| View previous topic :: View next topic |
| Author |
Message |
keith
Joined: 19 Sep 2005
Posts: 3355
Location: near Detroit, Michigan, USA
|
 Posted: Sat Sep 27, 2008 7:24 pm Post subject: LA Times / Freep 26 Sep, 2008 |
 |
|
Not yet attempted: | Code: | Puzzle: FP092608
+-------+-------+-------+
| . . . | . . . | 8 . . |
| . 5 . | 6 . . | 9 1 . |
| 6 3 . | . 1 . | . . . |
+-------+-------+-------+
| . . . | 8 . . | . . 2 |
| . 9 . | . 6 . | 5 8 . |
| 4 . . | 5 . 3 | . . . |
+-------+-------+-------+
| . . . | . 3 . | . 4 1 |
| . 1 6 | . . 7 | . 5 . |
| . . 7 | . . . | . . . |
+-------+-------+-------+ |
Keith |
|
| Back to top |
|
 |
nataraj
Joined: 03 Aug 2007
Posts: 1048
Location: near Vienna, Austria
|
| Posted: Sat Sep 27, 2008 8:20 pm Post subject: |
|
|
This (after one kite r5c1<>7 ) is tough:
| Code: |
+--------------------------+--------------------------+--------------------------+
| 1279 27 1249 | 3 24579 2459 | 8 267 4567 |
| 278 5 248 | 6 2478 248 | 9 1 3 |
| 6 3 2489 | 24-79 1 24589 | 24a 27* 457 |
+--------------------------+--------------------------+--------------------------+
| 157 67 15 | 8 479 149 | 1346b 3679 2 |
| 12 9 3 | 1247* 6 124 | 5 8 47c |
| 4 2678 128 | 5 279 3 | 16 679 679 |
+--------------------------+--------------------------+--------------------------+
| 2589 28 2589 | 29 3 6 | 7 4 1 |
| 239 1 6 | 249 2489 7 | 23 5 89 |
| 239 4 7 | 129 2589 12589 | 236 2369 689 |
+--------------------------+--------------------------+--------------------------+
|
There is an elimination based on strong links and/or bi-values in 4 and 7 in col 7, box 6 and row 5:
(7=2)r3c8-(2=4)r3c7-(4)r4c7=(4-7)r5c9=(7)r5c4; r3c4<>7
[Notice that if r3c7=4 then r5c9 is also 4. (Most "wings" start with an observation like this one)
But if r5c9=4 it cannot be 7 and then r5c4=7.
How will r3c7 be 4 ? Only if it is not 2 and that happens when r3c8 is not 7.
Take it all together and we get: if r3c8<>7 then r5c4=7]
7 is now a naked single in r5c4.
After cleanup, another roadblock here:
| Code: |
+--------------------------+--------------------------+--------------------------+
| 1279 27* 1249 | 3 24579 2459 | 8 26 56 |
| -278* 5 248 | 6 2478 248 | 9 1 3 |
| 6 3 289 | 29 1 2589 | 4 27 57 |
+--------------------------+--------------------------+--------------------------+
| 157 67 15 | 8 49 149 | 136 379 2 |
| 12 9 3 | 7 6 12 | 5 8 4 |
| 4 2678 128 | 5 29 3 | 16 679 679 |
+--------------------------+--------------------------+--------------------------+
| 2589b 28a 2589 | 29 3 6 | 7 4 1 |
| 239 1 6 | 4 289 7 | 23 5 89 |
| 239 4 7 | 1 2589 2589 | 236 369 689 |
+--------------------------+--------------------------+--------------------------+
|
Again, a short AIC is the best I can do:
(7=2)r1c2-(2=8)r7c2-(8)r7c1=(8)r2c1; r2c1<>2
With 2 gone from r2c1, there is now an xy-wing 78-27-28 and the puzzle gets going once more.
Finally, here
| Code: |
+--------------------------+--------------------------+--------------------------+
| 19 7 149 | 3 459 459 | 8 2 6 |
| 8 5 24 | 6 7 24 | 9 1 3 |
| 6 3 29 | 29 1 8 | 4 7 5 |
+--------------------------+--------------------------+--------------------------+
| 7 6 5 | 8 49 149 | 13 39 2 |
| 12 9 3 | 7 6 12 | 5 8 4 |
| 4 28 18 | 5 29 3 | 16 69 7 |
+--------------------------+--------------------------+--------------------------+
| 5 28 89 | 29 3 6 | 7 4 1 |
| 239 1 6 | 4 289 7 | 23 5 89 |
| 239 4 7 | 1 2589 259 | 236 36 89 |
+--------------------------+--------------------------+--------------------------+
|
there is a UR (89) type 4, but coloring on 2 yields a better harvest.
