| View previous topic :: View next topic |
| Author |
Message |
daj95376
Joined: 23 Aug 2008
Posts: 3854
|
 Posted: Tue Dec 02, 2008 9:49 pm Post subject: Set E Puzzle 4 |
 |
|
Since my alerts seem to be extraneous. I won't suggest a Swordfish or chain for this puzzle.
| Code: | +-----------------------+
| . . . | . . . | 7 2 . |
| . . . | . . 8 | 1 . . |
| . . . | 2 3 . | . . . |
|-------+-------+-------|
| . . 2 | . . 7 | 9 . . |
| . . 3 | . 9 6 | 4 . . |
| . 7 . | 8 4 . | . . 6 |
|-------+-------+-------|
| 5 8 . | 7 6 . | . 9 3 |
| 6 . . | . . . | 8 4 . |
| . . . | . . 3 | 6 . 7 |
+-----------------------+
|
Play this puzzle online at the Daily Sudoku site |
|
| Back to top |
|
 |
storm_norm
Joined: 18 Oct 2007
Posts: 1741
|
| Posted: Wed Dec 03, 2008 3:41 am Post subject: |
|
|
| another alert for this puzzle is "classic remote naked pair" as there are two of them from the get go. |
|
| Back to top |
|
|
Marty R.
Joined: 12 Feb 2006
Posts: 5770
Location: Rochester, NY, USA
|
| Posted: Wed Dec 03, 2008 4:49 am Post subject: |
|
|
| I might have to try this again. I had 11 moves involving W-Wings, Remote Pairs, coloring, Hidden URs and X-Wings before grinding to a halt and getting out my Medusa crutch. |
|
| Back to top |
|
|
daj95376
Joined: 23 Aug 2008
Posts: 3854
|
| Posted: Wed Dec 03, 2008 5:47 am Post subject: Re: Set E Puzzle 4 |
|
|
| daj95376 wrote: | Since my alerts seem to be extraneous. I won't suggest a Swordfish or chain for this puzzle.
|
Let me rephrase the above.
Since my alerts seem so often not to be needed, I'm not going to issue an alert for techniques that I used in this puzzle.  |
|
| Back to top |
|
|
storm_norm
Joined: 18 Oct 2007
Posts: 1741
|
| Posted: Wed Dec 03, 2008 8:30 am Post subject: |
|
|
| Code: | .------------------------.------------------------.------------------------.
| 3 -(15)49 145689 | 14569 #15 1459 | 7 2 489 |
|*29 *2459 4569 | 4569 7 8 | 1 3 49 |
| 7 149 1489 | 2 3 149 | 5 6 489 |
:------------------------+------------------------+------------------------:
| 4 6 2 | 3 #15% 7 | 9 8 %15 |
| 8 #15 3 |#15% 9 6 | 4 7 2 |
| 19 7 159 | 8 4 2 | 3 15 6 |
:------------------------+------------------------+------------------------:
| 5 8 14 | 7 6 14 | 2 9 3 |
| 6 3 7 |-(15)9 2 159 | 8 4 %15 |
|*129 *124-9 149 |415 8 3 | 6 15 7 |
'------------------------'------------------------'------------------------' |
visually, the two tangent classic remote pairs are present on {1,5} means {1,5} is gone from r8c4 and r1c2.
as well as the type 5 UR (I think) on{2,9} r29c12 with two strong links on 2, which means 9 is gone from r9c2.
| Code: | .------------------.------------------.------------------.
| 3 49 1568 | 1456 15 1459 | 7 2 489 |
| 2 459 56 | 456 7 8 | 1 3 49 |
| 7 149 18 | 2 3 149 | 5 6 489 |
:------------------+------------------+------------------:
| 4 6 2 | 3 15 7 | 9 8 15 |
| 8 15 3 | 15 9 6 | 4 7 2 |
| 19 7 159 | 8 4 2 | 3 15 6 |
:------------------+------------------+------------------:
| 5 8 14 | 7 6 14 | 2 9 3 |
| 6 3 7 | 9 2 15 | 8 4 15 |
| 19 2 149 | 145 8 3 | 6 15 7 |
'------------------'------------------'------------------' |
the UR {1,9} r69c13 makes two of the same eliminations as the x-wing on 1... r69c3 and r9c4 is not 1.
