| View previous topic :: View next topic |
| Author |
Message |
storm_norm
Joined: 18 Oct 2007
Posts: 1741
|
 Posted: Sun Feb 24, 2008 9:29 am Post subject: feb 24 bundle |
 |
|
these first three are from the sudokulist site. sudoku.org.uk
#959 daily competition
| Code: | . 6 . | . . 7 | . . .
. . 8 | . 9 . | 2 . .
. 9 7 | 5 . 3 | . . .
- - - + - - - + - - -
. . 2 | . 6 . | 9 . .
9 4 . | . . . | . 2 8
. . 5 | . 2 . | 7 . .
- - - + - - - + - - -
. . . | 6 . 8 | 5 1 .
. . 4 | . 5 . | 6 . .
. . . | 4 . . | . 9 . |
#957 daily competition
| Code: | 3 2 . | . . . | . . .
5 . . | 6 . . | . 4 7
. 9 4 | . . . | 5 . .
- - - + - - - + - - -
. . . | . 9 3 | . . .
. . 6 | 4 . 8 | 2 . .
. . . | 1 6 . | . . .
- - - + - - - + - - -
. . 2 | . . . | 8 7 .
4 8 . | . . 7 | . . 6
. . . | . . . | . 3 9 |
weekly extreme #74
| Code: | . . . | . . 9 | . 4 .
1 . . | . 4 . | . . 6
7 . . | . . 5 | . . .
- - - + - - - + - - -
. 2 . | 4 . 6 | . . .
. . 6 | . 5 . | 9 . .
. 1 . | 9 . 2 | . 8 .
- - - + - - - + - - -
. . . | 2 . . | . . 5
9 . . | . 3 . | . . 7
. 8 . | 7 . . | . . . SE 7.2 |
lastly is the feb 22 daily telegraph which is actually from the sudokulist site as well.
| Code: | . 8 . | . 4 7 | . 2 .
. . . | 2 . . | . 6 .
4 . 2 | . . . | 3 . .
- - - + - - - + - - -
. 2 . | 3 1 . | . . .
6 . . | . . . | . . 7
. . . | . 2 6 | . 1 .
- - - + - - - + - - -
. . 9 | . . . | 1 . 8
. 5 . | . . 3 | . . .
. 6 . | 9 5 . | . 4 . |
the extreme puzzle's rating is the lowest I have seen it in a long time so I am optimistic about solving it. its still a work in progress tho.
the daily telegraph puzzle contains nice patterns, easy on the eyes.
enjoy,
norm |
|
| Back to top |
|
 |
keith
Joined: 19 Sep 2005
Posts: 3355
Location: near Detroit, Michigan, USA
|
| Posted: Sun Feb 24, 2008 5:37 pm Post subject: |
|
|
After basic moves, #957 is:
| Code: | +----------------+----------------+----------------+
| 3 2 7 | 58 48 45 | 69 69 1 |
| 5 1 8 | 6 2 9 | 3 4 7 |
| 6 9 4 | 37 37 1 | 5 8 2 |
+----------------+----------------+----------------+
| 28 457 15 | 257 9 3 | 67 156 48 |
| 19 57 6 | 4 57 8 | 2 19 3 |
| 28 3457 39 | 1 6 25 | 79 59 48 |
+----------------+----------------+----------------+
| 19 36 2 | 39 14 46 | 8 7 5 |
| 4 8 39 | 359 35 7 | 1 2 6 |
| 7 56 15 | 28 18 26 | 4 3 9 |
+----------------+----------------+----------------+
|
Now what?
Keith |
|
| Back to top |
|
|
Marty R.
