dailysudoku.com Forum Index dailysudoku.com
Discussion of Daily Sudoku puzzles
 
 FAQFAQ   SearchSearch   MemberlistMemberlist   UsergroupsUsergroups   RegisterRegister 
 ProfileProfile   Log in to check your private messagesLog in to check your private messages   Log inLog in 

Free Press March 25, 2011

 
Post new topic   Reply to topic    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

PostPosted: Fri Mar 25, 2011 4:28 pm    Post subject: Free Press March 25, 2011 Reply with quote

Not yet started. Seems to be on the difficult side.
Code:
Puzzle: FP032511
+-------+-------+-------+
| . 1 4 | 9 . 5 | 7 . . |
| . . 5 | . . 4 | 3 . . |
| 7 . . | . . . | . . . |
+-------+-------+-------+
| 8 . . | 2 . . | . 6 . |
| 5 . . | . 9 . | . . 7 |
| . 2 . | . . 6 | . . 9 |
+-------+-------+-------+
| . . . | . . . | . . 8 |
| . . 9 | 5 . . | 1 . . |
| . . . | 1 . . | 4 2 . |
+-------+-------+-------+

Play this puzzle online at the Daily Sudoku site

Keith
Back to top
View user's profile Send private message
Marty R.



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

PostPosted: Fri Mar 25, 2011 7:17 pm    Post subject: Reply with quote

Coloring + two extensions; r3c3, r6c1, r56c3<>3
W-Wing (28); r3c6<>2
X-Wing; r7c5<>3
Back to top
View user's profile Send private message
keith



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

PostPosted: Sat Mar 26, 2011 2:06 pm    Post subject: Reply with quote

After basics:
Code:
+----------------+----------------+----------------+
| 236  1    4    | 9    23   5    | 7    8    26   |
| 9    68   5    | 67   278  4    | 3    1    26   |
| 7    368  2368 | 36   1    238  | 9    5    4    |
+----------------+----------------+----------------+
| 8    9    37   | 2    4    37   | 5    6    1    |
| 5    346  36   | 8    9    1    | 2    34   7    |
| 134  2    137  | 37   5    6    | 8    34   9    |
+----------------+----------------+----------------+
| 123  5    123  | 4    237  237  | 6    9    8    |
| 246  468  9    | 5    268  28   | 1    7    3    |
| 36   7    368  | 1    368  9    | 4    2    5    |
+----------------+----------------+----------------+

A skyscraper on 8 and a couple of X-wings on 3 will do it.

Keith
Back to top
View user's profile Send private message
tlanglet



Joined: 17 Oct 2007
Posts: 2468
Location: Northern California Foothills

PostPosted: Sat Mar 26, 2011 2:10 pm    Post subject: Reply with quote

Here is a fun one step and a funner two step solution..........

Code:
AUR(36)r35c23[(3)r5c8=(28)r3c23]-als(28=3)r3c236-r3c4=r6c4; r6c8<>3

or

anp(68=3)r23c2-r3c4=r6c4-r4c6=(3)r4c3-(3=6)r5c3; r5c2<>6
                                     \
                                       -als(36=8)r59c3; r8c2<>8

anp(46=2)r8c12-(2=8)r8c6-als(8=23)b2q29-(3=6)r3c4-(6=7)r2c4-(7=3)r6c4-r4c6=r4c3-(3=4)r5c2-r8c2=(4)r8c1; Conflict r8c1<>2


Ted
Back to top
View user's profile Send private message
keith



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

PostPosted: Sat Mar 26, 2011 2:36 pm    Post subject: Reply with quote

Another two-stepper is a swordfish on 3, then the skyscraper on 8.

Keith
Back to top
View user's profile Send private message
ronk



Joined: 07 May 2006
Posts: 398

PostPosted: Sat Mar 26, 2011 11:10 pm    Post subject: Reply with quote

tlanglet wrote:
Here is a fun one step
...
AUR(36)r35c23[(3)r5c8=(28)r3c23]-als(28=3)r3c236-r3c4=r6c4; r6c8<>3

I'm quite sure the portion I highlighted in red is invalid. A "quantum naked triple" requires four cells, including the two cells of the AUR. The reason for the "quantum" term is that the two AUR cells are, for purposes of the naked triple, counted as one cell.

