| View previous topic :: View next topic |
| Author |
Message |
dongrave
Joined: 06 Mar 2014
Posts: 568
|
 Posted: Sat Aug 29, 2015 3:50 pm Post subject: Help! Another Evil one from 1sudoku.net |
 |
|
After seeing Clement's post from 1sudoku.net, I went and pulled down a few 'evil' puzzles that I had a tough time with. It looked like they all required chains to solve and here's one that I gave up on. Even my 'trial and error' chain method didn't solve it. Can this one be solved in one step?
| Code: |
nš521627 Evil
+-------+-------+-------+
| 6 . . | . . . | 2 . . |
| . 5 . | . 2 . | 7 . . |
| . . 2 | . . 5 | . 4 3 |
+-------+-------+-------+
| . . . | 6 4 . | 8 . . |
| . 6 . | 7 . 8 | . 1 . |
| . . 8 | . 5 3 | . . . |
+-------+-------+-------+
| 8 7 . | 4 . . | 1 . . |
| . . 9 | . 3 . | . 7 . |
| . . 6 | . . . | . . 8 |
+-------+-------+-------+
|
Last edited by dongrave on Sat Aug 29, 2015 4:30 pm; edited 1 time in total |
|
| Back to top |
|
 |
Marty R.
Joined: 12 Feb 2006
Posts: 5770
Location: Rochester, NY, USA
|
| Posted: Sat Aug 29, 2015 4:22 pm Post subject: |
|
|
| OK, I'm assuming that 1 in r5 is in c8. |
|
| Back to top |
|
|
dongrave
Joined: 06 Mar 2014
Posts: 568
|
| Posted: Sat Aug 29, 2015 4:29 pm Post subject: |
|
|
| Marty R. wrote: | | OK, I'm assuming that 1 in r5 is in c8. |
Whoops! Yes, it is Marty. Sorry about that. I'll go fix it. |
|
| Back to top |
|
|
Marty R.
Joined: 12 Feb 2006
Posts: 5770
Location: Rochester, NY, USA
|
| Posted: Sat Aug 29, 2015 8:33 pm Post subject: |
|
|
| Code: |
+----------------+----------+-----------+
| 6 1389 137 | 139 78 4 | 2 5 19 |
| 349 5 134 | 139 2 6 | 7 8 19 |
| 79 189 2 | 19 78 5 | 6 4 3 |
+----------------+----------+-----------+
| 23579 239 357 | 6 4 1 | 8 23 257 |
| 2345 6 345 | 7 9 8 | 35 1 245 |
| 147 14 8 | 2 5 3 | 9 6 47 |
+----------------+----------+-----------+
| 8 7 35 | 4 6 9 | 1 23 25 |
| 145 14 9 | 8 3 2 | 45 7 6 |
| 234 234 6 | 5 1 7 | 34 9 8 |
+----------------+----------+-----------+
|
Play this puzzle online at the Daily Sudoku site
Better check my basics, they often are incorrect..
I have a very nice two-stepper. The 14 UR in r48c12 offered
numerous possibilities but they all fizzed out.
Sashimi X-Wing on 3 in r47 = -3r5c3
That exposed a W-Wing in r5c3 and r8c7. SL4 in c2=> -5r5c7
This was a very enjoyable puzzle with all the possibilities. |
|
| Back to top |
|
|
bat999
Joined: 09 Jul 2015
Posts: 55
Location: UK
|
| Posted: Sat Aug 29, 2015 11:02 pm Post subject: |
|
|
Mine is a 2-stepper also. 