Joined: 12 Feb 2006
Posts: 5770
Location: Rochester, NY, USA
|
| Posted: Sun Feb 24, 2008 5:57 pm Post subject: |
|
|
Looks to me like it's tailor-made for the snake-haired one.  |
|
| Back to top |
|
|
nataraj
Joined: 03 Aug 2007
Posts: 1048
Location: near Vienna, Austria
|
| Posted: Sun Feb 24, 2008 6:18 pm Post subject: |
|
|
xy-wing pivot r9c6 removes 5 from r6c2
w-wing r5c8-r7c1 removed 1 from r5c1
xy-chain removes 9 from r7c1,r6c3:
'19'(r5c1)...r4c3,r5c2,r5c5,r8c5...'39'(r8c3)
no need for the snake-haired one... |
|
| Back to top |
|
|
keith
Joined: 19 Sep 2005
Posts: 3355
Location: near Detroit, Michigan, USA
|
| Posted: Sun Feb 24, 2008 6:30 pm Post subject: |
|
|
Actually,
| Code: | +----------------+----------------+----------------+
| 3 2 7 | 58 48 45 | 69 69 1 |
| 5 1 8 | 6 2 9 | 3 4 7 |
| 6 9 4 | 37 37 1 | 5 8 2 |
+----------------+----------------+----------------+
| 28 4-57 15c |2-57 9 3 | 67 156 48 |
| 19 57a 6 | 4 57A 8 | 2 19 3 |
| 28 34-57 39 | 1 6 25a | 79 59 48 |
+----------------+----------------+----------------+
| 19 36 2 | 39 14 46 | 8 7 5 |
| 4 8 39 | 359 35 7 | 1 2 6 |
| 7 56c 15C | 28 18 26b | 4 3 9 |
+----------------+----------------+----------------+ |
The XY-wing abc takes out a <5>. Then, you can color on <5> from each of the pincers to remove two more <5>'s.
Any cell that sees both a and c cannot be <5>. Leaving a BUG+1.
Keith |
|
| Back to top |
|
|
Victor
Joined: 29 Sep 2005
Posts: 207
Location: NI
|
| Posted: Sun Feb 24, 2008 7:16 pm Post subject: |
|
|
| Nice logic Keith. |
|
| Back to top |
|
|
Victor
Joined: 29 Sep 2005
Posts: 207
Location: NI
|
| Posted: Sun Feb 24, 2008 7:23 pm Post subject: |
|
|
| The extreme. Got a little way into this, but found it hard going after a while. Following your advice in a previous thread, Storm_norm, I'm walking away from this one! |
|
| Back to top |
|
|
tlanglet
Joined: 17 Oct 2007
Posts: 2468
Location: Northern California Foothills
|
| Posted: Sun Feb 24, 2008 8:50 pm Post subject: |
|
|
| keith wrote: |
The XY-wing abc takes out a <5>. Then, you can color on <5> from each of the pincers to remove two more <5>'s.
Any cell that sees both a and c cannot be <5>. Leaving a BUG+1.
Keith |
Keith,
I have looked around for a definition of BUG+1, BUG+2, etc but all I have found is a simple BUG definition with a single cell having more than 2 values which I have always assumed was the same as a BUG+1. But in this puzzle, I observe three cells with polyvalues and yet you referred to it as a BUG+1 condition. I am once again confused.
Could you direct me to a general BUG outline including definitions and solution techniques?
Ted |
|
| Back to top |
|
|
keith
Joined: 19 Sep 2005
Posts: 3355
Location: near Detroit, Michigan, USA
|
| Posted: Sun Feb 24, 2008 9:18 pm Post subject: |
|
|
Ted,
There are simplifications I did not mention: A hidden pair in R4, and a couple of pinned cells, that get you to here:
| Code: | +-------------+-------------+-------------+
| 3 2 7 | 58 48 45 | 9 6 1 |
| 5 1 8 | 6 2 9 | 3 4 7 |
| 6 9 4 | 37 37 1 | 5 8 2 |
+-------------+-------------+-------------+
| 28 47 15 | 27 9 3 | 6 15 48 |
| 19 57 6 | 4 57 8 | 2 19 3 |
| 28 34 39 | 1 6 25 | 7 59 48 |
+-------------+-------------+-------------+
| 19 36 2 | 39 14 46 | 8 7 5 |
| 4 8 39 | 359 35 7 | 1 2 6 |
| 7 56 15 | 28 18 26 | 4 3 9 |
+-------------+-------------+-------------+ |
This is the BUG+1 I mentioned. R8C4 must be <3>.