If you backtest your deduction by setting r6c8=3 and follow the inferences backwards, you'll see there is no resultant DP.
Back to top
View user's profile Send private message
storm_norm



Joined: 18 Oct 2007
Posts: 1741

PostPosted: Sun Mar 27, 2011 12:14 am    Post subject: Reply with quote

sometimes chains such as the ones posted by Ted are nice to see as a graphic.

here they are.

Code:
anp(68=3)r23c2-r3c4=r6c4-r4c6=(3)r4c3-(3=6)r5c3; r5c2<>6
                                     \
                                       -als(36=8)r59c3; r8c2<>8





Code:
anp(46=2)r8c12-(2=8)r8c6-als(8=23)b2q29-(3=6)r3c4-(6=7)r2c4-(7=3)r6c4-r4c6=r4c3-(3=4)r5c2-r8c2=(4)r8c1; Conflict r8c1<>2


Back to top
View user's profile Send private message
daj95376



Joined: 23 Aug 2008
Posts: 3854

PostPosted: Sun Mar 27, 2011 3:04 am    Post subject: Reply with quote

Except for a few instances, I never did warm up to graphs.

Here's the grid after basics. My concern is that r3c4=36 will prevent any possibility of Ted's UR existing.

Code:
 +--------------------------------------------------------------+
 |  236   1     4     |  9     23    5     |  7     8     26    |
 |  9     68    5     |  67    278   4     |  3     1     26    |
 |  7    *36+8 *36+28 |  36    1     238   |  9     5     4     |
 |--------------------+--------------------+--------------------|
 |  8     9     37    |  2     4     37    |  5     6     1     |
 |  5    *36+4 *36    |  8     9     1     |  2     34    7     |
 |  134   2     137   |  37    5     6     |  8     34    9     |
 |--------------------+--------------------+--------------------|
 |  123   5     123   |  4     237   237   |  6     9     8     |
 |  246   468   9     |  5     268   28    |  1     7     3     |
 |  36    7     368   |  1     368   9     |  4     2     5     |
 +--------------------------------------------------------------+
 # 49 eliminations remain

As for his first anp() chain, I believe he missed two eliminations. Still, an interesting find.

Code:
anp(68=3)r23c2-r3c4=r6c4-r4c6=(3)r4c3-(3=6)r5c3; r3c3,r5c2<>6
                                     \
                                       -als(36=8)r59c3; r3c3,r8c2<>8

As far as I'm concerned, Eureka notation is insufficient when a sequence of assignments are present.
I submit this notation for Ted's second chain:

Code:
(68=3)r23c2-r3c4=r6c4-r4c6=(3+6+8)r459c3; r3c3<>68, r5c2<>6, r8c2<>8
Back to top
View user's profile Send private message
tlanglet



Joined: 17 Oct 2007
Posts: 2468
Location: Northern California Foothills

PostPosted: Sun Mar 27, 2011 7:59 pm    Post subject: Reply with quote

I did not receive an email indicating any new posts following my initial submission; sorry about the delay in responding.

Ronk, I am not familiar with the term "quantum", and do not understand the reference to a naked triple.

My view of the pattern was that of using a mixed sis provided by the AUR(36)r35c23. For r5c23 I used external inferences which was only r5c8=3. For r3c23, I used internal inferences which resulted in a pseudocell (28)r3c23. I then combined the (28)r3c23 pseudocell, which I consider a single cell, with (238)r3c6 to form an als(28=3) consisting of three variable in two cells.

Where is the flaw in this perspective?

Danny, Yup, I did not post two additional deletions for the first chain. I did notice them when working the pattern and also noticed at that time that r3c3 would <>68 as a result of clean-up. I should have posted them anyway......

As I have noted before, I like your notation for a sequence of assignments and have used it in the past. I also like, and have sometimes used, your scheme for notating cells in a box (bNqC where N is the box number and C is any number of cells within the box). In the past, I have had PMs asking about both notations so I tend not to use them. So how to make it "standard" notation?