(These are probably 2-String Kites)
Step 1
| Code: | .-------------------.-----------.---------------.
| 6 38 137 | 39 78 4 | 2 5 19 |
| 34 5 134 | 39 2 6 | 7 8 19 |
| 79 89 2 | 1 78 5 | 6 4 3 |
:-------------------+-----------+---------------:
| 23579 239 357 | 6 4 1 | 8 23 257 |
| 2345 6 45-3 | 7 9 8 | a35 1 245 |
| 147 14 8 | 2 5 3 | 9 6 47 |
:-------------------+-----------+---------------:
| 8 7 d35 | 4 6 9 | 1 c23 25 |
| 145 14 9 | 8 3 2 | 45 7 6 |
| 234 23 6 | 5 1 7 | b34 9 8 |
'-------------------'-----------'---------------' |
(3)r5c7 = r9c7 - r7c8 = (3)r7c3 => -3 r5c3
Step 2
| Code: | .------------------.-----------.---------------.
| 6 38 137 | 39 78 4 | 2 5 19 |
| 34 5 134 | 39 2 6 | 7 8 19 |
| 79 89 2 | 1 78 5 | 6 4 3 |
:------------------+-----------+---------------:
| 23579 239 357 | 6 4 1 | 8 23 257 |
| 2345 6 4-5 | 7 9 8 | d35 1 245 |
| 147 14 8 | 2 5 3 | 9 6 47 |
:------------------+-----------+---------------:
| 8 7 a35 | 4 6 9 | 1 23 b25 |
| 145 14 9 | 8 3 2 | c45 7 6 |
| 234 23 6 | 5 1 7 | 34 9 8 |
'------------------'-----------'---------------' |
(5)r7c1 = r7c9 - r8c7 = (5)r5c7 => -5 r5c3; stte
---------------------------------------------------------------------
Those eliminations could have been done using 'Empty Rectangle' method (or whatever you prefer to call it).
Step 1
| Code: | .-------------------.-----------.--------------.
| 6 38 137 | 39 78 4 | 2 5 19 |
| 34 5 134 | 39 2 6 | 7 8 19 |
| 79 89 2 | 1 78 5 | 6 4 3 |
:-------------------+-----------+--------------:
| 23579 239 357 | 6 4 1 | 8 *23 257 |
| 2345 6 45-3 | 7 9 8 | 35 1 245 |
| 147 14 8 | 2 5 3 | 9 6 47 |
:-------------------+-----------+--------------:
| 8 7 *35 | 4 6 9 | 1 *23 25 |
| 145 14 9 | 8 3 2 | 45 7 6 |
| 234 23 6 | 5 1 7 | 34 9 8 |
'-------------------'-----------'--------------' |
Both of the threes in box 6 (r4c8,r5c7) will eliminate the 3 in r5c3 (r4c8 makes a detour*).
Step 2
| Code: | .-------------------.-----------.--------------.
| 6 38 137 | 39 78 4 | 2 5 19 |
| 34 5 134 | 39 2 6 | 7 8 19 |
| 79 89 2 | 1 78 5 | 6 4 3 |
:------------------+-----------+---------------:
| 23579 239 357 | 6 4 1 | 8 23 257 |
| 2345 6 4-5 | 7 9 8 | *35 1 245 |
| 147 14 8 | 2 5 3 | 9 6 47 |
:-------------------+-----------+--------------:
| 8 7 35 | 4 6 9 | 1 23 25 |
| *145 14 9 | 8 3 2 | *45 7 6 |
| 234 23 6 | 5 1 7 | 34 9 8 |
'-------------------'-----------'--------------' |
Both of the fives in box 7 (r7c3,r8c1) will eliminate the 5 in r5c3 (r8c1 makes a detour*).
 |
|
| Back to top |
|
|
Marty R.
Joined: 12 Feb 2006
Posts: 5770
Location: Rochester, NY, USA
|
| Posted: Sat Aug 29, 2015 11:50 pm Post subject: |
|
|
| Bat, both moves look like simple coloring to me. The 5 might be a two-string kite, but to me that kind of kite is really simple coloring by another name. |
|
| Back to top |
|
|
Marty R.
