7

..9.8.7...6...5.4.3.....6.2.......3.5...7.1...2.......4.7.2...1.3.8...7...1...9..;6.70;6.70;6.70  8

98.7.....7.....6....6.5.....4...5.3...79..5......2...1..85..9......1...4.....3.2.;11.90;11.90;11.80;GP;champagne dry;1;22; 9
98.7.....6.....87...7.....5.4..3.5....65...9......2..1..86...5.....1.3.......4..2;11.90;11.90;11.60;GP;kz0;11523;23;10
12..3....4....1.2...52..1..5..4..2......6..7......3..8.5....9....9.7..3......8..6;11.90;11.90;11.30;elev;second flush;2;23;11
.......39.....1..5..3.5.8....8.9...6.7...2...1..4.......9.8..5..2....6..4..7.....;11.90;11.90;11.30;tax;Golden-Nugget;3;21;12
.2.4...8.....8...68....71..2..5...9..95.......4..3.........1..7..28...4.....6.3..;11.90;11.90;9.90;elev;3;4;22;13
........1....23.45..51..2....25...1..6...27..8...9......42....7.3...6...9...8....;11.90;11.90;9.90;dob;12_12_03;248078;23;14
12.3.....4.5...6...7.....2.6..1..3....453.........8..9...45.1.........8......2..7;11.90;11.90;2.60;elev;1;5;22;15
..3..6.8....1..2......7...4..9..8.6..3..4...1.7.2.....3....5.....5...6..98.....5.;11.90;1.20;1.20;elev;2;6;22;16



4 "not valid" puzzles that the first version of the "zhou" process had difficulties to reject
3 easy puzzles and one "medium rating" ,puzzle out of the pattern game
8 puzzles from the data base of potential hardest (top of the list when I picked them).

The fss2 clock in milliseconds for 100 000 calls for that file is about (small variations from one run to another one)

Code: Select all

17 17 17 17 for the "non valid"
82 82 130 for the "easy"
155 for the "medium rating"
11208 7079 9607 9703 12140 5868 5557 12194


I delivered a wrong information when I used the results of the first puzzle in my file (1 millisecond for 10 000 calls) as a result for the first "easy".

I hope that all is now clear.

Best regards

champagne

[m_b_metcalf]

I hope that all is now clear.

Crystal clear. Thanks.

Mike

[Pat]

Any puzzle published with multiple solutions is simply a mistake.

Other than here, I simply haven't seen mention of this restriction to one solution.

A. in my own view, a valid SuDoku can indeed have any number of answers.
(classified as: 0ans, 1ans, 2+ans.)

B. others reject 0ans,
and call it "valid" only with 1-answer-or-more.

C. and some call it "valid" only with exactly-1-answer.
(perhaps the majority on this forum.)

D. there's even one who restricts "valid" even more,
requiring exactly-1-answer and not-too-tough.
(using his own definition of too-tough.)

[champagne]

[list]A. in my own view, a valid SuDoku can indeed have any number of answers.
(classified as: 0ans, 1ans, 2+ans.)

interesting,

so

... ... ...
... ... ...
... ... ...

... ... ...
... ... ..
... ... ...

... ... ...
... ... ...
... ... ...

is a valid sudoku

[JasonLion]

The earliest definitions of Sudoku from Number Place and Nikoli never explicitly say that the puzzle will have exactly one solution. On the other hand, every puzzle ever published by those two sources does have exactly one solution. So while the single solution rule is never mentioned by the most authoritative sources, it is strongly implied. In practice, while there is some debate about wording, nearly everyone agrees that a properly formed puzzle, or valid puzzle, or whatever you want to call it, must have exactly one solution.

The presumption here on the forum is that puzzles are presumed to have exactly one solution unless stated otherwise. Discussions of puzzles with other numbers of solutions comes up often enough. Whatever you call those puzzles, they are of some interest.

[coloin]

interesting,

so
... ... ...
... ... ...
... ... ...

... ... ...
... ... ..
... ... ...

... ... ...
... ... ...
... ... ...
is a valid sudoku

no it is 6670903752021072936960 valid sudoku - as first found by dukuso

Quantum computers will be able to quickly solve EVERY POSSIBLE Sudoku puzzle with every possible number of clues INCLUDING all the puzzles with just one clue. It will happen with such massively parallel processing that everything that can possibly be known about Sudoku, will be known. Nothing will be left to discover.

well the number of solution grids can be calculated in less than ? 0.1 sec here
but i doubt we can know everything .....

[eleven]

... But I must warn you, if you haven't figured this out already, the coming of quantum computing will make all of the current programs completely obsolete.

Quantum computers will be able to quickly solve EVERY POSSIBLE Sudoku puzzle with every possible number of clues INCLUDING all the puzzles with just one clue. It will happen with such massively parallel processing that everything that can possibly be known about Sudoku, will be known. Nothing will be left to discover.

I must warn you not to be too optimistic about the possibilities of quantum computers, see e-g- https://en.wikipedia.org/wiki/Quantum_computing#Relation_to_computational_complexity_theory:
"There is a common misconception that quantum computers can solve NP-complete problems in polynomial time. That is not known to be true, and is generally suspected to be false."