Awfully advanced for a newspaper sudoku, if you ask me ... I must have missed something there. |
|
| Back to top |
|
|
Marty R.
Joined: 12 Feb 2006
Posts: 5770
Location: Rochester, NY, USA
|
| Posted: Sun Sep 28, 2008 12:08 am Post subject: |
|
|
| I can't make a dent in this thing. I had a worthless ER and Finned X-Wing. I tried it a second time to see if I might've missed something, but it was the same. |
|
| Back to top |
|
|
keith
Joined: 19 Sep 2005
Posts: 3355
Location: near Detroit, Michigan, USA
|
| Posted: Sun Sep 28, 2008 2:59 am Post subject: |
|
|
I think we should throw this one on the trash pile.
I have checked that I posted the correct puzzle. Sudoku Susser uses four very obscure chains to solve it.
RIP.
Keith |
|
| Back to top |
|
|
daj95376
Joined: 23 Aug 2008
Posts: 3854
|
| Posted: Sun Sep 28, 2008 3:43 am Post subject: |
|
|
Epilog:
| Code: | after basics
*--------------------------------------------------------------------*
| 1279 27 1249 | 3 24579 2459 | 8 267 4567 |
|*278# 5 248 | 6 *2478# 248 | 9 1 3 |
| 6 3 2489 |*2479 1 24589 | 24 27 457 |
|----------------------+----------------------+----------------------|
| 157 67 15 | 8 479# 149 | 1346 3679 2 |
| 12-7 9 3 |*1247# 6 124 | 5 8 47 |
| 4 2678 128 | 5 279# 3 | 16 679 679 |
|----------------------+----------------------+----------------------|
| 2589 28 2589 | 29 3 6 | 7 4 1 |
| 239 1 6 | 249 2489 7 | 23 5 89 |
| 239 4 7 | 129 2589 12589 | 236 2369 689 |
*--------------------------------------------------------------------*
2-String Kite (*) or Empty Rectangle (#) => [r5c1]<>7
[r2c1]=2, [r5c1]=1, [r4c3]=5, [r7c3]<>5, [r7c1]=5, [r2c1]=8 => [r2c1]<>2
XY-Wing [r1c2]/[r2c1]+[r7c2] => [r7c1]<>8
Skyscraper => [r89c5]<>2
|
|
|
| Back to top |
|
|
Marty R.
Joined: 12 Feb 2006
Posts: 5770
Location: Rochester, NY, USA
|
| Posted: Sun Sep 28, 2008 3:55 am Post subject: |
|
|
How do you differentiate between a 2-string kite and a kite? What I see with the (*) cells is what I was taught was a simple case of strong links, except the two strong links are perpendicular, as opposed to a parallel skyscraper. This was later named a kite to the best of my knowledge. |
|
| Back to top |
|
|
daj95376
Joined: 23 Aug 2008
Posts: 3854
|
| Posted: Sun Sep 28, 2008 4:22 pm Post subject: |
|
|
| Marty R. wrote: |
How do you differentiate between a 2-string kite and a kite? What I see with the (*) cells is what I was taught was a simple case of strong links, except the two strong links are perpendicular, as opposed to a parallel skyscraper. This was later named a kite to the best of my knowledge. |
I learned it as a 2-String Kite. It gets its name from the box being the kite and the two strong links being visualized as the two strings.
In Sudopedia, under Solving Techniques: Single Digit Patterns, it lists: Skyscraper, 2-String Kite, and Empty Rectangle. For the most part, they are all specialized forms of Turbot Fish.
I just assumed that everyone here was abbreviating it to kite.
Note: Many patterns have differing definitions and are sometimes implemented contrary to their definitions.
