superior plus puzzles

Everything about Sudoku that doesn't fit in one of the other sections

Postby ab » Mon Sep 04, 2006 5:39 pm

i haven't looked at all your puzzles, but the techniques you mention are not all required to solve the puzzles daj95376. For instance the last puzzle solves with just a jellyfish. However I think the puzzle I submitted requires all the techniques I listed (given the list of techniques available for the purposes of this thread).
Last edited by ab on Mon Sep 04, 2006 8:32 pm, edited 1 time in total.
ab
 
Posts: 451
Joined: 06 September 2005

Postby daj95376 » Mon Sep 04, 2006 7:11 pm

ab wrote:i haven't looked at all your puzzles, but the techniques you mention are not all required to solve the puzzles daj95376. For instance the last puzzle solves with just a swordfish. However I think the puzzle I submitted requires all the techniques I listed (given the list of techniques available for the purposes of this thread).

Well, if we're discussing required, then you might want to retest your puzzle without Naked Pairs ... or without Naked Triples. Other techniques in your list will fill in for them if they're missing.
daj95376
2014 Supporter
 
Posts: 2624
Joined: 15 May 2006

Postby ab » Mon Sep 04, 2006 9:42 pm

yes that's true. But if i turn off naked triples i have to turn on hidden triples and hidden pairs and if i turn off naked pairs i have to turn on hidden pairs.
ab
 
Posts: 451
Joined: 06 September 2005

Postby ab » Tue Sep 05, 2006 12:29 am

sorry daj95376, I just realised how my initial response must have looked to you. Thanks for the nice set of puzzles. I've added puzzles 1, 3, 4, 9, 11, 13, 14 and 15 to the list, since I think they're the pick of the bunch. I've found that swordfish and naked pairs are required quite often the ones I've added are the ones that go beyond that!
ab
 
Posts: 451
Joined: 06 September 2005

Postby ab » Tue Sep 05, 2006 9:58 am

another beaut:
Code: Select all
 . 3 1 | . . . | 7 2 .
 . . . | 9 . 8 | 6 . .
 . . . | . 7 . | . . .
 ------+-------+------
 . 5 . | . . . | . 4 .
 . . 2 | 7 . 6 | 3 . .
 . 8 . | . . . | . 1 .
 ------+-------+------
 . . . | . 9 . | . . .
 . . 8 | 5 . 4 | . . .
 . 6 3 | . . . | 8 9 .

locked candidates, naked pair, hidden triple + swordfish
ab
 
Posts: 451
Joined: 06 September 2005

Postby daj95376 » Tue Sep 05, 2006 5:23 pm

ab wrote:sorry daj95376, I just realised how my initial response must have looked to you. Thanks for the nice set of puzzles. I've added puzzles 1, 3, 4, 9, 11, 13, 14 and 15 to the list, since I think they're the pick of the bunch. I've found that swordfish and naked pairs are required quite often the ones I've added are the ones that go beyond that!

No problem ab. I'm glad that we are past the confusion. I half expected it and that's why I mentioned technique order in my initial reply. Thanks for adding some of my puzzles to your list!!!

BTW: If you place all of the Naked n-tuples before the Hidden n-tuples, your last example has two Naked Quads pop up along with the Swordfish. However, I concede that they aren't required ... just pretty! (This would also give you an example with a triple in the plus category!)
daj95376
2014 Supporter
 
Posts: 2624
Joined: 15 May 2006

Postby ab » Fri Sep 08, 2006 12:07 am

here's another really nice one. it's fully symmetric and minimal (I think). requires 2 swordfish, an x wing and a naked pair.
Code: Select all
 . . 6 | 9 . 2 | 3 . .
 . . . | . . . | . . .
 2 . . | . 8 . | . . 4
 ------+-------+------
 3 . . | . 7 . | . . 1
 . . 9 | 5 . 6 | 8 . .
 5 . . | . 4 . | . . 6
 ------+-------+------
 1 . . | . 6 . | . . 7
 . . . | . . . | . . .
 . . 8 | 2 . 4 | 6 . .
ab
 
Posts: 451
Joined: 06 September 2005

Postby daj95376 » Fri Sep 08, 2006 12:45 am

Hello ab. Nice Cloverleaf. You may wish to double-check the required. Here's what I get.