the type 5 UR on {5,6} r12c34 removes 5 from r1c4
and still have the swordfish on 5 which removes the 5 in r1c3
leaves this xy-chain- (4=9)r9c3 - (9=5)r6c3 - (5=6)r2c3 - (6=4)r2c4; r9c4 <>4... to finish it. |
|
| Back to top |
|
|
daj95376
Joined: 23 Aug 2008
Posts: 3854
|
| Posted: Wed Dec 03, 2008 10:09 am Post subject: |
|
|
| storm_norm wrote: | visually, the two tangent classic remote pairs are present on {1,5} means {1,5} is gone from r8c4 and r1c2.
|
Fold in [c8] and you have a third remote pair that eliminates {1} from r9c2.
(Yeah, I know, it doesn't help much.)
Whew!!! Not your boring ole' XY-Wing single-stepper, huh? |
|
| Back to top |
|
|
Marty R.
Joined: 12 Feb 2006
Posts: 5770
Location: Rochester, NY, USA
|
| Posted: Wed Dec 03, 2008 4:58 pm Post subject: |
|
|
| Quote: | | as well as the type 5 UR (I think) on{2,9} r29c12 with two strong links on 2, which means 9 is gone from r9c2. |
That's not a Type 5, at least based on Keith's UR treatise, but it is a type of Hidden UR as described by Sudopedia (and perhaps others).
http://www.sudopedia.org/wiki/Uniqueness_Test
| Quote: | | Since my alerts seem so often not to be needed, I'm not going to issue an alert for techniques that I used in this puzzle. |
Without going to a lot of trouble, is there a way you could provide a degree of difficulty? |
|
| Back to top |
|
|
daj95376
Joined: 23 Aug 2008
Posts: 3854
|
| Posted: Wed Dec 03, 2008 7:40 pm Post subject: |
|
|
| Marty R. wrote: | | Without going to a lot of trouble, is there a way you could provide a degree of difficulty? |
I posted several puzzles that I thought would take several steps to solve. Others found them to be single/double-steppers. 
Many of the puzzles in Set E aren't going to be solved so easily because they were generated with XY-Chains being allowed (but not necessarily required).
This one qualifies as extreme because of the Swordfish, XY-Chains, and complex URs present.
Fortunately, there are some puzzles in Set E that only qualify as darn ugly.  |
|
| Back to top |
|
|
cgordon
Joined: 04 May 2007
Posts: 769
Location: ontario, canada
|
| Posted: Wed Dec 03, 2008 7:48 pm Post subject: |
|
|
Very straightforward at first!
Two ER's on <1>
an x wing on <1>
an ER on<5>
a bit of colouring on <5>
an xyz wing
then (and this was brilliant) a Sue de Coq from Box 1 C2
...... but then I get stuck.
I'm left with a whole bunch of <15>s but they don't form any Classic Remote Pair - at least not for me - the chains are all odd numbered
And I sure don't see ant Type UR for <29>. How can you have a UR with only one pair.
This was a cruel one! |
|
| Back to top |
|
|
storm_norm
Joined: 18 Oct 2007
Posts: 1741
|
| Posted: Wed Dec 03, 2008 8:18 pm Post subject: |
|
|
| Quote: | That's not a Type 5, at least based on Keith's UR treatise, but it is a type of Hidden UR as described by Sudopedia (and perhaps others).
|
yep, you are right, its not a type 5, its a hidden one. at least the eliminations are correct 
oh by the way, this rated in at a 7.2 SE |
|
| Back to top |
|
|
daj95376
Joined: 23 Aug 2008
Posts: 3854
|
| Posted: Wed Dec 03, 2008 8:27 pm Post subject: |
|
|
| cgordon wrote: | | This was a cruel one! |
Yes, in retrospect, I shouldn't have posted this puzzle. My apologies
After basics, X-Wing, Swordfish, and Colors, I single-stepped several XY-Chains by using a Remote Pair chain, alternately colored Blue and Green, to force a UR contradiction.
