dailysudoku.com :: View topic - DB Saturday Puzzle

dailysudoku.com
Discussion of Daily Sudoku puzzles

FAQ

Memberlist

Usergroups

Profile

DB Saturday Puzzle - July 29, 2006

dailysudoku.com Forum Index -> Other puzzles

View previous topic :: View next topic

Author

Message

keith

Joined: 19 Sep 2005
Posts: 3355
Location: near Detroit, Michigan, USA

Posted: Sat Jul 29, 2006 12:29 pm Post subject: DB Saturday Puzzle - July 29, 2006

More difficult than usual?

Code:

Puzzle: DB072906 ******
+-------+-------+-------+
| . 9 . | 5 3 . | 4 . 8 |
| . . . | . . . | . 9 1 |
| . . 5 | . . 6 | . . 3 |
+-------+-------+-------+
| . . . | 9 . . | . 1 7 |
| . . . | . . . | . . . |
| 6 1 . | . . 8 | . . . |
+-------+-------+-------+
| 5 . . | 7 . . | 9 . . |
| 3 8 . | . . . | . . . |
| 9 . 1 | . 2 3 | . 8 . |
+-------+-------+-------+

Keith

keith

Joined: 19 Sep 2005
Posts: 3355
Location: near Detroit, Michigan, USA

Posted: Sat Jul 29, 2006 1:36 pm Post subject: My solution

I get it to here:

Code:

+----------------+----------------+----------------+
| 1 9 27a | 5 3 27b | 4 6 8 |
| 27b 36 36 | 8 47 247 | 5 9 1 |
| 8 4 5 | 1 9 6 | 27 27 3 |
+----------------+----------------+----------------+
| 24 235 238 | 9 6 45 | 238 1 7 |
| 247a 2357 89 | 23 1 457c | 68 24 69 |
| 6 1 79 | 23 47 8 | 23 5 49 |
+----------------+----------------+----------------+
| 5 26 246 | 7 8 1 | 9 3 246 |
| 3 8 247 | 6 5 9 | 1 47 24 |
| 9 67 1 | 4 2 3 | 67 8 5 |
+----------------+----------------+----------------+

Note the hidden triple <689> in R5.

There is an interesting UR in R78C39. R8C9 cannot be <4>, it must be <2>, but this does not solve the puzzle.

Coloring on <7> (either "a" or "b" is <7>) eliminates <7> from R5C6 ("c") and the puzzle is solved.

Keith

Marty R.

Joined: 12 Feb 2006
Posts: 5770
Location: Rochester, NY, USA

Posted: Sat Jul 29, 2006 4:00 pm Post subject:

Quote:

Note the hidden triple <689> in R5.

Keith, just to clarify terminology: when I was learning the most basic stuff, this would have been a naked triple, or just plain triple. I thought hidden meant it had to be somewhat disguised by having other candidates in one or more of the cells.

Marty R.

Joined: 12 Feb 2006
Posts: 5770
Location: Rochester, NY, USA

Posted: Sat Jul 29, 2006 5:17 pm Post subject: Re: My solution

keith wrote:

I get it to here:

Code:

I must've reached a similar, if not the same, position as you did Keith. I noticed that rectangle but didn't try anything with it because I was looking for other options first.

Now I have a very interesting observation. There is a strong link on "7" in cols. 1 and 6 ( what you might call a "fork"); I used that to eliminate the "7" from r2c5 (either r1c6 or r6c5 must be "7") and that broke the puzzle wide open.

So we each solved the puzzle by eliminating a "7" from c5, but from two different cells. Therefore, one of us has made an error or else this puzzle does not have a unique solution.

keith

Joined: 19 Sep 2005
Posts: 3355
Location: near Detroit, Michigan, USA

Posted: Sat Jul 29, 2006 5:34 pm Post subject: Unhidden triple

Marty,

One step back from what I posted, the situation is:

Code:

