The New Sudoku Players' Forum

by **daj95376** » Sat Mar 31, 2007 7:34 pm

Well ... sure ... take the easy one!!!

Thanks Ron!

by **ronk** » Sat Mar 31, 2007 8:55 pm

Code: Select all: 1...23495.39.....6..6....2...8.56..1....3....3..28.6...1....5..6.....81.82451...3 #G055 SSTS gets to here 1 8 7 | 6 2 3 | 4 9 5 2 3 9 | 478 47 5 | 1 78 6 45 45 6 | 1789 79 1789 | 3 2 78 -------------------+-------------------+------------------ 479 479 8 | 479 5 6 | 2 3 1 4579 6 2 | 1479 3 1479 | 79 4578 4789 3 4579 1 | 2 8 479 | 6 457 479 -------------------+-------------------+------------------ 79 1 3 | 4789 6 24789 | 5 47 2479 6 79 5 | 3 479 2479 | 8 1 2479 8 2 4 | 5 1 79 | 79 6 3 . . . | . . . | . . . . . . | . . . | . . . . . . | 9 9 9 | . . . ---------+----------+---------- 9 9 . | 9 . . | . . . 9 . . | 9 . 9 | 9 . 9 / *9 / | / / *9 | / / *9 ---------+----------+---------- #9 . . | 9 . 9 | . . -9 / *9 / | . 9 9 | . . 9 / / / | / / *9 | #9 . . sashimi franken swordfish r69b7\c269 plus fins r7c1 and r9c7, implies r7c9<>9 . . . | . . . | / . . . . . | 7 7 . | / *7 . . . . | 7 7 7 | / . *7 ---------+----------+---------- 7 7 . | 7 . . | . . . 7 . . | 7 . 7 | 7 7 7 / *7 / | / / *7 | / *7 *7 ---------+----------+---------- #7 . . | 7 . 7 | . -7 -7 / *7 / | . 7 7 | . . 7 / / / | / / *7 | #7 / / sashimi franken jellyfish r69b37\c2689 plus fins r7c1 and r9c7, implies r7c89<>7

... and cascading singles solves the puzzle.

daj95376 wrote:Well ... sure ... take the easy one!!!

LOL! I'm actually picking on the three my solver won't solve

... which means G039 is next.

by **Mike Barker** » Sun Apr 01, 2007 1:09 am

Actually G063 is easier than shown. Here's an alternative which uses a Swordfish and two Jellyfish. By the way, did I miss the memo that said that squirmbags have become starfish?

Code: Select all: Column Finned Franken Swordfish (c458/r28b5, fin=r3c8): r2458c4|r258c5|r238c8 => r2c9<>6 +---------------+--------------------+----------------+ | 6 59 1 | 59 8 3 | 7 2 4 | | 34 7 34 | 2569* 256* 12569 | 8 569* 15-6 | | 2 8 59 | 4 7 1569 | 3 569# 156 | +---------------+--------------------+----------------+ | 57 2 8 | 567* 9 4 | 1 3 567 | | 39 1 39 | 25678* 256* 2568 | 56 4 2567 | | 457 6 45 | 3 1 25 | 9 8 257 | +---------------+--------------------+----------------+ | 159 459 6 | 259 3 7 | 24 15 8 | | 159 459 7 | 25689* 256* 25689 | 24 156* 3 | | 8 3 2 | 1 4 56 | 56 7 9 | +---------------+--------------------+----------------+ Franken Jellyfish (c58b67/r2587, fins=r3c8|r46c9): r258c5|r2378c8|r5c79|r46c9|r78c12 => r2c9<>5 +---------------+---------------------+-----------------+ | 6 59 1 | 59 8 3 | 7 2 4 | | 34 7 34 | 2569 256* 12569 | 8 569* 1-5 | | 2 8 59 | 4 7 1569 | 3 569# 156 | +---------------+---------------------+-----------------+ | 57 2 8 | 567 9 4 | 1 3 567# | | 39 1 39 | 25678 256* 2568 | 56* 4 2567* | | 457 6 45 | 3 1 25 | 9 8 257# | +---------------+---------------------+-----------------+ | 159* 459* 6 | 259 3 7 | 24 15* 8 | | 159* 459* 7 | 25689 256* 25689 | 24 156* 3 | | 8 3 2 | 1 4 56 | 56 7 9 | +---------------+---------------------+-----------------+ Endofin Franken Jellyfish (r17b13/r37c48, endofin=r1c2): r1c224|r7c1248|r3c389|r2c8 => r8c2<>5 +-----------------+--------------------+----------------+ | 6 59# 1 | 59* 8 3 | 7 2 4 | | 34 7 34 | 2569 256 2569 | 8 59* 1 | | 2 8 59* | 4 7 1 | 3 569* 56* | +-----------------+--------------------+----------------+ | 57 2 8 | 567 9 4 | 1 3 567 | | 39 1 39 | 25678 256 2568 | 56 4 2567 | | 457 6 45 | 3 1 25 | 9 8 257 | +-----------------+--------------------+----------------+ | 159* 459* 6 | 259* 3 7 | 24 15* 8 | | 159 49-5 7 | 25689 256 25689 | 24 156 3 | | 8 3 2 | 1 4 56 | 56 7 9 | +-----------------+--------------------+----------------+

