The New Sudoku Players' Forum

[gurth]

JADE RATINGS AND OYSTERS

I think more now that the attempt to rate jades might end in the belief that it serves little purpose. It remains too subjective an appreciation. One can't count beauty, or compare them. We know what we like, and in any case we have the SE number to think about and the number of singles, apart from other factors less easy to enumerate.

In fact I like JPF's "challenger" as much as any of them.

OYSTERS:

My latest craving is for oysters. (No, I am not a pregnant woman.) Oysters can be with or without pearls. Pearl-oysters are hard then easy. The ultimate pearl-oyster would be a puzzle starting with one ultra-hard step followed by singles only. You only have to crack an oyster once. Is there such a thing?

Also good are oysters without pearls. They can start easy. But essentially the oyster makes its point once: it requires you to rise to a necessary level to achieve success. Once you have met that demand, it doesn't go on endlessly making the same demand on you again.

Even if there is no perfect oyster, I would like to get closer to the oyster ideal of short and sweet. Steps of maximum difficulty, but as few as possible.

A puzzle with say all singles except for one swordfish would be an acceptable oyster on the easier levels.

Any offerings? sorry I have none of my own.

_______________________________________________________________________

[gurth]

RW, my apologies for denying your 29 singles. As you rightly surmised, I got fooled by SE.

[RW]

My latest craving is for oysters. (No, I am not a pregnant woman.) Oysters can be with or without pearls. Pearl-oysters are hard then easy. The ultimate pearl-oyster would be a puzzle starting with one ultra-hard step followed by singles only. You only have to crack an oyster once. Is there such a thing?

The Effortless Extreme puzzles are all some kind of oysters, with the requirement of one hard move and the rest nothing more than singles, locked candidates, pairs and triplets. There's several puzzles there that need nothing but the one move and singles, can't remember straight away if there's any oysters with pearls. Personally I like oysters that can be cracked with some very simple, but rare, technique. Like this one from the Reverse BUG-lite thread:

Code: Select all

 *-----------*
 |...|...|.83|
 |..1|38.|.46|
 |.23|...|.1.|
 |---+---+---|
 |7..|15.|.34|
 |...|..7|8..|
 |.6.|.4.|.7.|
 |---+---+---|
 |..5|8..|...|
 |...|5..|...|
 |2.7|4.1|...|
 *-----------*

Not quite a pearl, some singles available in the beginning, SE 7.8, make the Reverse BUG-lite elimination and it get's down to 2.8.

I also like oysters where the tough move is visible immediately in the starting grid, when you don't need to do any eliminations before you can use the "oyster cracking technique".

RW

[Ocean]

OYSTERS:
[...]
Also good are oysters without pearls. They can start easy. But essentially the oyster makes its point once: it requires you to rise to a necessary level to achieve success. Once you have met that demand, it doesn't go on endlessly making the same demand on you again.

Even if there is no perfect oyster, I would like to get closer to the oyster ideal of short and sweet. Steps of maximum difficulty, but as few as possible.

A puzzle with say all singles except for one swordfish would be an acceptable oyster on the easier levels.

Any offerings? sorry I have none of my own.

Here is an attempt on "Jade & Easy Oyster without pearl".

Code: Select all

#Jade & Easy Oyster without pearl (M21).
......12.
1.....3..
64...5...
...6.78..
....1....
..64.2...
...9...71
..5.....4
.72......


[JPF]

The ultimate pearl-oyster would be a puzzle starting with one ultra-hard step followed by singles only. You only have to crack an oyster once. Is there such a thing?

Still with Gordon’s help, here is a pearl-oyster :

Code: Select all

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



S.S. : 1 locked candidates (1). End

S.E. : 1 Forcing X-Chain (SE = 6.7) : r6c8 <>8 and the puzzle is solved = 64 singles.

JPF

[Ocean]

An "easy" pearl-oyster.
XY-wing at start, after HD and ND. After the xy-wing, the rest is singles (but narrow path).

Code: Select all

# "Easy" pearl-oyster with oyster symmetry.
..1...2..
.3.....4.
5...6...7
...1.5...
.2..4..3.
...6.7...
7.......1
.4..5..2.
..6...8..


[gurth]

RW, Ocean and JPF,

Many thanks for your Oyster offerings. I'll have a meal!


ORIGINAL JADE ?

Compare the two jades:

Code: Select all

..6.7.45.
.1.8.....
.2.......
7...4....

4......8.
........1
.8.2.....
3.....7..
...1.....  "newcomer" SE=8.3 n=37 mp=17

..6...45.
.1.8.....
.2....1..
..1.4....
4......87
...7...4.
.8.2.7...
3........
.7.1.....  GC01 SE=8.3 n=34 mp=20


I have "doctored" NEWCOMER by shifting 3 7s and adding a few more clues to make it valid, getting a jade that is most excellent, yet can be viewed as just an inferior version of NEWCOMER.

My problem and question is this: Exactly how much doctoring does one have to do to any puzzle to earn the right to call the new puzzle one's own? ESPECIALLY IF THE NEW VERSION SEEMS WORSE THAN THE OLD? Anybody with any ideas that might give me food for thought?
_____________________________________________________________

[Mauricio]

Here is my contribution to jades.
It starts with 5 singles and then a SE step of rating 9.1.

