The New Sudoku Players' Forum

[Pappocom]

I was wondering......

The best thing for these probably is to appear in a sudoku special with one of the newspapers.....

I'm sure Pappocom can pull a few strings & then we may wake up to find our puzzles printed & our names cited as members of the Sudoku Players' Forums....

Just a thought...

Sorry, tarek and the other contributors to this thread. I wasn't paying attention. For the past week I've been doing a lot of travelling, culminating in attending Will Shortz's "Sudoku Smackdown", held as part of the 29th Annual USA Crossword Competition.

Anyway, I still haven't had time to study this topic carefully. But let me know when you all think you have "finished" and we can take a look at what you have come up with.

I like the idea of a book created by contributors to the Sudoku Players Forums, with individual attribution of puzzles to their originators ... perhaps with a brief bio section as well?

However, I don't guarantee that any strings I might try to pull wouldn't break whilst being tugged. But let's see what you come up with first.

- Wayne

[tarek]

However, I don't guarantee that any strings I might try to pull wouldn't break whilst being tugged. But let's see what you come up with first

not to worry,
at the moment we're just having some fun with these, but as we pass 100 mark (which is probably the target), I think we might call it a day. at that stage the puzzles compiled in this thread (& the inferior puzzles thread) would be definitely worthy of publication.

So you don't rate locked candidates at all Tarek? I think they're fun.

Well, I'm not sure about that...it is debatable & is open for discussion...at the moment locked candidates a.k.a box-line reductions don't carry any weight in my rankings unless all other variables are the same [1 Double + 14 locked candidates score less than 2 doubles and 1 locked candidates].....

Tarek

[Red Ed]

I was admiring these sum-to-10 submissions by Red Ed, when I suddenly realized that they are not minimal.

Your "minimisation" of my puzzles takes clues away but it also requires that you add something, namely the information (now lost) that the puzzle is a sum-to-10.

My original puzzles didn't require this property to be stated up-front: they are solved by the normal rules of sudoku. If you remove any single clue from my puzzles, multiple solutions arise. In this (traditional) sense, my puzzles are minimal.

[Red Ed]

Hehe, quite correctly observed, Ruud.

And while we're on the subject, your excellent puzzle #66 which needs a hidden triple and an x-wing to solve ...

Code: Select all

Puzzle #066 (vidarino)
 . . 3 | . 9 . | 1 . .
 . . . | . . . | . . .
 9 . . | 2 . 7 | . . 4
-------+-------+------
 5 . 4 | 1 . 2 | 6 . 9
 7 . . | . . . | . . 2
 2 . 6 | 9 . 3 | 4 . 7
-------+-------+------
 1 . . | 4 . 6 | . . 5
 . . . | . . . | . . .
 . . 9 | . 1 . | 8 . .

... is not minimal You can remove the 9 in the top row, for example

But then again, minimality was never required by the rules, so maybe we shouldn't care.

[ab]

Hehe, quite correctly observed, Ruud.

And while we're on the subject, your excellent puzzle #66 which needs a hidden triple and an x-wing to solve ...

Code: Select all

Puzzle #066 (vidarino)
 . . 3 | . 9 . | 1 . .
 . . . | . . . | . . .
 9 . . | 2 . 7 | . . 4
-------+-------+------
 5 . 4 | 1 . 2 | 6 . 9
 7 . . | . . . | . . 2
 2 . 6 | 9 . 3 | 4 . 7
-------+-------+------
 1 . . | 4 . 6 | . . 5
 . . . | . . . | . . .
 . . 9 | . 1 . | 8 . .

... is not minimal You can remove the 9 in the top row, for example

But then again, minimality was never required by the rules, so maybe we shouldn't care.

I managed to remove 2 clues from that puzzle and retain the symmetry. see puzzle #67. So if Tarek and pappocom publish these puzzles in a book they'd have to choose between them and also between puzzles 1 and 3 and puzzles 14 and 25.

[Ruud]

We are aiming at a 100.

This one has a naked triple, a hidden triple and 2 X-Wings, but may only need one of them. It is not pretty, but that's not what superiority is all about, heh?

Code: Select all

. . 8|. . .|. 9 .
. . .|. 9 5|6 . .
. 1 .|. 7 .|3 . .
-----+-----+-----
. . 3|. . .|. . 9
. 4 6|. 2 .|7 1 .
7 . .|. . .|4 . .
-----+-----+-----
. . 2|. 8 .|. 3 .
. . 5|9 4 .|. . .
. 6 .|. . .|2 . .

The next one also qualifies with a hidden pair, hidden triple, naked triple and an X-Wing. Variety rules! And it's a bit prettier.

Code: Select all

. 5 .|. . .|. . 7
3 . .|2 . 6|5 . 9
. 4 .|. 5 .|6 . .
-----+-----+-----
. . 7|. 6 8|. . .
. . .|. . .|. . .
. . .|4 2 .|8 . .
-----+-----+-----
. . 5|. 9 .|. 2 .
6 . 2|8 . 3|. . 4
9 . .|. . .|. 1 .

72 down, 28 to go!

Ruud.

[Red Ed]

This one has a naked triple, a hidden triple and 2 X-Wings, but may only need one of them.

The hidden triple isn't needed to solve, though both X-wings are.