Otherwise, if you are looking for practice, I have a theory that any BUG+1 pattern will fall to Medusa coloring.
A BUG is a pattern where every cell has two candidates, and each candidate occurs exactly twice in the unsolved cells in each row, column, and box.
A BUG+1 is as given above. One cell has more than two candidates, I believe it must be three. The candidate in the +1 cell that destroys the BUG is the correct value in that cell.
A BUG+2 has two cells that have more than two candidates, a BUG+3 has three such cells, etc.
I do not know of any description of a BUG+n that both describes what it is, and what to do about it. The original posts on the topic are quite dense.
If you'll excuse my (lack of) modesty, I have posted a few messages on this site where I attempted to explain what to do with a BUG+n. Perhaps if you search on
keith BUG+
you will find something.
There are two steps:
1. Establish what the BUG pattern is, if it exists.
2. Make inferences from the candidates that destroy the BUG.
Some of this logic can be really cool, but there is no fixed recipe.
Best wishes,
Keith |
|
| Back to top |
|
|
keith
Joined: 19 Sep 2005
Posts: 3355
Location: near Detroit, Michigan, USA
|
| Posted: Sun Feb 24, 2008 9:21 pm Post subject: |
|
|
| Victor wrote: | | The extreme. Got a little way into this, but found it hard going after a while. Following your advice in a previous thread, Storm_norm, I'm walking away from this one! |
The extreme, after basics:
| Code: | +-------------------+-------------------+-------------------+
| 236 5 238 | 1368 128 9 | 7 4 128 |
| 1 9 238 | 38 4 7 | 23 5 6 |
| 7 346 2348 | 1368 128 5 | 123 1239 1289 |
+-------------------+-------------------+-------------------+
| 38 2 9 | 4 18 6 | 5 7 13 |
| 48 7 6 | 18 5 3 | 9 12 124 |
| 345 1 345 | 9 7 2 | 6 8 34 |
+-------------------+-------------------+-------------------+
| 346 346 7 | 2 69 148 | 148 169 5 |
| 9 46 12 | 5 3 148 | 1248 126 7 |
| 25 8 125 | 7 69 14 | 1234 12369 129 |
+-------------------+-------------------+-------------------+ |
Sudoku Susser does it with six chains and a wing. Not my cup of tea, but not that impossible.
Keith |
|
| Back to top |
|
|
Johan
Joined: 25 Jun 2007
Posts: 206
Location: Bornem Belgium
|
| Posted: Sun Feb 24, 2008 9:27 pm Post subject: |
|
|
The Daily Telegraph puzzle:
For avoiding the [59] DP in R16C79*, either R1C9=1 or/both R6C7=4, transporting <4> in R6C7 via Box 4 and Box 1 pins <1> in R1C9, after that a 4-cell xy-chain with pincer ends in R5C4(A) and R7C6(D) wipes out both <4>'s in R7C4 and R5C6.
