| View previous topic :: View next topic |
| Author |
Message |
wapati
Joined: 10 Jun 2008
Posts: 472
Location: Brampton, Ontario, Canada.
|
 Posted: Fri Jul 18, 2008 6:23 am Post subject: Another M-wing. |
 |
|
Jsudoku helped me generate this interesting puzzle.
I found an M-wing that got rid of:finned-x, W-wing, x-wing, sue de coq (2 of them) and an xy-wing. Good deal!
1 X-Wing
1 Swordfish
2 Skyscrapers
2 Two String Kites
1 Finned X-Wing
2 XY-Wings
1 Y-Wings
3 Finned Swordfish
2 Sue de Coq up to 4 cells
| Code: | . . .|9 4 .|. 8 .
. . .|. . 8|7 . 6
. . 1|. . .|. 3 .
-----+-----+-----
8 . .|. 9 .|. 6 .
6 . .|2 . 5|. . 3
. 1 .|. 8 .|. . 2
-----+-----+-----
. 3 .|. . .|5 . .
2 . 9|8 . .|. . .
. 4 .|. 2 6|. . .
|
|
|
| Back to top |
|
 |
nataraj
Joined: 03 Aug 2007
Posts: 1048
Location: near Vienna, Austria
|
| Posted: Fri Jul 18, 2008 7:31 pm Post subject: |
|
|
One more, I did not see the shortcut m-wing 
skyscraper (1)
multi-coloring (5)
xy-wing 35-57-73 (r2c5,r3c4,r4c4)
kite (7)
... got me here:
| Code: |
+--------------------------+--------------------------+--------------------------+
| 357 6 357 | 9 4 123 | 12 8 15 |
| 349 259 2345 | 15 35# 8 | 7 1459 6 |
| 459 8 1 | 57* 6 27 | 249 3 459 |
+--------------------------+--------------------------+--------------------------+
| 8 257 2457 | 3 9 147 | 14 6 1457 |
| 6 79 47 | 2 17 5 | 8 1479 3 |
| 3479 1 3457 | 6 8 47 | 49 4579 2 |
+--------------------------+--------------------------+--------------------------+
| 17+ 3 6 | 4 17+ 9 | 5 2 8 |
| 2 57* 9 | 8 35# 137 | 6 147 147 |
| 157 4 8 | 15-7 2 6 | 3 179 179 |
+--------------------------+--------------------------+--------------------------+
|
A w-wing (57*-57* via 5# in col 5) + transport 17-17
Then, after cleanup:
kite (1), multi-coloring(7)
w-wing(4) (47-47 in r5c3,r8c8) via strong link (7) col 2
coloring (7)
m-wing(4) r4c9,r2c8 and m-wing(9) r5c8,r3c9
xy-wing 57-79-95
I thought this was never going to end...
("I dub you wapiti de sade"  ) |
|
| Back to top |
|
|
storm_norm
Joined: 18 Oct 2007
Posts: 1741
|
| Posted: Fri Jul 18, 2008 8:57 pm Post subject: |
|
|
is this reasoning valid???
taking a look at the grid after basics, you have the useless w-wing on {5,7} in r3c4 and r8c2
| Code: | .---------------------.---------------------.---------------------.
| 357 6 357 | 9 4 1237 | 12 8 15 |
| 3459 259 2345 | 135 35 8 | 7 1459 6 |
| 4579 8 1 |#57 6 27 | 249 3 459 |
:---------------------+---------------------+---------------------:
| 8 257 23457 | 137 9 1347 | 14 6 1457 |
| 6 79 47 | 2 17 5 | 8 1479 3 |
| 34579 1 3457 | 6 8 347 | 49 4579 2 |
:---------------------+---------------------+---------------------:
| 17 3 6 | 4 1+{7} 9 | 5 2 8 |
| 2 #57 9 | 8 35 137 | 6 147 147 |
| 157 4 8 | 157 2 6 | 3 179 179 |
'---------------------'---------------------'---------------------' |
now they would eliminate any 7's they see. so if either one is true, then can't we say that the 7 in r7c5 has to be true?
it is the only 7 in box 8 that can't see the pincers of the w-wing.
any thoughts?? |
|
| Back to top |
|
|
wapati
Joined: 10 Jun 2008
Posts: 472
Location: Brampton, Ontario, Canada.
|
| Posted: Fri Jul 18, 2008 9:45 pm Post subject: |
|
|
| nataraj wrote: |
I thought this was never going to end...