Finally, could you expand on your comment: Here's the grid after basics. My concern is that r3c4=36 will prevent any possibility of Ted's UR existing.

Ted
Back to top
View user's profile Send private message
daj95376



Joined: 23 Aug 2008
Posts: 3854

PostPosted: Sun Mar 27, 2011 10:49 pm    Post subject: Reply with quote

tlanglet wrote:
As I have noted before, I like your notation for a sequence of assignments and have used it in the past. I also like, and have sometimes used, your scheme for notating cells in a box (bNqC where N is the box number and C is any number of cells within the box). In the past, I have had PMs asking about both notations so I tend not to use them. So how to make it "standard" notation?

Ted,

There's "standard" Eureka notation, and then there's Eureka notation that's bastardized to describe things the notation was never intended to describe. In the latter case, I'm sure there are a number of people who contend they know what's "correct". I think your notation is understandable -- especially with the additional information you provide about your objectives.

As for myself, I try to find reasonable extensions to the notation that can be used in a number of situations. I also try to learn from the notation used by others. RonK has been a big help to me in this area because we interact across several forums and varying topics of discussion. Unfortunately, my memory is "swiss cheese" anymore and I've developed a bad habit lately of not taking notes dilligently; or worse, not remembering that a note exists.

Do what feels right when notating your solutions. It's my opinion that's what everyone else is doing! I'm sure others will offer suggestions, and you can pick-'n-choose what you like from them. For me, even if I don't decide to follow a suggestion, I've gained by learning how others view a situation.

Personally, I would love to see a thread where standard/basic Eureka notation is introduced and additional messages are added to describe complex or innovative extensions.

===== ===== ===== ===== ===== ===== =====

Quote:
Finally, could you expand on your comment: Here's the grid after basics. My concern is that r3c4=36 will prevent any possibility of Ted's UR existing.

I do not have a firm grasp on deadly patterns. However, I have managed to learn a few things about UR -- some of which you actually helped me to improve my understanding. One of the first _ Idea _ for me about URs/DPs was when ronk/eleven/someone told me that the presence of certain bivalue cells exclude the possibility of a UR/DP existing in a grid. Consider this grid w/o the extra candidates present in the UR cells:

Code:
 original grid
 +--------------------------------------------------------------+
 |  236   1     4     |  9     23    5     |  7     8     26    |
 |  9     68    5     |  67    278   4     |  3     1     26    |
 |  7    *36+8 *36+28 |  36    1     238   |  9     5     4     |
 |--------------------+--------------------+--------------------|
 |  8     9     37    |  2     4     37    |  5     6     1     |
 |  5    *36+4 *36    |  8     9     1     |  2     34    7     |
 |  134   2     137   |  37    5     6     |  8     34    9     |
 |--------------------+--------------------+--------------------|
 |  123   5     123   |  4     237   237   |  6     9     8     |
 |  246   468   9     |  5     268   28    |  1     7     3     |
 |  36    7     368   |  1     368   9     |  4     2     5     |
 +--------------------------------------------------------------+
 # 49 eliminations remain

Code:
 grid w/o extra UR candidates
 +--------------------------------------------------------------+
 |  236   1     4     |  9     23    5     |  7     8     26    |
 |  9     68    5     |  67    278   4     |  3     1     26    |
 |  7    *36   *36    | @36    1     238   |  9     5     4     |
 |--------------------+--------------------+--------------------|
 |  8     9     37    |  2     4     37    |  5     6     1     |
 |  5    *36   *36    |  8     9     1     |  2     34    7     |
 |  134   2     137   |  37    5     6     |  8     34    9     |
 |--------------------+--------------------+--------------------|
 |  123   5     123   |  4     237   237   |  6     9     8     |
 |  246   468   9     |  5     268   28    |  1     7     3     |
 |  36    7     368   |  1     368   9     |  4     2     5     |
 +--------------------------------------------------------------+

It's impossible to place <36> in r3c23 w/o eliminating both candidates in r3c4. So, in my mind, the extra candidates in the (*) cells aren't necessary to prevent the UR because the presence of r3c4=36 takes precedence.