+----------------------+----------------------+----------------------+
| 1 9 27 | 5 3 27 | 4 6 8 |
| 27 36 36 | 8 47 247 | 5 9 1 |
| 8 4 5 | 1 9 6 | 27 27 3 |
+----------------------+----------------------+----------------------+
| 24 235 2348 | 9 6 45 | 238 1 7 |
| 247 2357 234789 | 23 1 457 | 2368 24 2469 |
| 6 1 479 | 23 47 8 | 23 5 49 |
+----------------------+----------------------+----------------------+
| 5 26 246 | 7 8 1 | 9 3 246 |
| 3 8 247 | 6 5 9 | 1 247 24 |
| 9 67 1 | 4 2 3 | 67 8 5 |
+----------------------+----------------------+----------------------+

The triple is hidden. I "unhid" it. It is quite difficult to spot, I think. The corresponding naked set in R5 is <23457>.

There is a UR on <23> in R56C47 which will help to partly reveal the triple.

Keith

keith

Joined: 19 Sep 2005
Posts: 3355
Location: near Detroit, Michigan, USA

Posted: Sat Jul 29, 2006 5:51 pm Post subject: The fork that I see

Marty,

The fork that I see is in C1 (ab) and C5 (cd):

Code:

+----------------+----------------+----------------+
| 1 9 27g | 5 3 27h | 4 6 8 |
| 27a 36 36 | 8 47c 247 | 5 9 1 |
| 8 4 5 | 1 9 6 | 27 27 3 |
+----------------+----------------+----------------+
| 24 235 238 | 9 6 45 | 238 1 7 |
| 247b 2357 89 | 23 1 457f | 68 24 69 |
| 6 1 79e | 23 47d 8 | 23 5 49 |
+----------------+----------------+----------------+
| 5 26 246 | 7 8 1 | 9 3 246 |
| 3 8 247 | 6 5 9 | 1 47 24 |
| 9 67 1 | 4 2 3 | 67 8 5 |
+----------------+----------------+----------------+

One of b or d must be <7>; You can eliminate <7> from cells e and f. The last one is the same elimination I made with coloring.

edit: Now I see the other fork in R1 (gh) and R5 (ed). One of h or d must be <7>; You can eliminate <7> from cells f and c.

Keith

David Bryant

Joined: 29 Jul 2005
Posts: 559
Location: Denver, Colorado

Posted: Sat Jul 29, 2006 6:01 pm Post subject: Rows or columns?

Marty R wrote:

... one of us has made an error or else this puzzle does not have a unique solution.

I don't see a "fork" in columns 1 and 6, Marty ... I see it in rows 1 & 6, instead. I think that might be a typo.

There's no conflict between your method and Keith's -- you eliminated "7" from r2c5, meaning that r2c5 = 4, and Keith eliminated a "7" from r5c6, leaving the {4, 5} pair in r45c6, and eliminating "4" from r2c6 (leaving r2c5 = 4, unique horizontal). Maybe we're all getting cross-eyed from staring at these things too long!

Oh -- I hit on the same basic structure, but I noticed the "fork" between row 1 and column 5 ({2, 7} pair in row 1 -- {4, 7} pair in column 5). That allows one to eliminate "7" from r6c3, and r6c3 = 9 also solves this puzzle. dcb

Marty R.

Joined: 12 Feb 2006
Posts: 5770
Location: Rochester, NY, USA

Posted: Sat Jul 29, 2006 6:28 pm Post subject:

Quote:

The triple is hidden. I "unhid" it. It is quite difficult to spot, I think. The corresponding naked set in R5 is <23457>.

There is a UR on <23> in R56C47 which will help to partly reveal the triple.

I did the same thing as you did Keith. It's just a difference what we call things. Rather than a hidden triple (which would be near-impossible for me to spot), I spotted the naked set which I would call a "quint", and that exposed the triple.

I also used that 23 rectangle, which I forgot to mention in my original post.

Quote:

Yes David, a typo. The fork is in rows, not columns.

As to no conflict between Keith's method and mine, again correct. Maybe I'm getting dyslexic in my dotage, as I saw his elimination of the "7" from r5c6 and I was thinking r6c5.

Sorry to start all this confusion and cause all the wasted keystrokes. Embarassed

Display posts from previous:

	dailysudoku.com Forum Index -> Other puzzles	All times are GMT
Page 1 of 1

You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum