| View previous topic :: View next topic |
| Author |
Message |
daj95376
Joined: 23 Aug 2008
Posts: 3854
|
 Posted: Sun Apr 18, 2010 4:38 pm Post subject: Puzzle 10/04/18 (B) |
 |
|
Single stepper ... and it doesn't use a chain, ALS, M-Wing, W-Wing, extension, transport, or pseudocell.
| Code: | +-----------------------+
| 4 2 . | . . 9 | 3 8 . |
| 3 5 . | . . 6 | 2 . . |
| . . 8 | 4 . . | 7 . . |
|-------+-------+-------|
| . . 3 | 9 7 . | 5 6 . |
| . . . | 6 . . | . . . |
| 6 7 . | . . . | . . . |
|-------+-------+-------|
| 7 8 1 | 3 . . | . 5 . |
| 5 . . | 8 . . | 1 3 . |
| . . . | . . . | . . 8 |
+-----------------------+
|
Play this puzzle online at the Daily Sudoku site |
|
| Back to top |
|
 |
arkietech
Joined: 31 Jul 2008
Posts: 1834
Location: Northwest Arkansas USA
|
| Posted: Sun Apr 18, 2010 6:25 pm Post subject: |
|
|
One Step: | Quote: | | ur 15 r9c68<>2, r9c68<>4 |
|
|
| Back to top |
|
|
daj95376
Joined: 23 Aug 2008
Posts: 3854
|
| Posted: Mon Apr 19, 2010 7:17 am Post subject: |
|
|
Sometimes I get impatient to share. That turns out to be especially true with this puzzle. Please forgive me for not waiting longer for others to post a message.
| Code: | after basics
+--------------------------------------------------------------------------------+
| 4 2 7 | *15 *15 9 | 3 8 6 |
| 3 5 9 | 7 8 6 | 2 14 14 |
| 1 6 8 | 4 23 23 | 7 9 5 |
|--------------------------+--------------------------+--------------------------|
| 28 14 3 | 9 7 1248 | 5 6 124 |
| 289 149 245 | 6 12345 123458 | 48 1247 12347 |
| 6 7 245 | 125 12345 123458 | 489 124 12349 |
|--------------------------+--------------------------+--------------------------|
| 7 8 1 | 3 2469 24 | 469 5 249 |
| 5 49 246 | 8 2469 247 | 1 3 2479 |
| 29 3 246 | *15+2 *15+2469 12457 | 469 247 8 |
+--------------------------------------------------------------------------------+
# 94 eliminations remain
|
| Code: | There is the <15> UR Type 3 eliminations in [r9] for <2469> => r 9c68<>24 (Dan's)
There is the <15> UR Type 3 eliminations in [b8] for <2469> => r89c6 <>24
|
But, most interesting of all, is r9c6=1|5 must be true to prevent the <15> DP. This results in r9c6<>247.
All of these UR eliminations result in r8c6,r9c8=7 to crack the puzzle. |
|
| Back to top |
|
|
Marty R.
Joined: 12 Feb 2006
Posts: 5770
Location: Rochester, NY, USA
|
| Posted: Tue Apr 20, 2010 12:50 am Post subject: |
|
|
| Quote: | | Single stepper ... and it doesn't use a chain, ALS, M-Wing, W-Wing, extension, transport, or pseudocell. |
Thanks for boosting my confidence. This borders on the impossible and I can't do a thing with it. I haven't read Dan's or your comments. |
|
| Back to top |
|
|
peterj
Joined: 26 Mar 2010
Posts: 974
Location: London, UK
|
| Posted: Tue Apr 20, 2010 7:56 am Post subject: |
|
|
Danny, very neat!
I got completely waylaid looking for giant fish in 2s - so nearly a Sashimi Jellyfish (only a single in one of the columns) and a Finned Swordfish (no eliminations)....
Found this a tough puzzle! |
|
| Back to top |
|
|
daj95376
Joined: 23 Aug 2008
Posts: 3854
|
| Posted: Tue Apr 20, 2010 3:23 pm Post subject: |
|
|
| This puzzle was tough because of the UR being the best way to proceed (from what my solver found). I now open it up to all of the techniques originally barred. Have fun!!! |
|
| Back to top |
|
|
tlanglet
Joined: 17 Oct 2007
Posts: 2468
Location: Northern California Foothills
|
| Posted: Tue Apr 20, 2010 4:03 pm Post subject: |
|
|
| daj95376 wrote: | Sometimes I get impatient to share. That turns out to be especially true with this puzzle. Please forgive me for not waiting longer for others to post a message.
| Code: | after basics
+--------------------------------------------------------------------------------+
| 4 2 7 | *15 *15 9 | 3 8 6 |
| 3 5 9 | 7 8 6 | 2 14 14 |
| 1 6 8 | 4 23 23 | 7 9 5 |
|--------------------------+--------------------------+--------------------------|
| 28 14 3 | 9 7 1248 | 5 6 124 |
| 289 149 245 | 6 12345 123458 | 48 1247 12347 |
| 6 7 245 | 125 12345 123458 | 489 124 12349 |
|--------------------------+--------------------------+--------------------------|
| 7 8 1 | 3 2469 24 | 469 5 249 |
| 5 49 246 | 8 2469 247 | 1 3 2479 |
| 29 3 246 | *15+2 *15+2469 12457 | 469 247 8 |
+--------------------------------------------------------------------------------+
# 94 eliminations remain
|
But, most interesting of all, is r9c6=1|5 must be true to prevent the <15> DP. This results in r9c6<>247.
|
To expand on Danny's post, his statement that "r9c6=1|5" is a result of looking at the implications of the UR candidate digits on non-UR cells in the houses of the UR. Consider the following.