by **daj95376** » Sun Apr 01, 2007 1:23 am

Mike Barker wrote:Actually G063 is easier than shown. Here's an alternative which uses a Swordfish and two Jellyfish. By the way, did I miss the memo that said that squirmbags have become starfish?

Thanks Mike for your contribution! I was about to sit down and review Ron's solutions. Now, I have yours as well. One day I might actually learn something. (Actually, I'm starting to recognize Franken fish.)

As for your question. The following is from TUFG. I'm partial to the term Starfish because I can relate to it. I'd never heard of a Squirmbag until I started reading the Sudoku forums.

Code: Select all: Size of Fish: 1. 1-Fish (a.k.a Cyclopsfish): All elements are in 1 sector * 1 sector 2. X-wing: All elements are in 2 sectors * 2 sectors 3. Swordfish: All elements are in 3 sectors * 3 sectors 4. Jellyfish: All elements are in 4 sectors * 4 sectors 5. Squirmbag (a.k.a Starfish): All elements are in 5 sectors * 5 sectors 6. Whale: All elements are in 6 sectors * 6 sectors 7. Leviathan: All elements are in 7 sectors * 7 sectors

by **ronk** » Sun Apr 01, 2007 3:13 am

Mike Barker wrote:Actually G063 is easier than shown.

I assume you meant it has smaller fish than shown. When fishing for frankens and mutants without a computer program, "easiest" is landing the first fish one sees. A computer program should be able to find the smallest.

by **daj95376** » Sun Apr 01, 2007 3:45 am

Easier for me is anything I can stumble upon without making too big of a mess.

Code: Select all: G063 elimination [r2c9]<>6: # [r3c9] => ![r2c9] # ![r3c9] => Skyscraper r39 => ![r8c8] => [r23c8] => ![r2c9] *-----------------------------------* | 6 . . | . . . | . . . | | . . . | 6 6 6 | . 6 -6 | | . . . | . . 6 | . 6 6 | |-----------+-----------+-----------| | . . . | 6 . . | . . 6 | | . . . | 6 6 6 | 6 . 6 | | . 6 . | . . . | . . . | |-----------+-----------+-----------| | . . 6 | . . . | . . . | | . . . | 6 6 6 | . 6 . | | . . . | . . 6 | 6 . . | *-----------------------------------* # It's easier to write 'Skyscraper r39' than 'Sashimi X-Wing r39/c68 w/fin [r9c7]'

by **ronk** » Sun Apr 01, 2007 11:37 am

daj95376 wrote:
Code: Select all
G063 elimination [r2c9]<>6: # [r3c9] => ![r2c9] # ![r3c9] => Skyscraper r39 => ![r8c8] => [r23c8] => ![r2c9]

While arguably 'franken' may be considered simpler than 'mutant', and 'finned' simpler than 'endofinned' ... I think you've got the best interpretation for that elimination. That's because it uses three conjugate links which are easy to see. It is slightly complicated by the fact that two of the links are grouped links which overlap, but it also has four potential eliminations instead of two.