("I dub you wapiti de sade" ) |
LOL, it is a sadistic puzzle !
| Code: | -------------------------- -------------------------- --------------------------
| 357 6 357 | 9 4 123 | 12 8 15 |
| 349 259 2345 | 15 35 8 | 7 1459 6 |
| 459 8 1 | 57 6 27 | 249 3 459 |
-------------------------- -------------------------- --------------------------
| 8 257 2457 | 3 9 147 | 14 6 1457 |
| 6 79 47 | 2 17 5 | 8 1479 3 |
| 3479 1 3457 | 6 8 47 | 49 4579 2 |
-------------------------- -------------------------- --------------------------
| 17 3 6 | 4 17 9 | 5 2 8 |
| 2 57 9 | 8 35 137 | 6 147 147 |
| 157 4 8 | 157 2 6 | 3 179 179 |
-------------------------- -------------------------- -------------------------- |
From your markup JSudoku (with my settings) uses:
Finned Swordfish on 1 in R148 and C679
Swordfish on 1 in C679 and R148
Finned Swordfish on 5 in C258 and R268
R6C8 = 5 (hidden single in R6)
R4C23 forms a hidden Pair on 25
Y-Wing with strong link on 7 in C2, R5C3 & R8C8
R5C3 = 4 (hidden single in R5)
Y-Wing with strong link on 7 in C2, R5C5 & R7C1
R7C5 = 7 , R5C5 = 1 , R7C1 = 1 , R3C4 = 7 (hidden single in C4)
R3C6 = 2 , R1C7 = 2 (hidden single in N3)
R4C7 = 1 (hidden single in C7)
5 of N2 locked in R2C45 -> not elsewhere in R2
to get to here:
| Code: | +----------------+----------------+----------------+
| 357 6 357 | 9 4 13 | 2 8 15 |
| 349 29 23 | 15 35 8 | 7 149 6 |
| 459 8 1 | 7 6 2 | 49 3 459 |
+----------------+----------------+----------------+
| 8 25 25 | 3 9 47 | 1 6 47 |
| 6 79 4 | 2 1 5 | 8 79 3 |
| 379 1 37 | 6 8 47 | 49 5 2 |
+----------------+----------------+----------------+
| 1 3 6 | 4 7 9 | 5 2 8 |
| 2 57 9 | 8 35 13 | 6 47 147 |
| 57 4 8 | 15 2 6 | 3 19 79 |
+----------------+----------------+----------------+ |
JSudoku (and the way I have it set up) expects:
Finned X-Wing on 7 in R58 and C28 with fin in R8C9
X-Wing on 7 in C28 and R58
Sue de Coq with 4 cells
Sue de Coq with 4 cells
XY-Wing on 5 with pivot R9C9
Instead I see an M-wing (of the type with the extra leg of 2 strong links).
Overall your method is shorter, I think.
I don't use coloring but I am learning that it is death to fish! |
|
| Back to top |
|
|
nataraj
Joined: 03 Aug 2007
Posts: 1048
Location: near Vienna, Austria
|
| Posted: Fri Jul 18, 2008 9:45 pm Post subject: |
|
|
Norm, I do not think your argument is valid.
True, if r3c4=5 then (by way of the w-wing) r8c2=7 and (by way of the 17-17 transport) r7c5=7.
But if r3c4=7 then all we know is that r9c4<>7. R5c7 could be 7 and r7c5=1. |
|
| Back to top |
|
|
storm_norm
Joined: 18 Oct 2007
Posts: 1741
|
| Posted: Fri Jul 18, 2008 10:38 pm Post subject: |
|
|
| nataraj wrote: | Norm, I do not think your argument is valid.
True, if r3c4=5 then (by way of the w-wing) r8c2=7 and (by way of the 17-17 transport) r7c5=7.
But if r3c4=7 then all we know is that r9c4<>7. R5c7 could be 7 and r7c5=1. |
ok, I have seen those kind of patterns before and wanted to throw that out there.
ehem...
coloring on 5 and 1
xy-wing
coloring 5 and 7
xy-chain |
|
| Back to top |
|
|
wapati
Joined: 10 Jun 2008
Posts: 472
Location: Brampton, Ontario, Canada.
|
| Posted: Sat Jul 19, 2008 12:59 am Post subject: |
|
|
I don't do colors so I have no idea how involved your steps are.
Is coloring in 5s about as quick as doing a swordfish on 5s?
If you are using pencil paper do you have to make, (and later erase?) lots of marks?
Thanks. |
|
| Back to top |
|
|
storm_norm
Joined: 18 Oct 2007
Posts: 1741
|
| Posted: Sat Jul 19, 2008 2:10 am Post subject: |
|
|
| wapati wrote: | I don't do colors so I have no idea how involved your steps are.
Is coloring in 5s about as quick as doing a swordfish on 5s?
If you are using pencil paper do you have to make, (and later erase?) lots of marks?