[tarek]

The first puzzle requires an x-wing, naked triple & a hidden double to solve
The second an x-wing & 2 hidden doubles.....

Neither requires the elusive hidden triple, but they definately display vatriety, both should be on the list.

The updated list at the start of this thread has 72 superiors.

tarek

[ab]

So you don't rate locked candidates at all Tarek? I think they're fun.

Well, I'm not sure about that...it is debatable & is open for discussion...
Tarek

You rate x wings quite highly. How about generalised x wings? They are a form of locked candidates. Surely you have to do the same amount of computation to spot a generalised x wing as you do to spot an x wing. That's why I differentiate between the two forms of locked candidates, one seems much harder to spot than the other.

[Deano]

From a newbie...

Code: Select all

. . . | . 8 9 | 4 2 .
. . . | . 5 1 | 6 9 .
. 5 . | . . . | . . .
---------------------
9 1 . | . . . | . . 6
4 . . | . . . | . . 8
5 . . | . . . | . 7 2
---------------------
. . . | . . . | . 3 .
. 7 1 | 9 2 . | . . .
. 2 5 | 8 1 . | . . .


Apart from one hidden pair there's an X/Y-Wing in here. Don't know if that makes it acceptable?
Anyway, the reason I posted it is that it's not computer-generated. I use a programme called Sudoku Assistenten to construct puzzles by hand; extremely satisfying because I just put the seeds in and SA does all the donkey work.

[Ruud]

Hi Deano,

The XY-Wing (with a BUG as alternative) disqualifies it as a superior sudoku.

Here is one aimed for the list:

Code: Select all

. 7 .|4 5 .|. . .
. . 8|2 . .|3 . .
5 . .|. . .|. . .
-----+-----+-----
1 . 2|. . 6|. . .
9 . .|. 1 .|. . 6
. . .|9 . .|7 . 4
-----+-----+-----
. . .|. . .|. . 9
. . 3|. . 7|8 . .
. . .|. 2 1|. 6 .

My analyzer found 8 line-box reductions, 3 naked pairs, 1 naked triple, 3 hidden pairs, 1 hidden triple and a good old X-Wing. It may be possible to bypass some of the subsets, but it shows a good variety.

73 and counting...

Ruud.

[vidarino]

I'm still well over the quota, I'm afraid, but here are a few more that I like. They all require a variety of method, all including at least one X-Wing and two hidden triples. Include them or not. Hmm, or perhaps I should review my submissions and remove a couple to make room for these?

Anyway, the goodies;

Code: Select all

+-------+-------+-------+
| 1 9 . | 6 . 8 | . 4 3 |
| 6 8 . | . . . | . 9 7 |
| . . . | . . . | . . . |
+-------+-------+-------+
| 2 . . | 7 . 4 | . . 6 |
| . . . | . 3 . | . . . |
| 3 . . | 5 . 1 | . . 2 |
+-------+-------+-------+
| . . . | . . . | . . . |
| 4 3 . | . . . | . 5 1 |
| 8 5 . | 1 . 7 | . 6 4 |
+-------+-------+-------+

+-------+-------+-------+
| 4 . . | . . . | . . 3 |
| . . 7 | . . . | 2 . . |
| . 8 2 | 7 . 1 | 9 5 . |
+-------+-------+-------+
| . . 1 | 3 . 2 | 6 . . |
| . . . | . 7 . | . . . |
| . . 3 | 8 . 9 | 5 . . |
+-------+-------+-------+
| . 2 6 | 4 . 7 | 1 9 . |
| . . 8 | . . . | 7 . . |
| 1 . . | . . . | . . 2 |
+-------+-------+-------+

+-------+-------+-------+
| . . 7 | . 3 . | 5 . . |
| . 6 . | 9 . 7 | . 3 . |
| . 1 . | . 6 . | . 9 . |
+-------+-------+-------+
| . 2 3 | . 8 . | 6 7 . |
| . . 9 | . . . | 2 . . |
| . 8 6 | . 2 . | 9 1 . |
+-------+-------+-------+
| . 5 . | . 9 . | . 6 . |
| . 3 . | 5 . 4 | . 2 . |
| . . 2 | . 1 . | 4 . . |
+-------+-------+-------+


Vidar

[Red Ed]

They all require a variety of method, all including at least one X-Wing and two hidden triples.

None of those require a hidden triple. They do all require a single X-wing, though (plus presumably a wide array of lower-order techniques, though I've not checked).

[Ruud]

Testing Vidar's latest:

#1 solves with 3 X-Wings, no subsets.

#2 solves with 2 X-Wings, no subsets.

#3 solves with a single X-Wing, no subsets.

I must admit, I often make the same mistake, but there is a difference between what is required to solve a sudoku and what a solver may find on its solving path. The reason is that we order our solvers to do the easier techniques first.

But as they require X-Wings, all of them are superior! 75 down....

Ruud.

[vidarino]

They all require a variety of method, all including at least one X-Wing and two hidden triples.

None of those require a hidden triple. They do all require a single X-wing, though (plus presumably a wide array of lower-order techniques, though I've not checked).

Hmm, that depends on the order of the methods. I have moved X-Wing before triples now, as they are debatably easier. (Not always, though. Sometimes triples are much easier to find when not working with candidates, since they're confined to one unit.)

So feel free to ignore my most recent submissions while I look for some less method-dependant ones.