Regards, Danny


Last edited by daj95376 on Sun Mar 27, 2011 11:11 pm; edited 1 time in total
Back to top
View user's profile Send private message
Marty R.



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

PostPosted: Sun Mar 27, 2011 11:03 pm    Post subject: Reply with quote

Quote:
when ronk/eleven/someone told me that the presence of certain bivalue cells exclude the possibility of a UR/DP existing in a grid.

Danny, you posted something along these lines some months ago and it's helped me in the sense of not wasting my time on DP moves that can't exist.
Back to top
View user's profile Send private message
ronk



Joined: 07 May 2006
Posts: 398

PostPosted: Sun Mar 27, 2011 11:59 pm    Post subject: Reply with quote

Marty R. wrote:
Quote:
when ronk/eleven/someone told me that the presence of certain bivalue cells exclude the possibility of a UR/DP existing in a grid.

Danny, you posted something along these lines some months ago and it's helped me in the sense of not wasting my time on DP moves that can't exist.

Since virtually all puzzles we work with have a unique solution, I doubt if that someone was me. With few exceptions IOW, there are ultimately no deadly patterns. However, associated with every almost-deadly-pattern is a strong inference set (SIS) which might yield a useful chain/network for an elimination.

Therefore, the existence of the <36> pair in r3c4 of the above puzzle is probably not, by itself, a "showstopper."
Back to top
View user's profile Send private message
tlanglet



Joined: 17 Oct 2007
Posts: 2468
Location: Northern California Foothills

PostPosted: Mon Mar 28, 2011 12:30 am    Post subject: Reply with quote

Danny,

I now understand your comment. However, my first reaction is "We already realize that a Deadly Pattern does not exit in a valid one solution puzzle". Knowing that fact is the bases for the Almost Deadly Pattern analysis we do using the strong inferences that insure this condition. In the situation you highlight in this puzzle, it is readily apparent that one of the internal digits in r3c23 must be true, but that fact does not help us in solving the puzzle nor, it seems to me, does it prevent us from doing a standard AUR analysis.

It may be that my simple minded viewpoint ignores some facet such as the recent discussion about the difference between a cell value being an original clue vs being derived. Maybe Ronk or Asellus or someone else will help us both.

Ted
[Edit: I started my post prior to seeing Ronk's latest input and our messages crossed in the internet. It is nice to have the answer to a question even before you complete posing the question.]
Back to top
View user's profile Send private message
Asellus



Joined: 05 Jun 2007
Posts: 865
Location: Sonoma County, CA, USA

PostPosted: Mon Mar 28, 2011 2:40 am    Post subject: Reply with quote

Ted wrote:
Where is the flaw in this perspective?

In a nutshell, the problem is that all or part of the two (28) groups in your weak inference exist on both sides of the inference:
AUR(36)r35c23[(3)r5c8=(28)r3c23]-als(28=3)r3c236
A weak inference requires that the two items cannot both be true. But it is possible for both sides to be true in this case due to the overlap. If that is not obvious, remember that the (28) group in r3c23 is true if either or both members are true. In the "either" cases, the r3c236 ALS is not necessarily reduced to a Locked Set. So it is still possible that it reduces to a LS containing 28 as well. (The ALS as you notate it must be considered a 4-digit, 3-cell ALS, not as a 3-digit ALS with a pseudo-cell.)

While, say, an ER strong inference can include the ERI digit on both sides, I cannot think of another example at the moment, and certainly cannot think of a weak inference example. So, be careful in such cases.

ronk wrote:
Therefore, the existence of the <36> pair in r3c4 of the above puzzle is probably not, by itself, a "showstopper."

I don't believe that "probably" is necessary. A no-solution pattern is just as deadly as a non-unique pattern. While my experience has been that cases such as this usually don't offer useful inferences, I have seen occasional examples in which they have.
Back to top
View user's profile Send private message Visit poster's website
keith



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

PostPosted: Mon Mar 28, 2011 2:47 am    Post subject: Reply with quote

Marty R. wrote:
Quote:
when ronk/eleven/someone told me that the presence of certain bivalue cells exclude the possibility of a UR/DP existing in a grid.