Last edited by daj95376 on Mon Sep 29, 2008 8:48 am; edited 1 time in total |
|
| Back to top |
|
|
arkietech
Joined: 31 Jul 2008
Posts: 1834
Location: Northwest Arkansas USA
|
| Posted: Sun Sep 28, 2008 7:52 pm Post subject: |
|
|
daj95376 said | Quote: | In Sudopedia, under Solving Techniques: Single Digit Patterns, it lists: Skyscraper, 2-String Kite, and Empty Rectangle. They are all specialized forms of Turbot Fish.
|
Can anyone show me an example where a kite or skyscraper is not an empty rectangle?
dan |
|
| Back to top |
|
|
cgordon
Joined: 04 May 2007
Posts: 769
Location: ontario, canada
|
| Posted: Sun Sep 28, 2008 8:29 pm Post subject: |
|
|
| Whoa! I spent a long time on this. I probably found the same useless ER as Marty - then nothing. Why would the LA Times produce a Suduko whose solution is unsolvable for the vast majority of their readers? Inexcusable I say! |
|
| Back to top |
|
|
keith
Joined: 19 Sep 2005
Posts: 3355
Location: near Detroit, Michigan, USA
|
| Posted: Sun Sep 28, 2008 9:39 pm Post subject: |
|
|
| arkietech wrote: | daj95376 said | Quote: | In Sudopedia, under Solving Techniques: Single Digit Patterns, it lists: Skyscraper, 2-String Kite, and Empty Rectangle. They are all specialized forms of Turbot Fish.
|
Can anyone show me an example where a kite or skyscraper is not an empty rectangle?
dan |
@ is a skyscraper, # are the eliminations.
| Code: | +-------+-------+-------+
| . @ . | . . . | # . # |
| . . . | . . . | . . . |
| # . # | . . . | . @ . |
+-------+-------+-------+
| . . . | . . . | . . . |
| . . . | . . . | . . . |
| . . . | . . . | . . . |
+-------+-------+-------+
| . . . | . . . | . . . |
| . . . | . . . | . . . |
| . @ . | . . . | . @ . |
+-------+-------+-------+ |
Another thing to consider is that an ER makes an elimination in only one cell..
Keith |
|
| Back to top |
|
|
Asellus
Joined: 05 Jun 2007
Posts: 865
Location: Sonoma County, CA, USA
|
| Posted: Sun Sep 28, 2008 10:04 pm Post subject: |
|
|
| nataraj wrote: | Again, a short AIC is the best I can do:
(7=2)r1c2-(2=8)r7c2-(8)r7c1=(8)r2c1; r2c1<>2 |
This AIC is valid, but not the elimination. It is <7> that is eliminated from r2c1. This is why it can be good practice to include the weak link discontinuity in your Eureka notation:
(7)r2c1 - (7=2)r1c2 - (2=8)r7c2 - (8)r7c1=(8-7)r2c1; r2c1<>7
Danny showed that there is also an AIC in the same grid that eliminates that <2>:
(2)r2c1 - (2=1)r5c1 - (1=5)r4c3 - (5)r7c3=(5-8)r7c1=(8-2)r2c1; r2c1<>2
Thus, r2c1=8.
Rather than two separate AICs, this can be done as a single (though not as easy to see) AIC:
(8)r2c1=(8)r23c3 - ALS[(8)r6c3=(1)r5c1|r6c3] - (1=5)r4c3 - (5)r7c3=(5-8)r7c1=(8)r2c1; r2c1=8
I expect that this is a good puzzle for practicing Medusa multicoloring. |
|
| Back to top |
|
|
Asellus
Joined: 05 Jun 2007
Posts: 865
Location: Sonoma County, CA, USA
|
| Posted: Sun Sep 28, 2008 10:30 pm Post subject: |
|
|
In nataraj's first grid (after the 7 Kite), there is a not very helpful Finned X-Wing that removes <7> from r1c9.
nataraj's <7> elimination in r3c4 can be revealed with a simple Medusa multi-coloring. However, the <4> can be eliminated from that same cell with an AIC that is even simpler in the sense that it is a basic Medusa elimination:
(4)r3c4 - (4)r3c7=(4)r4c7 - (4=7)r5c9 - (7)r5c4=(7-4)r3c4; r3c4<>4 |
|
| Back to top |
|
|
arkietech
Joined: 31 Jul 2008
Posts: 1834
Location: Northwest Arkansas USA
|
| Posted: Sun Sep 28, 2008 11:14 pm Post subject: |
|
|
keith said: | Quote: | Another thing to consider is that an ER makes an elimination in only one cell..
|
Doesn't a turbot also remove only one? Can't overlapping turbots and ers be considered? This is interesting  |
|
| Back to top |
|
|
daj95376
Joined: 23 Aug 2008
Posts: 3854
|
| Posted: Sun Sep 28, 2008 11:28 pm Post subject: |
|
|
| keith wrote: | | Another thing to consider is that an ER makes an elimination in only one cell. |
Keith, I wish you hadn't opened this bag of worms.