The grid after basics:
| Code: | +--------------------------+--------------------------+--------------------------+
| 3 8 6 | 15 4 7 |*59 2 *159 |
| 59 7 15 | 2 3 19 | 8 6 4 |
| 4 19 2 | 1568 689 1589 | 3 7 15 |
+--------------------------+--------------------------+--------------------------+
| 59 2 7 | 3 1 459 | 49 8 6 |
| 6 149 15 | 458 89 4589 | 2 3 7 |
| 8 49 3 | 7 2 6 |*459 1 *59 |
+--------------------------+--------------------------+--------------------------+
| 27 3 9 | 46 67 24 | 1 5 8 |
| 17 5 4 | 18 78 3 | 6 9 2 |
| 12 6 8 | 9 5 12 | 7 4 3 |
+--------------------------+--------------------------+--------------------------+ |
After <59> UR:
| Code: | +--------------------------+--------------------------+--------------------------+
| 3 8 6 | 5 4 7 | 9 2 1 |
| 59 7 15 | 2 3 19 | 8 6 4 |
| 4 19 2 | 168 689 189 | 3 7 5 |
+--------------------------+--------------------------+--------------------------+
| 59 2 7 | 3 1 59 | 4 8 6 |
| 6 19 15 | 48 A 89 -[4]589 | 2 3 7 |
| 8 4 3 | 7 2 6 | 5 1 9 |
+--------------------------+--------------------------+--------------------------+
| 27 3 9 |-[4]6 67 D 24 | 1 5 8 |
| 17 5 4 | 18 B 78 3 | 6 9 2 |
| 12 6 8 | 9 5 C 12 | 7 4 3 |
+--------------------------+--------------------------+--------------------------+ |
|
|
| Back to top |
|
|
keith
Joined: 19 Sep 2005
Posts: 3355
Location: near Detroit, Michigan, USA
|
|
| Back to top |
|
|
Marty R.
Joined: 12 Feb 2006
Posts: 5770
Location: Rochester, NY, USA
|
| Posted: Sun Feb 24, 2008 10:15 pm Post subject: |
|
|
| Quote: | | the extreme puzzle's rating is the lowest I have seen it in a long time so I am optimistic about solving it. its still a work in progress tho. |
If that's a low rating, I think I'll avoid them. I had an unhelpful ER and Finned X-Wing, but after that it took four or five iterations of extended Medusa to solve. |
|
| Back to top |
|
|
storm_norm
Joined: 18 Oct 2007
Posts: 1741
|
| Posted: Sun Feb 24, 2008 11:32 pm Post subject: |
|
|
| Marty R. wrote: | | Quote: | | the extreme puzzle's rating is the lowest I have seen it in a long time so I am optimistic about solving it. its still a work in progress tho. |
If that's a low rating, I think I'll avoid them. I had an unhelpful ER and Finned X-Wing, but after that it took four or five iterations of extended Medusa to solve. |
I am sure Ravel can say more about the rating. sudoku explainer says it needs 11 forcing chains. ugh!
for my money, when I see that a puzzle needs more than 2 or 3 forcing chains then I kind of tell myself that it probably needs a couple AICs or a whole lot of medusa and medusa multicoloring. |
|
| Back to top |
|
|
Steve R
Joined: 24 Oct 2005
Posts: 289
Location: Birmingham, England
|
| Posted: Mon Feb 25, 2008 3:32 am Post subject: |
|
|
You can make some progress with the extreme puzzle if you are prepared to assume that it has a unique solution.
Here is Keith’s grid:
| Code: | +-------------------+-------------------+-------------------+
| 236 5 238 | 1368 128 9 | 7 4 128 |
| 1 9 238 | 38 4 7 | 23 5 6 |
| 7 346 2348 | 1368 128 5 | 123 1239 1289 |
+-------------------+-------------------+-------------------+
| 38 2 9 | 4 18 6 | 5 7 13 |
| 48 7 6 | 18 5 3 | 9 12 124 |
| 345 1 345 | 9 7 2 | 6 8 34 |
+-------------------+-------------------+-------------------+
| 346 346 7 | 2 69 148 | 148 169 5 |
| 9 46 12 | 5 3 148 | 1248 126 7 |
| 25 8 125 | 7 69 14 | 1234 12369 129 |
+-------------------+-------------------+-------------------+ |
Consider r789c67. Only a MUG would confine the entries for these cells to (148) so, bearing in mind the (23) in r2c7, you can place 1 in r3c7.