In fish terms ...

Code: Select all: . . . | . . . | . . . . . . | 6 . 6 | . *6 -6 / / / | / / *6 | . @6 *6 ---------+----------+---------- . . . | 6 . . | . / 6 . . . | 6 6 6 | 6 / 6 . . . | . . . | . / . ---------+----------+---------- . . . | . . . | . . . . . . | 6 6 6 | . *6 . / / / | / / *6 | *6 / . mutant swordfish, r39c8\c6b39 plus endofin r3c8, implies r12c79<>6 (14 empties and 4 potential exclusions)

In nice loop terms (showing all permissible candidate locations and all potential exclusions) ...

Code: Select all: r12c79-6-r123c8=6=r78c8-6-r9c79=6=r9c6-6-r3c6=6=r3c789-6-r12c79, implies r12c79<>6

by **Mike Barker** » Sun Apr 01, 2007 1:49 pm

I agree. I think Danny has the best solution especially because it wasn't computer generated! I also like Ron's fish interpretation. Looking at the skyscraper and the additional links, it was pretty easy to see the Mutant Swordfish. Maybe looking for simpler fish and skyscrapers and "kraken" links is one way to help find Mutants.

by **daj95376** » Sun Apr 01, 2007 4:05 pm

Thank You Ron and Mike for the kind words! It took some work and some luck for me to catch that particular interaction.

Ron: Thanks for finally demonstrating a Mutant fish that I could (mostly) understand. I'm particulary impressed because it has rows and columns in the Base Set. Something I've been striving to accomplish/understand for a long time.

Mike: My fish techniques are limited to Basic, Finned, Sashimi, Kraken, and the occasional Franken. Almost all of my difficult fish solutions rely on Kraken fish where reduction of the fin cell(s) require intermediate Basic, Sashimi, and (rarely) Finned X-Wings. Often, I have this feeling that I'm thissssss close to describing a mutant fish when I'm done. It's very frustrating!

I have another 140 puzzles in Set G that I need to scan. Most aren't worth anything more that finned/Sashimi practice, but maybe I'll come up with some that will produce interesting solutions.

by **tarek** » Sun Apr 01, 2007 4:20 pm

Vary nice indeed daj95376....

ronk wrote:
Code: Select all
....mutant swordfish, r39c8\c6b39 plus endofin r3c8, ....

Is it ??!!

by **ronk** » Sun Apr 01, 2007 4:22 pm

daj95376 wrote:I have another 140 puzzles in Set G that I need to scan.

Were these also in Set G :?:

If so, they aren't labeled that way.

by **daj95376** » Sun Apr 01, 2007 5:40 pm

ronk wrote:
daj95376 wrote:I have another 140 puzzles in Set G that I need to scan.

Were these also in Set G If so, they aren't labeled that way.

No, they aren't part of Set G. They occurred while I was experimenting with different configuration logic in my puzzle generator. Set G popped out of one configuration attempt, but it was only marginally successful. However, you'll notice that this set no longer has annoying XY-Chains to deal with like Set E.

BTW: Your solutions for G055 overlap my single chain solution. With it, I'm able to conclude that one of your fin cells -- [r7c1],[r9c8] -- must be true.

by **ronk** » Sun Apr 01, 2007 7:45 pm

tarek wrote:
ronk wrote:
Code: Select all
....mutant swordfish, r39c8\c6b39 plus endofin r3c8, ....

Is it ??!!