Thanks. |
coloring is a type of chain that involves only one candidate. its also called a singles chain.
here is a link that should get you started.
http://www.scanraid.com/Singles_Chains |
|
| Back to top |
|
|
storm_norm
Joined: 18 Oct 2007
Posts: 1741
|
| Posted: Sat Jul 19, 2008 2:12 am Post subject: |
|
|
| nataraj wrote: | Norm, I do not think your argument is valid.
True, if r3c4=5 then (by way of the w-wing) r8c2=7 and (by way of the 17-17 transport) r7c5=7.
But if r3c4=7 then all we know is that r9c4<>7. R5c7 could be 7 and r7c5=1. |
so could the 7 in r7c5 be strongly/weakly linked to the w-wing?? and therefore allow more chain type options? |
|
| Back to top |
|
|
keith
Joined: 19 Sep 2005
Posts: 3355
Location: near Detroit, Michigan, USA
|
| Posted: Sat Jul 19, 2008 7:49 am Post subject: |
|
|
I am just now looking at this. Full disclosure: I used Sudoku Susser to do the basics. I get to here: | Code: | +-------------------+-------------------+-------------------+
| 357 6 357 | 9 4 1237 | 12 8 15 |
| 3459 259 2345 | 135 35# 8 | 7 1459 6 |
| 4579 8 1 | 57@ 6 27 | 249 3 459 |
+-------------------+-------------------+-------------------+
| 8 257 23457 | 137 9 1347 | 14 6 1457 |
| 6 79 47 | 2 17 5 | 8 1479 3 |
| 34579 1 3457 | 6 8 347 | 49 4579 2 |
+-------------------+-------------------+-------------------+
| 17% 3 6 | 4 17% 9 | 5 2 8 |
| 2 57@ 9 | 8 35# 137 | 6 147 147 |
| 157 4 8 | 15-7 2 6 | 3 179 179 |
+-------------------+-------------------+-------------------+ |
(Already noted): The cells @ are a W-wing with strong link #. Extending by coloring (%) we can take out <7> in R9C4.
An XYZ-wing takes out <5> in R3C4. Further basics lead to: | Code: | +----------------+----------------+----------------+
| 357 6 3-57 | 9 4 13# | 2 8 15@ |
| 349 29 234 | 15 35 8 | 7 149 6 |
| 459 8 1 | 7 6 2 | 49 3 459 |
+----------------+----------------+----------------+
| 8 25% 25% | 3 9 47 | 1 6 47 |
| 6 79 47 | 2 1 5 | 8 479 3 |
| 3479 1 347 | 6 8 47 | 49 5 2 |
+----------------+----------------+----------------+
| 1 3 6 | 4 7 9 | 5 2 8 |
| 2 57% 9 | 8 35 13# | 6 147 147 |
| 57% 4 8 | 15@ 2 6 | 3 179 179 |
+----------------+----------------+----------------+ |
The cells @ are a W-wing with strong link #. Extending by coloring (%) we can take out <5> in R1C3. Solving the puzzzle.
Keith |
|
| Back to top |
|
|
keith
Joined: 19 Sep 2005
Posts: 3355
Location: near Detroit, Michigan, USA
|
| Posted: Sat Jul 19, 2008 8:15 am Post subject: |
|
|
| wapati wrote: | I don't do colors so I have no idea how involved your steps are.
Is coloring in 5s about as quick as doing a swordfish on 5s?
If you are using pencil paper do you have to make, (and later erase?) lots of marks?
Thanks. |
wapati,
You should learn coloring. Then, you can forget about swordfish (which can be found by multi-coloring). An X-wing and a skyscraper are the simplest examples of multi-coloring.
If you understand the extensions % in my previous post, you already understand coloring.
If you solve with pencil and paper, coloring does not involve marks and erasing on your grid. It does involve little sketches on the side. Take a look at this:
http://www.dailysudoku.com/sudoku/forums/viewtopic.php?t=2409
Please click through the links to see the the pdf images.
Keith |
|
| Back to top |
|
|
wapati
Joined: 10 Jun 2008
Posts: 472
Location: Brampton, Ontario, Canada.
|
| Posted: Sat Jul 19, 2008 1:48 pm Post subject: |
|
|
| keith wrote: |
If you solve with pencil and paper, coloring does not involve marks and erasing on your grid. It does involve little sketches on the side. |
Thanks, exactly what I wanted to know.
| keith wrote: |
Please click through the links to see the the pdf images.
Keith |
I liked your sheet but it is too small for my feeble vision.
I rotated it and doubled the size and it is just fine for me.
I hope you don't mind, and I won't distribute it.
Thanks again.
BTW, I looked at scanraid for multi-colors and saw the final example as a reason to learn this method.
http://www.scanraid.com/Multi_Colouring_Strategy
The rest I saw very quickly as simple patterns that I already know. |
|
| Back to top |
|
|
|
FAQ
Search
Memberlist
Usergroups
Register
Profile
Log in to check your private messages
Log in