Joined: 12 Feb 2006
Posts: 5770
Location: Rochester, NY, USA
|
| Posted: Sun Aug 30, 2015 12:14 am Post subject: |
|
|
| Quote: | | Both of the threes in box 6 (r4c8,r5c7) will eliminate the 3 in r5c3 (r4c8 makes a detour*). |
What's the path with the 3 in r4c8?
| Quote: | | Both of the fives in box 7 (r7c3,r8c1) will eliminate the 5 in r5c3 (r8c1 makes a detour*). I have the same question about the 5 in r8c1. It doesn't see r5c3. |
|
|
| Back to top |
|
|
bat999
Joined: 09 Jul 2015
Posts: 55
Location: UK
|
| Posted: Sun Aug 30, 2015 5:21 am Post subject: |
|
|
| Marty R. wrote: |
What's the path with the 3 in r4c8?
I have the same question about the 5 in r8c1. |
From * to * to *. |
|
| Back to top |
|
|
Marty R.
Joined: 12 Feb 2006
Posts: 5770
Location: Rochester, NY, USA
|
| Posted: Sun Aug 30, 2015 5:56 am Post subject: |
|
|
| bat999 wrote: | | Marty R. wrote: |
What's the path with the 3 in r4c8?
I have the same question about the 5 in r8c1. |
From * to * to *. |
With both the 3 and the 5 I see only three cells in the asterisked chain and don't see pincers. |
|
| Back to top |
|
|
bat999
Joined: 09 Jul 2015
Posts: 55
Location: UK
|
| Posted: Sun Aug 30, 2015 5:54 pm Post subject: |
|
|
| Marty R. wrote: | | ... don't see pincers. |
The Empty Rectangle method doesn't use pincers. 
Empty Rectangles can deal with up to five of the relevant candidates in their box.
But these here both have only two of the relevant candidates in their box so they can be expressed as chains instead (with pincers at the chain ends).
Step 1
| Code: | .-------------------.-----------.---------------.
| 6 38 137 | 39 78 4 | 2 5 19 |
| 34 5 134 | 39 2 6 | 7 8 19 |
| 79 89 2 | 1 78 5 | 6 4 3 |
:-------------------+-----------+---------------:
| 23579 239 357 | 6 4 1 | 8 b23 257 |
| 2345 6 45-3 | 7 9 8 | a35 1 245 |
| 147 14 8 | 2 5 3 | 9 6 47 |
:-------------------+-----------+---------------:
| 8 7 d35 | 4 6 9 | 1 c23 25 |
| 145 14 9 | 8 3 2 | 45 7 6 |
| 234 23 6 | 5 1 7 | 34 9 8 |
'-------------------'-----------'---------------' |
(3)r5c7 = r4c8 - r7c8 = (3)r7c3 => -3 r5c3
Step 2
| Code: | .-------------------.-----------.--------------.
| 6 38 137 | 39 78 4 | 2 5 19 |
| 34 5 134 | 39 2 6 | 7 8 19 |
| 79 89 2 | 1 78 5 | 6 4 3 |
:-------------------+-----------+--------------:
| 23579 239 357 | 6 4 1 | 8 23 257 |
| 2345 6 4-5 | 7 9 8 | d35 1 245 |
| 147 14 8 | 2 5 3 | 9 6 47 |
:-------------------+-----------+--------------:
| 8 7 a35 | 4 6 9 | 1 23 25 |
| b145 14 9 | 8 3 2 | c45 7 6 |
| 234 23 6 | 5 1 7 | 34 9 8 |
'-------------------'-----------'--------------' |
(5)r7c3 = r8c1 - r8c7 = (5)r5c7 => -5 r5c3; stte |
|
| Back to top |
|
|
Marty R.
Joined: 12 Feb 2006
Posts: 5770
Location: Rochester, NY, USA
|
| Posted: Sun Aug 30, 2015 7:43 pm Post subject: |
|
|
OK, I'm familiar with ERs. I misinterpreted this statement to mean conventional chains that could ALSO be done by ERs.
| Quote: | | Those eliminations could have been done using 'Empty Rectangle' method (or whatever you prefer to call it). |
|
|
| Back to top |
|
|
dongrave
Joined: 06 Mar 2014
Posts: 568
|
| Posted: Sun Aug 30, 2015 8:59 pm Post subject: |
|
|
Thanks for the responses guys! Now I've got to go read about what you two are talking about. I've never heard of Sashimi X-Wings or Empty Rectangles - and I'm still no good at finding Kites or W-Wings - but at least I'm getting used to coloring since learning it from Marty about a year ago.