The two UR candidate digits are 1 & 5. We can look at the outside row, column or box implications since any one is sufficient; using row1 and row9, we find:
Digit 1: No outside 1s in row1; r9c6 has a 1 in row9.
Digit 5: No outside 5s in row1; r9c6 has a 5 in row5.
Thus to prevent the deadly pattern r9c6=1 or r9c6=5 from which we can conclude that r9c6<>247.
Also note that in this case, looking at rows or boxes provides the exact same result.
Looking at columns, the steps are slightly more involved but results in a more positive event.
Digit 1: 1s in r6c4 & r56c5
Digit 5: 5s in r6c4 & r56c5
Thus, r456c6<>1 & r56c6<>5 which makes r9c6=1, a more positive result that completes the puzzle without additional deletions.
Ted |
|
| Back to top |
|
|
daj95376
Joined: 23 Aug 2008
Posts: 3854
|
| Posted: Tue Apr 20, 2010 4:47 pm Post subject: |
|
|
To capsulate Ted's observation:
The beauty of having a Locked Pair in a UR is that it automatically eliminated the row/column and box containing it from consideration. We can then focus our attention on the row/column and box containing the remaining two cells. In this grid, that's [r9] and [b8].
This is why I always resolve basics and X-Wings before searching for a UR pattern.
(UR Types 4/6 need an X-Wing pattern for one candidate.) |
|
| Back to top |
|
|
peterj
Joined: 26 Mar 2010
Posts: 974
Location: London, UK
|
| Posted: Tue Apr 20, 2010 4:52 pm Post subject: |
|
|
| daj95376 wrote: | | This puzzle was tough because of the UR being the best way to proceed (from what my solver found). I now open it up to all of the techniques originally barred. Have fun!!! |
I did manage without using your barred steps - but I was hoping for a monster fish pattern in 2s, instead a two step
| Quote: | Finned Swordfish (2) in columns r569c348 fin at r8c3; r9c1<>2
xy-wing(248); r5c1<>8
Plenty of other finned fish in there also - including a Sashimi Finned x-wing! |
|
|
| Back to top |
|
|
daj95376
Joined: 23 Aug 2008
Posts: 3854
|
| Posted: Tue Apr 20, 2010 5:06 pm Post subject: |
|
|
| peterj wrote: | I was hoping for a monster fish pattern in 2s ...
|
There are four (total) fish eliminations -- r56c3,r6c9,r9c1<>2 -- that can be derived through various X-Wing/Swordfish patterns after basics. Together, they are insufficient to crack the puzzle.
BTW: I noticed your use of Tarek's "Sashimi finned" preference over RonK and my choice of "Sashimi" by itself. No, I don't want to reopen that discussion in this forum.  |
|
| Back to top |
|
|
ronk
Joined: 07 May 2006
Posts: 398
|
| Posted: Tue Apr 20, 2010 5:20 pm Post subject: |
|
|
| tlanglet wrote: | Looking at columns, the steps are slightly more involved but results in a more positive event.
Digit 1: 1s in r6c4 & r56c5
Digit 5: 5s in r6c4 & r56c5
Thus, r456c6<>1 & r56c6<>5 which makes ... |
As you say, at least one of r6c4=1, r56c5=1, r6c4=5 and r56c5=5 must be true. However, I don't see how your conclusions follow. |
|
| Back to top |
|
|
daj95376
Joined: 23 Aug 2008
Posts: 3854
|
| Posted: Tue Apr 20, 2010 5:31 pm Post subject: |
|
|
Welcome back ronk 
We now have another active member to keep my misconceptions and wanton rantings in check. He also does a much more thorough check of details than I.
Regards, Danny |
|
| Back to top |
|
|
tlanglet
Joined: 17 Oct 2007
Posts: 2468
Location: Northern California Foothills
|
| Posted: Tue Apr 20, 2010 6:21 pm Post subject: |
|
|
| ronk wrote: | | tlanglet wrote: | Looking at columns, the steps are slightly more involved but results in a more positive event.
Digit 1: 1s in r6c4 & r56c5
Digit 5: 5s in r6c4 & r56c5
Thus, r456c6<>1 & r56c6<>5 which makes ... |
As you say, at least one of r6c4=1, r56c5=1, r6c4=5 and r56c5=5 must be true. However, I don't see how your conclusions follow. |
Ronk,
I jumped to a conclusion after looking at only half the constraints: (1)r6c4|r56c5 - r456c6 = (1)r9c6
I totally forgot to deal with the 5s. 
Ted |
|
| Back to top |
|
|
|
FAQ
Search
Memberlist
Usergroups
Register
Profile
Log in to check your private messages
Log in
