| View previous topic :: View next topic |
| Author |
Message |
daj95376
Joined: 23 Aug 2008
Posts: 3854
|
 Posted: Wed Mar 31, 2010 12:53 am Post subject: Puzzle 10/03/30 ___ Chain |
 |
|
For those who like a chain after basics ... plus non-chain step(s).
| Code: | +-----------------------+
| 6 7 . | 3 2 . | 5 . 1 |
| 5 . . | . . . | . 4 . |
| . . . | . . . | . 7 3 |
|-------+-------+-------|
| 7 . . | 9 . 1 | 4 . . |
| 2 . . | . 4 . | . 1 . |
| . . . | 7 . . | . . . |
|-------+-------+-------|
| 4 . . | 1 . . | 2 . . |
| . 2 3 | . 5 . | . 6 9 |
| 9 . 5 | . . . | . 3 . |
+-----------------------+
|
Play this puzzle online at the Daily Sudoku site |
|
| Back to top |
|
 |
Mogulmeister
Joined: 03 May 2007
Posts: 1151
|
| Posted: Wed Mar 31, 2010 10:36 am Post subject: |
|
|
At your word Danny, a chain followed by something else - feels back to front but it works:
| Quote: | 1. AN AIC that creates a contradiction if you start with a 1:
(8=1)r3c1 - r8c1 = (1-7)r8c7 = r8c6 - r2c6 = (7-1)r2c5 = r2c3 - (1)r3c1
r3c1 <> 1
2. This opens up an X wing on 8's and the puzzle is done. |
|
|
| Back to top |
|
|
wapati
Joined: 10 Jun 2008
Posts: 472
Location: Brampton, Ontario, Canada.
|
| Posted: Wed Mar 31, 2010 3:19 pm Post subject: |
|
|
I had fun playing with this one, mostly because of the Finned Jelly on 8s.
I like almost W-wings and I found 3 all using the 18s in c1. |
|
| Back to top |
|
|
wapati
Joined: 10 Jun 2008
Posts: 472
Location: Brampton, Ontario, Canada.
|
| Posted: Wed Mar 31, 2010 5:18 pm Post subject: |
|
|
I had this in an earlier post with chains included but BBCode or HTML deleted a lot of it. I'm leaving out the chains this time. 
Jelly for r3c5 <>8
Then
xy-wing 16-8
W-wing using composite 19 endpoint.
89-6 xy-wing.
or
W-wing using composite 18 endpoint.
W-wing using composite 17 endpoint.
X-wing on 8s.
Every none-basic step involves 8s  |
|
| Back to top |
|
|
Mogulmeister
Joined: 03 May 2007
Posts: 1151
|
| Posted: Wed Mar 31, 2010 8:22 pm Post subject: |
|
|
| Yes it was all about the 8s - my AIC started on an 8 and did you notice the type 4 UR and guess what was eliminated ? |
|
| Back to top |
|
|
oaxen
Joined: 10 Jul 2006
Posts: 96
|
| Posted: Thu Apr 01, 2010 7:11 am Post subject: |
|
|
In the 36 free boxes there are 26 Two of a Kinds. 18 of them can be used for succesful one-stepper-chains.
Danny is it possible for your program to minimize or completely avoid Two of Kinds or stop the possibilities to use them for One-steppers? |
|
| Back to top |
|
|
Mogulmeister
Joined: 03 May 2007
Posts: 1151
|
| Posted: Thu Apr 01, 2010 7:41 am Post subject: |
|
|
Do you mean bivalues ? When I have completed basics I have 47 boxes left unsolved of whih 20 are bivalues.
I have to say that these puzzles are very enjoyable to those of us who like to go at it without solvers. |
|
| Back to top |
|
|
oaxen
Joined: 10 Jul 2006
Posts: 96
|
| Posted: Thu Apr 01, 2010 3:30 pm Post subject: Mogulmeister |
|
|
OK, may be bialvues is the correct word. (Same two numbers on a row, column or block)
I anyhow have 36 boxes left after basics.
And of course none of us use a solver. I normally solve a a puzzle and AFTER dry to diagnostice how many other possibilities there are for a one stepper. And in this case 18 where each one also can bring me to a happy end.