After searching for a long time, I finally found a solution similar to Bat's that I would almost consider a single-step solution (even though it involves two branches) because both of the branches begin with the same cell and end up removing candidates from that cell. If r5c3 is 3, then you have the contradiction 3r5c3-r7c3=r7c8-r4c8=r5c7-r5c3; and if it's 5, you have the contradiction 5r5c3-r7c3=r7c9-r45c9=r5c7-r5c3; therefore r5c3=4. Does that qualify as a one-stepper? (Oh well, probably not since it has two distinct braches with two separate contradictions.) Pretty ugly, isn't it? And this is actually a cleaned-up version of a 3-branch solution that I found first that began by assuming that r7c8=2 and ended up with two 4's in box 4. All I did was first verify that r7c8=3 solved it (to make sure it didn't leave us with another unsolved mess) and then assumed that r7c8<>3 and looked for the quickest possible contradiction. (Pretty elegant approach, huh?)  |
|
| Back to top |
|
|
Marty R.
Joined: 12 Feb 2006
Posts: 5770
Location: Rochester, NY, USA
|
| Posted: Sun Aug 30, 2015 9:53 pm Post subject: |
|
|
| Don, I wouldn't call it a one-stepper because no elimination was made after the first chain. Are you familiar with Skyscrapers? If not, check them out before Sashimi. Thanks for posting. It's always nice to get some puzzles here other than what the site publishes. |
|
| Back to top |
|
|
dongrave
Joined: 06 Mar 2014
Posts: 568
|
| Posted: Mon Aug 31, 2015 2:49 am Post subject: |
|
|
| Thanks for the advice Marty! I'll do that! I remember reading about Skyscrapers and Kites and W-Wings (and a couple of fish techniques too) a while ago but I'm sure I'll have to go over them at least a couple more times before they stick. |
|
| Back to top |
|
|
Marty R.
Joined: 12 Feb 2006
Posts: 5770
Location: Rochester, NY, USA
|
| Posted: Mon Aug 31, 2015 3:13 am Post subject: |
|
|
| Code: | .-------------------.-----------.---------------.
| 6 38 137 | 39 78 4 | 2 5 19 |
| 34 5 134 | 39 2 6 | 7 8 19 |
| 79 89 2 | 1 78 5 | 6 4 3 |
:-------------------+-----------+---------------:
| 23579 239 357 | 6 4 1 | 8 23 257 |
| 2345 6 45-3 | 7 9 8 | a35 1 245 |
| 147 14 8 | 2 5 3 | 9 6 47 |
:-------------------+-----------+---------------:
| 8 7 d35 | 4 6 9 | 1 c23 25 |
| 145 14 9 | 8 3 2 | 45 7 6 |
| 234 23 6 | 5 1 7 | b34 9 8 |
'-------------------'-----------'---------------' |
I'll cover the Sashimi really quickly. Note the 3s in r47. They'd form an X-Wing except for the extra candidates in r4c12. 3 can't be present in both r4c8 and r7c8. Therefore, 3r4c123 or r7c3 must be =3. R5c3 sees both and gets zapped. It's a fun move but doesn't come up as often as I'd like. Actually, it's a fancier name for a Finned X-Wing. |
|
| Back to top |
|
|
dongrave
Joined: 06 Mar 2014
Posts: 568
|
| Posted: Mon Aug 31, 2015 10:18 pm Post subject: |
|
|
| Marty R. wrote: | | Code: | .-------------------.-----------.---------------.