Yes

Code: Select all: Exclusions: If r3c8<>X If r3c8=X . . . | . . * | * X * . . . | . . . | * * * . . . | . . * | * X * . . . | . . . | * * * / / / | / / X | X / X * * * | * * * | * X * ---------+----------+---------- ---------+----------+---------- . . . | . . * | . / . . . . | . . . | . * . . . . | . . * | . / . . . . | . . . | . * . . . . | . . * | . / . . . . | . . . | . * . ---------+----------+---------- ---------+----------+---------- . . . | . . * | * X * . . . | . . . | . * . . . . | . . * | * X * . . . | . . . | . * . / / / | / / X | X / X . . . | . . . | . * . mutant swordfish r39c8\c6b39 Exclusions common to r3c8<>X and r3c8=X . . . | . . . | * X * . . . | . . . | * X * / / / | / / X | X @ X ---------+----------+---------- . . . | . . . | . / . . . . | . . . | . / . . . . | . . . | . / . ---------+----------+---------- . . . | . . . | . X . . . . | . . . | . X . / / / | / / X | X / X mutant swordfish r39c8\c6b39 plus endo-fin r3c8, implies r12c79<>X

The unusual thing about this is that b3 would be used twice as a cover sector in Obi-Wahn's "N+1 finless cover notation." I'm afraid mutant swordfish r39c8\c6b339 OR r39c8\c6b39b3 would each lead to confusion.

[edit: added endo-fin to combined illustration]

by **daj95376** » Mon Apr 02, 2007 6:05 am

Code: Select all: .9...185...8.....945.9.7.1...98........154........34...2.4.9.839.....1...135...4. #G103 ..254.3...1...3...3....7..2.5628..3.14.....28.3..1564.5..4....6...1...5...1.582.. #G125 .5.7..6..32...58.77.........93.5..866.5.3.2.424..6.53.........29.42...58..2..8.6. #G127 37....2.86.........24....67..52.9..4.4..3..2.2..4.61..95....83.........97.3....52 #G136 .7.6.....6.9.72...5...91....35...67.8.61.79.5.97...18....71...6...52.7.9.....8.5. #G140 3.1.45.9....37..4..2..9..........6.4..6.3.9..2.7..........1..3..6..23....1.96.4.2 #G144 ..65..478..4...5...3.8..9...7.4...9..41.5.78..8...1.5...7..3.4...8...3..359..48.. #G148 1....2...82.1...3...5...9.2.73.6.29.6.1.5.7.3.82.7.65.2.7...1...1...5.69...6....7 #G150 ...5...148.....9....4.39.5.4.5.1...7..1.2.5..2...6.4.1.2.19.6....6.....959...4... #G156 .65.8..2392...36...34..7.........7.9.4.....5.6.7.........7..36...65...4787..6.59. #G166 ..1.23..5.5......27..5.849..8.95.1.3.1.....4.3.2.41.5..783.5..42......1.1..27.8.. #G175 .61....8...4..7..272..6...14...521.3....3....3.274...86...1..242..6..5...4....83. #G184 213..8...5.67....1....1.2.5...5....4...462...3....1...6.1.7....8....51.6...1..743 #G194 ...62.17.....7...2.2.1.4.8...2...81.....9.....76...9...3.8.9.2.6...4.....91.63... #G200 ...7......59.....227.5.8.9...3....78.2.831.4.98....5...4.9.3.577.....93......7... #G227

Code: Select all: .196...3..4.3...6.6.....7....6.2794..9.1.6.5..3794.1....1.....4.6...1.7..7...469. #G900

by **tarek** » Mon Apr 02, 2007 10:11 am

ronk wrote:
tarek wrote:
ronk wrote:
Code: Select all
....mutant swordfish, r39c8\c6b39 plus endofin r3c8, ....

Is it ??!!

Yes

I've been away for 2 months & blood levels of Caffeine are not high eough at this time of day.... That is whay I'm still seeing r3c8 as a "vertix" not an endo-fin & the fish as a mutant swordfish only.... That is becuase of Box3 as you've mentioned.

some rehabilitation is needed & 'will be back........

tarek

The New Sudoku Players' Forum

Is There a Fish ?