I also am very fond of Dannys puzzles, actually the only ones I game with. So I hope Danny can find ways to make them even more difficult. |
|
| Back to top |
|
|
daj95376
Joined: 23 Aug 2008
Posts: 3854
|
| Posted: Thu Apr 01, 2010 3:51 pm Post subject: |
|
|
oaxen:
Yes, I can make harder puzzles. Generating one that doesn't have a backdoor single from a bivalue cell is more difficult but not impossible. There are two backdoor singles in this puzzle. Have fun finding them.
Ext_Out Puzzle #27
| Code: | +-----------------------+
| 3 . 1 | . 4 7 | . . 8 |
| . 7 . | 8 . . | 1 . . |
| 5 . . | . . . | . . . |
|-------+-------+-------|
| . 1 . | 6 5 . | . . 2 |
| 6 . . | 2 . 8 | 5 . . |
| 7 . . | . 1 3 | 8 . . |
|-------+-------+-------|
| . 6 . | . 8 2 | 7 . . |
| . . . | . . . | . . . |
| 8 . . | 1 . . | . . 5 |
+-----------------------+
|
Play this puzzle online at the Daily Sudoku site |
|
| Back to top |
|
|
oaxen
Joined: 10 Jul 2006
Posts: 96
|
| Posted: Fri Apr 02, 2010 5:53 pm Post subject: |
|
|
Thank you Danny! It was amusing. I can solve the puzzle but only through successive eliminations. But what you are telling me is that if I somewhere find just one digit more (and I have two chances among the existing bivalues) the rest can be solved with pure bascis again to make it a one stepper.
I give up, please send the solution.
Lars |
|
| Back to top |
|
|
daj95376
Joined: 23 Aug 2008
Posts: 3854
|
| Posted: Fri Apr 02, 2010 7:52 pm Post subject: |
|
|
| oaxen wrote: | I give up, please send the solution.
|
| I wrote: | Generating one that doesn't have a backdoor single from a bivalue cell is more difficult but not impossible.
|
| Code: | +-----------------------------------------------------------------------+
| 3 29 1 | 59 4 7 | 269 2569 8 |
| 249 7 2469 | 8 2369 569 | 1 23459 349 |
| 5 8 2469 | 39 2369 1 | 249 2349 7 |
|-----------------------+-----------------------+-----------------------|
| 49 1 8 | 6 5 49 | 3 7 2 |
| 6 349 349 | 2 7 8 | 5 149 149 |
| 7 25 25 | 49 1 3 | 8 469 469 |
|-----------------------+-----------------------+-----------------------|
| 149 6 3459 | 3459 8 2 | 7 1349 1349 |
| 1249 23459 23459 | 7 369 4569 | 2469 8 13469 |
| 8 2349 7 | 1 369 469 | 2469 23469 5 |
+-----------------------------------------------------------------------+
# 103 eliminations remain
r2c1 = 9 simple SSTS backdoor (needs Naked Pair in solution)
r3c5 = 3 simple SSTS backdoor (Singles only)
|
|
|
| Back to top |
|
|
oaxen
Joined: 10 Jul 2006
Posts: 96
|
| Posted: Sat Apr 03, 2010 9:32 am Post subject: |
|
|
Thank you Danny. But in your solution you have "three of a kind" (What is the Sudokoword for that?) on rows, columns and blocks. How do you know to start with your suggestions? With my technique I can only start with "bivalues" to make it a one stepper. If the number I chose is wrong the other number must be correct.
Anyhow, this puzzle was impossible to solve as a one stepper and that is exactly what I am looking for. Definitely harder than the puzzles where I have 15-20 possible chances to find a successful start |
|
| Back to top |
|
|
daj95376
Joined: 23 Aug 2008
Posts: 3854
|
| Posted: Sat Apr 03, 2010 9:53 am Post subject: |
|
|
| oaxen wrote: | | How do you know to start with your suggestions? |
It's a logical extension of your approach. Once the search of bivalue cells fails, I simply expand my testing to candidates in increasingly larger poly-value cells. |
|
| Back to top |
|
|
oaxen
Joined: 10 Jul 2006
Posts: 96
|
| Posted: Sat Apr 03, 2010 3:54 pm Post subject: |
|
|
| daj95376 wrote: | | oaxen wrote: | | How do you know to start with your suggestions? |
It's a logical extension of your approach. Once the search of bivalue cells fails, I simply expand my testing to candidates in increasingly larger poly-value cells. |
OK, when you have a computersolver it works. But I have to start with pure bivalues. |
|
| Back to top |
|
|
|
FAQ
Search
Memberlist
Usergroups
Register
Profile
Log in to check your private messages
Log in