| 6 38 137 | 39 78 4 | 2 5 19 |
| 34 5 134 | 39 2 6 | 7 8 19 |
| 79 89 2 | 1 78 5 | 6 4 3 |
:-------------------+-----------+---------------:
| 23579 239 357 | 6 4 1 | 8 23 257 |
| 2345 6 45-3 | 7 9 8 | a35 1 245 |
| 147 14 8 | 2 5 3 | 9 6 47 |
:-------------------+-----------+---------------:
| 8 7 d35 | 4 6 9 | 1 c23 25 |
| 145 14 9 | 8 3 2 | 45 7 6 |
| 234 23 6 | 5 1 7 | b34 9 8 |
'-------------------'-----------'---------------' |
I'll cover the Sashimi really quickly. Note the 3s in r47. They'd form an X-Wing except for the extra candidates in r4c12. 3 can't be present in both r4c8 and r7c8. Therefore, 3r4c123 or r7c3 must be =3. R5c3 sees both and gets zapped. It's a fun move but doesn't come up as often as I'd like. Actually, it's a fancier name for a Finned X-Wing. |
Hey! Thanks for explaining that, Marty! That easy to understand! (How easy they'll be to find I'm sure is a different story though.) I've really got to read up on all these techniques that I'm not familiar with yet. Thanks again! |
|
| Back to top |
|
|
ronk
Joined: 07 May 2006
Posts: 398
|
| Posted: Sat Sep 05, 2015 5:41 pm Post subject: |
|
|
| Marty R. wrote: | | Code: | .-------------------.-----------.---------------.
| 6 38 137 | 39 78 4 | 2 5 19 |
| 34 5 134 | 39 2 6 | 7 8 19 |
| 79 89 2 | 1 78 5 | 6 4 3 |
:-------------------+-----------+---------------:
| 23579 239 357 | 6 4 1 | 8 23 257 |
| 2345 6 45-3 | 7 9 8 | a35 1 245 |
| 147 14 8 | 2 5 3 | 9 6 47 |
:-------------------+-----------+---------------:
| 8 7 d35 | 4 6 9 | 1 c23 25 |
| 145 14 9 | 8 3 2 | 45 7 6 |
| 234 23 6 | 5 1 7 | b34 9 8 |
'-------------------'-----------'---------------' |
I'll cover the Sashimi really quickly. Note the 3s in r47. .... |
Since the 3r4c3 candidate exists, the pattern is a finned x-wing. If it did not exist, then the pattern would be a sashimi x-wing. |
|
| Back to top |
|
|
Clement
Joined: 24 Apr 2006
Posts: 1111
Location: Dar es Salaam Tanzania
|
| Posted: Sun Sep 06, 2015 8:52 pm Post subject: nš521627-Evil |
|
|
| Code: |
+---------------+---------+-----------+
| 6 38 137 | 39 78 4 | 2 5 19 |
| 1349 5 134 | 39 2 6 | 7 8 19 |
| 79 89 2 | 1 78 5 | 6 4 3 |
+---------------+---------+-----------+
| 23579 239 357 | 6 4 1 | 8 23 25*7 |
| 2345 6 4-35 | 7 9 8 | 35# 1 245* |
| 147 14 8 | 2 5 3 | 9 6 47 |
+---------------+---------+-----------+
| 8 7 35# | 4 6 9 | 1 23 25* |
| 145 14 9 | 8 3 2 | 45 7 6 |
| 23*4 23* 6 | 5 1 7 | 3*4 9 8 |
+---------------+---------+-----------+
|
35 in r7c3 & r5c7 is a W-Wing with two grouped SL,s; 3 in row 9 and 5 in col 9; r5c3<>35; stte |
|
| Back to top |
|
|
Marty R.
Joined: 12 Feb 2006
Posts: 5770
Location: Rochester, NY, USA
|
| Posted: Mon Sep 07, 2015 2:15 am Post subject: |
|
|
| Nice move, Clement. |
|
| Back to top |
|
|
Clement
Joined: 24 Apr 2006
Posts: 1111
Location: Dar es Salaam Tanzania
|
| Posted: Mon Sep 07, 2015 8:13 am Post subject: |
|
|
| Marty R. wrote: | | Nice move, Clement. |
Thanks |
|
| Back to top |
|
|
|
FAQ
Search
Memberlist
Usergroups
Register
Profile
Log in to check your private messages
Log in
