The New Sudoku Players' Forum

by **wapati** » Sun Jun 03, 2007 1:51 pm

Interesting, but not so hard, perhaps.

Swordfish and a Jellyfish are in there with numerous "chum".

SA and SS solve it because the little patterns eat up the big fish.

SE finds it pretty easy because only one chain. (No fins on the fish.)

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

by **udosuk** » Sun Jun 03, 2007 4:12 pm

Page 3 is not broken. It's just that the incident involving rep'nA/re'born (which is not his fault at all) resulted in some posts being deleted, but the forum database didn't intelligently reduce the post counts for this thread (or topic). So while this thread currently has 27 posts in total (not including this very post), the not-so-smart database thinks it had 31, and created an empty page 3.

After this post is recorded, this thread will have 28 posts, and 3 more posts later a post will appear on a legitimate page 3. :idea:

by **wapati** » Mon Jun 04, 2007 1:59 am

This is a dilly. A swordfish and a Jelly!

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

by **wapati** » Mon Jun 04, 2007 2:02 am

This one is fairly easy, SE 7.1 though, I am filling up the "page" to try to fix Page 3.

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

by **wapati** » Mon Jun 04, 2007 2:04 am

A Swordfish and many more mini-fish.

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

by **wapati** » Mon Jun 04, 2007 2:09 am

Udosuk was right, Page 3 is fixed.

Here is the puzzle that I posted in the "new" thread, and which is now gone.

This puzzle is contorted and finned. (Must be an eel? ) Uniqueness may come in early but doesn't help much. Odd that it is early is why I mention it.

Some big finned fishies.

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

by **daj95376** » Mon Jun 04, 2007 2:23 am

Yes, and what will happen once this page is nearly full?

by **wapati** » Mon Jun 04, 2007 2:45 am

daj95376 wrote:Yes, and what will happen once this page is nearly full?

I don't yet know. My worry was that the missing posts might return and mess up the new ones.

by **udosuk** » Mon Jun 04, 2007 4:35 am

daj95376 wrote:Yes, and what will happen once this page is nearly full?

After this post, we have 35 posts in this thread and the forum DB thinks it has 39. When we have 42 posts in this thread the DB will think it has 46 and create a dummy page 4, which will come into effect when 4 more posts are recorded. Unfortunately, this will happen over and over again. :!:

wapati wrote:I don't yet know. My worry was that the missing posts might return and mess up the new ones.

If the missing posts returned then all problems would be solved as the DB post-count will match the actual one in reality. However, as the administrator suggested, it's highly unlikely to happen.

by **wapati** » Mon Jun 04, 2007 8:54 pm

This an OK puzzle, not easy for true.

Still, maybe only one big fishy, finned at that.

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

by **wapati** » Mon Jun 04, 2007 9:09 pm

This is a puzzle I like, not too hard yet hard enough.

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

by **wapati** » Wed Jun 06, 2007 8:28 am

A swordfish or two.

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

by **re'born** » Wed Jun 06, 2007 10:14 am

wapati wrote:This is a puzzle I like, not too hard yet hard enough.

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

Ooh, that one is tricky with the x-wing coming so early. It seems one can solve it with that x-wing, an xy-wing, a skyscraper and a UR. But here is a different approach using BUG-lites. First, grant me the early x-wing so that I can get to here:

Code: Select all: *-----------------------------------------------------------* | 12 6 8 | 12(5) 9 3 | 157* 57* 4 | | 13 37 4 | 6 1578 15 | 2 9 158 | | 5 9 27 | 12(8) 178 4 | 18 6 3 | |-------------------+-------------------+-------------------| | 8 25 19& | 157& 4 12(5)6| 379*& 37* 56 | | 7 4 3 | 58% (58)6 9 | 568 1 2 | | 6 25 19& | 1578& 3 12(5) | 79& 4 58 | |-------------------+-------------------+-------------------| | 9 1 6 | 4 2 8 | 35* 35* 7 | | 23 37 27 | 9 15 156 | 4 8 16 | | 4 8 5 | 3 16 7 | 16 2 9 | *-----------------------------------------------------------*

To avoid a deadly pattern in r147c78, we must have r1c7=1 or r4c7=9. On the other hand, to avoid a deadly pattern in r46c347, we must have r4c4=5, r4c7=3, r6c4=5 or r6c4=8. The key cell is now r5c4=[58]. Take any cell that sees r456c4. If it contains a 5 or 8, then r5c4 will contain the other and hence we must have r4c7=3. But an easy argument (using the strong link on 7 in row 1) then shows that r1c7 = 7, a contradiction.

We conclude that r1c4, r5c5, r46c6<>5 and r3c4, r5c5<>8, solving the puzzle. I'm not sure how this should be written up as a chain.

by **re'born** » Wed Jun 06, 2007 12:54 pm

wapati wrote:This an OK puzzle, not easy for true.

Still, maybe only one big fishy, finned at that.

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

I'm still looking for the one big finned fishy that will solve it. However, it will solve with one 2-string kite, one remote naked pair and one xy-wing.

by **ronk** » Wed Jun 06, 2007 5:27 pm

re'born wrote:It seems one can solve it with that x-wing, an xy-wing, a skyscraper and a UR. But here is a different approach using BUG-lites. First, grant me the early x-wing so that I can get to here:
Code: Select all
*-----------------------------------------------------------* | 12 6 8 | 12(5) 9 3 | 157* 57* 4 | | 13 37 4 | 6 1578 15 | 2 9 158 | | 5 9 27 | 12(8) 178 4 | 18 6 3 | |-------------------+-------------------+-------------------| | 8 25 19& | 157& 4 12(5)6| 379*& 37* 56 | | 7 4 3 | 58% (58)6 9 | 568 1 2 | | 6 25 19& | 1578& 3 12(5) | 79& 4 58 | |-------------------+-------------------+-------------------| | 9 1 6 | 4 2 8 | 35* 35* 7 | | 23 37 27 | 9 15 156 | 4 8 16 | | 4 8 5 | 3 16 7 | 16 2 9 | *-----------------------------------------------------------*

To avoid a deadly pattern in r147c78, we must have r1c7=1 or r4c7=9. On the other hand, to avoid a deadly pattern in r46c347, we must have r4c4=5, r4c7=3, r6c4=5 or r6c4=8. The key cell is now r5c4=[58]. Take any cell that sees r456c4. If it contains a 5 or 8, then r5c4 will contain the other and hence we must have r4c7=3. But an easy argument (using the strong link on 7 in row 1) then shows that r1c7 = 7, a contradiction.

We conclude that r1c4, r5c5, r46c6<>5 and r3c4, r5c5<>8, solving the puzzle. I'm not sure how this should be written up as a chain.

re'born, you have an uncanny ability to spot some tricky deductions. Even with the details of your post, it took me a long time to realize the underlying reason for the deduction is ...

{BUG-Lite r147c78: r4c7=9|1=r1c7}=7=r1c8-7-r4c8-3-r4c7, implies r4c7<>3

Then a BUG-Lite+2 "Type 3" in r46c347 remains ... where "Type 3" is meant in the same sense as UR Type 3.

The New Sudoku Players' Forum

Hard puzzles, chains not required.