Danny, you posted something along these lines some months ago and it's helped me in the sense of not wasting my time on DP moves that can't exist.
I am not sure what this means. Does anyone have a link?

I believe that ADP's stand on their own For example:
Code:
+-----------+-----------+-----------+
|  .  .  .  |  12 .  12 |  .  .  .  |
|  .  .  .  |  .  .  .  |  .  .  .  |
|  .  .  .  |  .  .  .  |  .  .  .  |
+-----------+-----------+-----------+
|  .  .  .  |  .  .  .  |  .  .  .  |
|  .  .  .  |  .  .  .  |  .  .  .  |
|  .  .  .  | 123 . 124 |  .  .  .  |
+-----------+-----------+-----------+
|  .  .  .  |  .  .  .  |  .  .  .  |
|  .  .  .  |  .  .  .  |  .  .  .  |
|  .  .  .  |  .  .  .  |  .  .  .  |
+-----------+-----------+-----------+
is a UR and the lower two cells make a pseudocell 34.

If we have
Code:
+-----------+-----------+-----------+
|  .  .  .  |  12 .  12 |  .  .  .  |
|  .  .  .  |  .  .  .  |  .  .  .  |
|  .  .  .  |  .  .  .  |  .  .  .  |
+-----------+-----------+-----------+
|  .  .  .  |  .  12 .  |  .  .  .  |
|  .  .  .  |  .  .  .  |  .  .  .  |
|  .  .  .  | 123 . 124 |  .  .  .  |
+-----------+-----------+-----------+
|  .  .  .  |  .  .  .  |  .  .  .  |
|  .  .  .  |  .  .  .  |  .  .  .  |
|  .  .  .  |  .  .  .  |  .  .  .  |
+-----------+-----------+-----------+
that does not change, even though the UR cannot now generate a Deadly Pattern. (In fact, the pseudocell can now be inferred without the UR, simply by looking at B5.)

In other words, I agree with ronk and Asellus. In the puzzle of this thread, the presence of 36 in R3C4 does not affect any deductions made with the 36 UR of R35C23.

(In fact, the 36 in R3C4 generates a pseudocell 28 in R3C23, which is perhaps a little more information than you get from the UR itself.)

Keith
Back to top
View user's profile Send private message
keith



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

PostPosted: Mon Mar 28, 2011 3:25 am    Post subject: Reply with quote

tlanglet wrote:
Ronk, I am not familiar with the term "quantum", and do not understand the reference to a naked triple.

Ted

Ted,

In a Type 3 UR, two of the cells combine to form what I usually call a pseudocell. Others call it a quantum cell. "Quantum" predates "pseudo", and I have not seen "pseudocell" used outside this discussion group. And, I have not seen "quantum" used outside the context of a Type 3 UR.

A "pseudo puzzle" seems to be a puzzle that does not have a unique solution.

Keith

PS: "Naked triple" reminds me of the joke that you should never have a threesome with a physicist. They can't handle the three-body problem.
Back to top
View user's profile Send private message
tlanglet



Joined: 17 Oct 2007
Posts: 2468
Location: Northern California Foothills

PostPosted: Mon Mar 28, 2011 12:18 pm    Post subject: Reply with quote

To All,
This puzzle is a great reminder of something I have know for a long time, namely that "The way to learn is to make a mistake". Thanks to all that posted comments; for me, each contributed some new or confirming info, and hopefully added something useful for everyone. I now realize the error in my solution; I violated a rule I thought I understood but failed to "see" the implication is this situation.

Ted
Back to top
View user's profile Send private message
keith



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

PostPosted: Tue Mar 29, 2011 6:25 am    Post subject: Reply with quote

This old thread may be relevant:

http://www.dailysudoku.com/sudoku/forums/viewtopic.php?t=2028

I am not sure we are wiser now, but I don't seem to have been very wise then!

Keith
Back to top
View user's profile Send private message
Display posts from previous:   
Post new topic   Reply to topic    dailysudoku.com Forum Index -> Other puzzles All times are GMT
Page 1 of 1

 
Jump to:  
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


Powered by phpBB © 2001, 2005 phpBB Group