| Code: | Remote Pair: (5=1)r5c2 - (1=5)r5c4 = (5-1)r4c5 - (1=5)r1c5
( [r5c2]=1 => [r1c5]=5 ) => (49) UR [r13c26] ==>> [r5c2]<>1
reduces puzzle to Naked Singles
+--------------------------------------------------------------+
| 3 49 4689 | 146 G15 49+5 | 7 2 489 |
| 29 2459 4569 | 46 7 8 | 1 3 49 |
| 7 49+1 1489 | 2 3 49 | 5 6 489 |
|--------------------+--------------------+--------------------|
| 4 6 2 | 3 B15 7 | 9 8 15 |
| 8 B15 3 | G15 9 6 | 4 7 2 |
| 19 7 59 | 8 4 2 | 3 15 6 |
|--------------------+--------------------+--------------------|
| 5 8 14 | 7 6 14 | 2 9 3 |
| 6 3 7 | 9 2 15 | 8 4 15 |
| 129 249 49 | 45 8 3 | 6 15 7 |
+--------------------------------------------------------------+
|
Another way to view the UR is [r3c2]=1 and/or [r1c6]=5. No matter what combination is true, all of the Blue cells are forced to {5} and all of the Green cells are forced to {1}. Not as elegant as my description! |
|
| Back to top |
|
|
cgordon
Joined: 04 May 2007
Posts: 769
Location: ontario, canada
|
| Posted: Wed Dec 03, 2008 8:57 pm Post subject: |
|
|
Actually I forgot to mention I also had a Type 6 UR (49) which knocked off the <9> in R3C2.
What's a 7.2 rating. Is that like a Class 7 Hurricane or 7.2 on the Richter Scale - i.e something that is life-threatening? |
|
| Back to top |
|
|
keith
Joined: 19 Sep 2005
Posts: 3355
Location: near Detroit, Michigan, USA
|
| Posted: Thu Dec 04, 2008 11:28 pm Post subject: |
|
|
| Quote: | | And I sure don't see any Type UR for <29>. How can you have a UR with only one pair. |
Consider this case:where the corners are also an X-wing on <2>. x, y, z are any candidates. <9> in the bottom right cell forces <2> in the upper right and lower left, which forces <9> in the upper left; a DP. So, the lower right cannot be <9>.
If you haveyou can immediately simplify it toThese have been called both "Type 6" and "Hidden" UR's.
In the end, these seem not to be very useful, although Mike Barker used them to crack a few previously "unsolvable" puzzles. The way I see it is this: The UR argument is maybe going to solve the X-wing candidates on one diagonal or the other. Since you have already made the X-wing eliminations (which have nothing to do with the UR), the effect on the rest of the puzzle is small.
That is the case in the puzzle of this thread: Solving the remaining <2>'s does not help much.
Keith |
|
| Back to top |
|
|
cgordon
Joined: 04 May 2007
Posts: 769
Location: ontario, canada
|
| Posted: Fri Dec 05, 2008 5:24 pm Post subject: |
|
|
| Quote: |
29 29x
29y 29z
where the corners are also an X-wing on <2>. x, y, z are any candidates. <9> in the bottom right cell forces <2> in the upper right and lower left, which forces <9> in the upper left; a DP. So, the lower right cannot be <9>. |
Keith:
I'm lost here. I understand Type 6 UR's in that if there was a strong link on <2> in the top row and the bottom right cell was a 29 (not a 29z) - you can remove the <2> from the 29y. (like a diagonal variant of a Type 4). Was this incorrect?
I also don't understand in your example why you can assume the bottom left is a <9> and not a <z>.
Confused.
Cheers,
Craig |
|
| Back to top |
|
|
|
FAQ
Search
Memberlist
Usergroups
Register
Profile
Log in to check your private messages
Log in