A single chain is then sufficient to solve the puzzle.
Steve |
|
| Back to top |
|
|
storm_norm
Joined: 18 Oct 2007
Posts: 1741
|
| Posted: Mon Feb 25, 2008 3:35 am Post subject: |
|
|
| Quote: | | Only a MUG would confine the entries for these cells |
steve,
please do fill me in on a MUG?? |
|
| Back to top |
|
|
storm_norm
Joined: 18 Oct 2007
Posts: 1741
|
|
| Back to top |
|
|
Steve R
Joined: 24 Oct 2005
Posts: 289
Location: Birmingham, England
|
| Posted: Mon Feb 25, 2008 4:54 am Post subject: |
|
|
So they did.
I hadn’t seen Myth Jellies' post when I wrote but the only other two examples of MUGs I have come across were those of Vidarino and Ron Kral earlier on the same thread.
Steve |
|
| Back to top |
|
|
Earl
Joined: 30 May 2007
Posts: 677
Location: Victoria, KS
|
| Posted: Mon Feb 25, 2008 2:13 pm Post subject: bundle |
|
|
Re #957
An xy-chain with pincers (3) at R7C2 and R8C5 eliminates 3's from R7C4 and R8C3, solving the puzzle in one step.
Re #959
UR then my old friend an xy-chain and the puzzle melts.
xy-chains may not be sophisticated but they often work more directly than other techniques.
Earl
Last edited by Earl on Mon Feb 25, 2008 2:36 pm; edited 1 time in total |
|
| Back to top |
|
|
ravel
Joined: 21 Apr 2006
Posts: 536
|
| Posted: Mon Feb 25, 2008 2:24 pm Post subject: |
|
|
#959 daily competition
| Code: | *-------------------------------------------*
| 5 6 1 | 2 48 7 | 34 38 9 |
| 4 3 8 | 1 9 6 | 2 #57 #57 |
| 2 9 7 | 5 48 3 | 14 68 16 |
|------------+--------------+---------------|
| 18 7 2 | 38 6 14 | 9 #345 #35 |
| 9 4 6 | 37 17 5 | 13 2 8 |
| 3 18 5 | 89 2 149 | 7 46 16 |
|------------+--------------+---------------|
| 7 2 9 | 6 3 8 | 5 1 4 |
| 18 18 4 | 79 5 29 | 6 #37 #237 |
| 6 5 3 | 4 17 12 | 8 9 27 |
*-------------------------------------------*
|
DP 57,35,37: r4c7<>3 (strong link for 3 in r8 or c9), r8c9<>7 (strong link for 5 in r4 or c8).
Leaves a BUG+1.
feb 22 daily telegraph
| Johan wrote: | | After <59> UR: |
Or x-wing for 9, r3c7<>9, r5c7<>9, leaves an interesting BUG+3 situation:
| Code: | +--------------------------+--------------------------+--------------------------+
| 3 8 6 | 5 4 7 | 9 2 1 |
| 59 7 15 | 2 3 19 | 8 6 4 |
| 4 19 2 | 16+8 69+8 18 | 3 7 5 |
+--------------------------+--------------------------+--------------------------+
| 59 2 7 | 3 1 59 | 4 8 6 |
| 6 19 15 | 48 89 458 | 2 3 7 |
| 8 4 3 | 7 2 6 | 5 1 9 |
+--------------------------+--------------------------+--------------------------+
| 27 3 9 | 46 67 24 | 1 5 8 |
| 17 5 4 | 18 78 3 | 6 9 2 |
| 12 6 8 | 9 5 12 | 7 4 3 |
+--------------------------+--------------------------+--------------------------+
|
Though there are only two 8's in column 6, an 8 in r3c6 would lead to a deadly BUG pattern. |
|
| Back to top |
|
|
|
FAQ
Search
Memberlist
Usergroups
Register
Profile
Log in to check your private messages
Log in