Code: Select all

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


[gurth]

GC04 : Easy Oyster

Fresh out of the sea this morning:

Code: Select all

  . . . . 4 . 3 . 2
  . 4 . . 1 . . . .
  3 . 7 . 8 2 . 1 .
  6 3 . . . . . . 1
  1 . . . . . 2 . .
  . . 4 . . . . . 3
  4 . 1 . . 5 . . 6
  2 . 5 6 . . . . .
  . . . . . 1 5 2 .


[gurth]

Mauricio, welcome to you and your heavyweight jade.

As heavyweights go, she's quite vicious. Even that Extreme heavyweight Jade, Ocean's "for gurth", after the 5 singles sweetener, slapped you in the face with a mere Turbot Fish before using chains on you.

Tarek's "Starter" was also considerate enough to slap you with a Turbot Fish before starting with the chains.

And even then, those gentle heavyweights used some 6.7 to 7.8 chains, before starting with the 9 plus stuff.

But no, your jade wastes no time in starting the rough stuff with a 9.1 chain!

I am endlessly fascinated by the range of personalities of sudokus, Turquoises, Pearls, Oysters and especially Jades.

What do you call her? Medea? Don't forget to give her credit for her full weight of 9.3 .

She also has 5 mutable clues, seems rather a lot for a 9.3 . You can put 6c3, 5d3, 2g7, 2h1 or 4k2: with what consequences I haven't checked.

___________________________

[gurth]

GC06 : A Very Generous Jade

JPF, remember your jade Challenger, the one I said I liked so much? Here's a companion for it, nearly as generous with 31 starting singles, a bit tougher at SE 7.1 :

Code: Select all

#La Generosa (GC06)

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


At old ratings:
Challenger .............. : JR = 6.6 x 34 = 224.4
La Generosa (GC06) : JR = 7.1 x 31 = 220.1

- A well-matched pair. Mine also has a Turbot Fish.

__________________________________________________________________________________

[ArkieTech]

daj95376 said:

Are there two Sashimi X-Wings in <1> (as well) that would produce the same eliminations?

Code: Select all

 *-----------*
 |...|...|8..|
 |9..|.4.|.6.|
 |.2.|.68|..5|
 |---+---+---|
 |.3.|...|...|
 |.4.|.5.|...|
 |..6|...|798|
 |---+---+---|
 |..1|3.6|..4|
 |...|4..|.5.|
 |2..|57.|6..|
 *-----------* # gurth GC06
 

Code: Select all

# at this point, Coloring could be used for these eliminations in <1>
 *-----------------------------------------------------------*
 | 4     6     5     | 1     2     3     | 8     7     9     |
 | 9     1     8     | 7     4     5     | 23    6     23    |
 | 3     2     7     | 9     6     8     | 4     1     5     |
 |-------------------+-------------------+-------------------|
 | 78    3     29    | 68   G19    179   | 5     4     26-1  |
 | 78    4     29    | 68    5     79-1  |B123   23    1236  |
 | 1     5     6     | 2     3     4     | 7     9     8     |
 |-------------------+-------------------+-------------------|
 | 5     7     1     | 3     89    6     | 29    28    4     |
 | 6     89    3     | 4    B189   2     |G19    5     7     |
 | 2     89    4     | 5     7     19    | 6     38    13    |
 *-----------------------------------------------------------*
 

The 2 Sashimi X-Wings looks like a turbot with a force.

I can't see a solution or what is next. Help

dan

[ArkieTech]

With Carcul's solution of gurth GC06 :

[r8c2]=8=[r9c2]=9=[r9c6]=1=[r9c9]-1-[r8c7]-9-[r8c2], => r8c2<>9 solving the puzzle.

The Sashimi X-Wings do not contribute to the solution.

dan

[daj95376]

With Carcul's solution of gurth GC06 :

[r8c2]=8=[r9c2]=9=[r9c6]=1=[r9c9]-1-[r8c7]-9-[r8c2], => r8c2<>9 solving the puzzle.

The Sashimi X-Wings do not contribute to the solution.

dan

ArkieTech, There are at least three chains -- [r7c2]<>2, [r9c2]<>8, and [r9c6]<>9 -- that will crack open the original puzzle without ever examining so much as an initial Naked/Hidden Single. However, some people frown at jumping over basic and lesser techniques to reach a solution.

[Mauricio]

On the same lines of my previous posted sudoku, I found this in the hardest sudoku thread, it starts with 5 singles, and then a 9.2 step, final rating 9.9

Code: Select all

9....67...4..3......71....4..8.....6.2..1..8.6....49..1....2..8.5..7..2....3..6..

A few nice jades:

Code: Select all

100008009400000002080400030004009070007000000000300005070006900060100800000502006
SE 7.3, 40 singles

Code: Select all

103060008056000200000024300004085009900001000000047100000000703008000005000800006
SE 8.3, 31 singles

Code: Select all

100597600000000900080004020902000008560021000000040000000000890000032706640000000
SE 8.4, 27 singles


Code: Select all

120804000050000002000503060000039050000600000307002100000000075000108000004370000
SE 9.0, 18 singles


Code: Select all

000050400050300000000002001007000080800000070090800003041009600600100207000003000
SE 9.1, 13 singles

Code: Select all

023000004000000070080500600008040009000001046000600000090087000304000000002093010

000000009406080100009000450000300608000070000008910045800204000004067090000890000
SE 9.2, 11 singles

Edit: Corrected a few typos when I changed 0's by .'s. I like 0's more.