[Withdrawn:] I don't want to get into another definition/name debate!
Last edited by daj95376 on Mon Sep 29, 2008 2:35 am; edited 2 times in total |
|
| Back to top |
|
|
daj95376
Joined: 23 Aug 2008
Posts: 3854
|
| Posted: Sun Sep 28, 2008 11:40 pm Post subject: |
|
|
| arkietech wrote: | Can anyone show me an example where a kite or skyscraper is not an empty rectangle?
|
In my PM where I have a 2-String Kite (*) and an Empty Rectangle (#) marked, if cell [r4c6] had contained a 7 as well, then the 2-String Kite would still be valid but the Empty Rectangle would no longer be valid. |
|
| Back to top |
|
|
arkietech
Joined: 31 Jul 2008
Posts: 1834
Location: Northwest Arkansas USA
|
| Posted: Mon Sep 29, 2008 1:03 am Post subject: |
|
|
| Thanks to all for the help. I will close this bag of worms. |
|
| Back to top |
|
|
nataraj
Joined: 03 Aug 2007
Posts: 1048
Location: near Vienna, Austria
|
| Posted: Mon Sep 29, 2008 5:24 am Post subject: |
|
|
| Asellus wrote: |
This AIC is valid, but not the elimination. It is <7> that is eliminated from r2c1. This is why it can be good practice to include the weak link discontinuity in your Eureka notation |
Absolutely right.  (And I'll adapt another good practice: not to post after midnight)
Thanks, Asellus. |
|
| Back to top |
|
|
keith
Joined: 19 Sep 2005
Posts: 3355
Location: near Detroit, Michigan, USA
|
| Posted: Mon Sep 29, 2008 5:36 am Post subject: |
|
|
| daj95376 wrote: | | keith wrote: | | Another thing to consider is that an ER makes an elimination in only one cell. |
Keith, I wish you hadn't opened this bag of worms.
[Withdrawn:] I don't want to get into another definition/name debate! |
Why not? I, for one, have learned recently that an ER is a distinct pattern. (I had thought that I could ignore ER's, for I would always find the elimination by other means, like a kite. Not true.)
Anyway, I think the term "skyscraper" was coined by Havard in his classic explanation of strong links. Seems pretty clear to me.
http://www.sudoku.com/boards/viewtopic.php?t=3326
He acknowledges a connection to Turbot fish. (The problem is, I think, no one ever goes looking for Turbot fish. We look for kites, skyscrapers and ER's.)
Keith |
|
| Back to top |
|
|
nataraj
Joined: 03 Aug 2007
Posts: 1048
Location: near Vienna, Austria
|
| Posted: Mon Sep 29, 2008 6:37 am Post subject: |
|
|
Thanks for the link, Keith! I think that is a great tutorial.
(remainder of original post deleted)
Last edited by nataraj on Mon Sep 29, 2008 10:23 am; edited 1 time in total |
|
| Back to top |
|
|
daj95376
Joined: 23 Aug 2008
Posts: 3854
|
| Posted: Mon Sep 29, 2008 8:31 am Post subject: |
|
|
| keith wrote: | Anyway, I think the term "skyscraper" was coined by Havard in his classic explanation of strong links. Seems pretty clear to me.
http://www.sudoku.com/boards/viewtopic.php?t=3326
He acknowledges a connection to Turbot fish. (The problem is, I think, no one ever goes looking for Turbot fish. We look for kites, skyscrapers and ER's.)
|
Havard's presentation is great. I just take exception with his statement:
| Quote: | This pattern is part of the Turbot Fish, and I have named it a "skyscraper" because of the way the two strong links looks a bit like two skyscrapers to me.
|
The definition of a Turbot Fish is a pentagon pattern that allows for at most one elimination from what I've been able to determine. This works okay for a (2-String) Kite and (basic) Empty Rectangle but it doesn't work IMO for the general case Skyscraper.
To perform the four eliminations in Havard's diagram of a Skyscraper, it would take 4x Turbot Fish eliminations ... or 2x Sashimi X-Wings ... or one Siamese Sashimi X-Wing. Yes, I said 2x Sashimi X-Wings ... and not one as indicated in Sudopedia.
With the introduction of grouped strong links, grouped versions of these old patterns have been adopted. I'm fine with that -- like in the puzzle CE_37 that I presented recently ... and everyone has ignored. I don't blame them. |
|
| Back to top |
|
|
|
FAQ
Search
Memberlist
Usergroups
Register
Profile
Log in to check your private messages
Log in