Code: Select all
Naked Pair: Replaceable with a (different candidates) Hidden Pair.
X-Wing:     Completely unnecessary.
Swordfish:  Both are necessary.
daj95376
2014 Supporter
 
Posts: 2624
Joined: 15 May 2006

Postby ab » Fri Sep 08, 2006 1:07 am

yes you may be right daj95376. there are actually 3 swordfish on the same step, but one of them is complemented (or maybe includes) the x wing, so i implemented the x wing to ensure that the other two swordfish are necessary. i didn't actually check to see that the x wing was required.
ab
 
Posts: 451
Joined: 06 September 2005

re(2): classification of quartets

Postby Pat » Wed Sep 13, 2006 8:59 am

— i see that we already have an example of 9=4+4+1

Ocean (2006.Jul.19) wrote:- - - - -24 clues

Code: Select all
 . . . | 1 . . | . 2 .
 . . 3 | . 4 5 | . . 6
 . 4 . | . 7 . | . . .
-------+-------+------
 2 . . | . . . | . 8 .
 . 8 7 | . . . | 4 5 .
 . 9 . | . . . | . . 1
-------+-------+------
 . . . | . 5 . | . 3 .
 9 . . | 7 3 . | 8 . .
 . 1 . | . . 8 | . . .
.



Pat (2006.Aug.31) wrote:- - - - -classification of quartets


9 empty cells in a unit:
    A. 9=4+5 (quartet with quintet)
    B. 9=4+4+1 (2 quartets - find either of them and you're left with a "single")
8 empty cells in a unit:
    C. 8=4+4 (2 quartets)
Last edited by Pat on Wed Sep 13, 2006 5:07 am, edited 1 time in total.
User avatar
Pat
 
Posts: 4056
Joined: 18 July 2005

Postby udosuk » Wed Sep 13, 2006 9:06 am

Code: Select all
 *--------------------------------------------------------------------*
 | 5678   567    5689   | 1      689    369    | 35     2      4      |
 | 18     2      3      | 89     4      5      | 179    179    6      |
 | 156    4      1569   | 2369   7      2369   | 35     19     8      |
 |----------------------+----------------------+----------------------|
 | 2      3      1456   |#456   *16    #467    | 679    8      79     |
 | 16     8      7      |*269   *1269  *269    | 4      5      3      |
 | 456    9      456    |#34568  68    #3467   | 2      67     1      |
 |----------------------+----------------------+----------------------|
 | 4678   67     2468   | 2469   5      12469  | 169    3      29     |
 | 9      56     2456   | 7      3      1246   | 8      146    25     |
 | 3      1      2456   | 2469   269    8      | 679    4679   2579   |
 *--------------------------------------------------------------------*

Are you talking about the naked quad *{1269} in n5? And the hidden quad #{3457}?
udosuk
 
Posts: 2698
Joined: 17 July 2005

re: 2 quartets

Postby Pat » Wed Sep 13, 2006 9:10 am

those who find the "naked" quartet
will get a "naked single"

those who see the "hidden" quartet
get a "hidden single"
User avatar
Pat
 
Posts: 4056
Joined: 18 July 2005

Postby daj95376 » Wed Sep 20, 2006 12:06 am

FYI on a hierarchy issue for Superior Plus puzzles. First, lets review the techniques.

Code: Select all
Superior: Naked/Hidden Singles/Pairs/Triples, Locked Candidates, X-Wings
Plus:     Naked/Hidden Quads, Swordfish, Jellyfish

I've discovered that Naked Quads are often missed if Hidden Pairs/Triples testing is performed first. The result is some awkward testing logic. Given their modest frequency, it's too bad that you didn't give Hidden Pairs/Triples a Plus designation after Naked Quads.
daj95376
2014 Supporter
 
Posts: 2624
Joined: 15 May 2006

Postby daj95376 » Thu Sep 21, 2006 6:30 pm

Deleted: Wrong hierarchy used!!!
Last edited by daj95376 on Thu Sep 21, 2006 7:58 pm, edited 1 time in total.
daj95376
2014 Supporter
 
Posts: 2624
Joined: 15 May 2006

Postby gsf » Thu Sep 21, 2006 7:07 pm

daj95376 wrote:ab, I'll let you decide (1) desirability, (2) hierarchy, and (3) required ... for these puzzles.

I have these not meeting the plus requirement laid out by ocean: 1-7, 35, 37, 38
for almost all of the other ones containing quads I got jellyfish instead, but
that's just a detail on how the plus constraints are ordered in each of our solvers

for mine the constraint groups and ordering are
Code: Select all
{T1H1B2B3T2H2W2T3H3}{W3T4H4W4}

where only the groups determine splus membership,
i.e., at least one constraint in the 2nd group is required

even when putting quads before swordfish
Code: Select all
{T1H1B2B3T2H2W2T3H3}{T4H4W3W4}

quads only showed up for: 13, 36, 41

what is your constraint order?
gsf
2014 Supporter
 
Posts: 7306
Joined: 21 September 2005
Location: NJ USA

PreviousNext

Return to General

cron