Daily Nightmare 2-5-07

Discuss the <a href="http://www.sudocue.net/daily.php">Daily Sudoku Nightmare</a> here
Post Reply
CathyW
Master
Master
Posts: 161
Joined: Wed Jan 31, 2007 5:39 pm
Location: Hertfordshire, UK

Daily Nightmare 2-5-07

Post by CathyW »

Don't look if you haven't done today's puzzle yet!!






I found a cool uniqueness move, an x-wing and finally a forcing chain to solve this puzzle :D

My walkthrough:

1. NP {28} in N5 -> r4c4=6 -> NP {39} in c6/N5 -> r1c6 = 1/6/7
2. NT {147} in c6/N8 -> r1c6 = 6, r9c4 = 2/5/8
3. NP {57} in c4/N2 -> r1c4 = 1, r23c5 = {39}, r9c4 = 2/8
4. HS r6c6 = 3, r4c6 = 9

5. Grid now:

Code: Select all

 
 *-----------------------------------------------------------------------------*
 | 2       379     3579    | 1       4       6       | 35      79      8       |
 | 457     3479    1       | 57      39      8       | 2       4679    4569    |
 | 4578    6       345789  | 57      39      2       | 345     1       459     |
 |-------------------------+-------------------------+-------------------------|
 | 48      234     348     | 6       1       9       | 7       5       249     |
 | 9       127     78      | 4       28      5       | 168     268     3       |
 | 148     5       6       | 28      7       3       | 148     2489    1249    |
 |-------------------------+-------------------------+-------------------------|
 | 14567   8       457     | 9       25      147     | 1456    3       12456   |
 | 14567   147     2       | 3       58      147     | 9       468     1456    |
 | 3       149     459     | 28      6       14      | 1458    248     7       |
 *-----------------------------------------------------------------------------*
r23c4 = {57} -> r23c1 cannot both be {57}
Either r2c1 = 4 or r3c1 = 4 or 8.
Since r4c1 = 4/8, 4 and 8 cannot go elsewhere in c1 -> r6c1 = 1 -> r78c1 = {567}
6. NT {278} in r5c235 -> r5c8 = 6, r5c7 = 1
7. 8 locked to r6 in N6 -> r6c4 = 2 -> several naked and hidden singles to here:

Code: Select all

 *--------------------------------------------------------------------*
 | 2      379    359    | 1      4      6      | 35     79     8      |
 | 457    3479   1      | 57     39     8      | 2      479    6      |
 | 4578   6      34589  | 57     39     2      | 345    1      459    |
 |----------------------+----------------------+----------------------|
 | 48     34     348    | 6      1      9      | 7      5      2      |
 | 9      2      7      | 4      8      5      | 1      6      3      |
 | 1      5      6      | 2      7      3      | 8      49     49     |
 |----------------------+----------------------+----------------------|
 | 57     8      45     | 9      2      147    | 6      3      145    |
 | 6      147    2      | 3      5      147    | 9      8      14     |
 | 3      149    459    | 8      6      14     | 45     2      7      |
 *--------------------------------------------------------------------*
8. x-wing on 5 in r1c37, r9c37 -> r3c3, r3c7, r7c3 <> 5 -> r7c3 = 4
9. Forcing chain:
If r7c6 = 1 -> r9c6 = 4, r9c7 = 5, r9c3 = 9, r9c2 = 1, r8c2 = 7
If r7c6 = 7 -> r7c1 = 5, r9c3 = 9, r9c2 = 1, r8c2 = 7
The rest is singles.
Para
Yokozuna
Yokozuna
Posts: 384
Joined: Wed Nov 08, 2006 7:42 pm
Location: The Netherlands

Re: Daily Nightmare 2-5-07

Post by Para »

Hi

I missed that UR and went through an alternative solution path. Tend to miss unique subset moves.
But for your walkthrough, that forcing chain can be replaced by: xy-wing: On R7C19 + R8C2: R8C9 <> 1

greetings

Para
CathyW
Master
Master
Posts: 161
Joined: Wed Jan 31, 2007 5:39 pm
Location: Hertfordshire, UK

Post by CathyW »

So it can! Thanks - should have spotted that. :roll:

I was pleased with the UR move - I can't remember using UR in combination with a bivalue cell before to make eliminations. I know some people don't hold with using uniqueness moves but I find them valuable and I'm 100% certain that Ruud would not supply a puzzle with more than one solution. :)
rep'nA
Hooked
Hooked
Posts: 49
Joined: Fri Jan 19, 2007 11:37 am
Location: Union City, California

Re: Daily Nightmare 2-5-07

Post by rep'nA »

Or at this point...

Code: Select all

 *--------------------------------------------------------------------*
 | 2      379    359    | 1      4      6      | 35     79     8      |
 | 457    3479   1      | 57     39     8      | 2      479    6      |
 | 4578   6      34589  | 57     39     2      | 345    1      459    |
 |----------------------+----------------------+----------------------|
 | 48     34     348    | 6      1      9      | 7      5      2      |
 | 9      2      7      | 4      8      5      | 1      6      3      |
 | 1      5      6      | 2      7      3      | 8      49     49     |
 |----------------------+----------------------+----------------------|
 | 57     8      45     | 9      2      147    | 6      3      145    |
 | 6      147    2      | 3      5      147    | 9      8      14     |
 | 3      149    459    | 8      6      14     | 45     2      7      |
 *--------------------------------------------------------------------*
you could skip both the x-wing and the forcing chain (or Para's alternative) with Subset Counting on r7c136 and r9c67 (or equivalently the ALS xz-rule). The candidates 1,4,5,7 appear with max multiplicities {1,2,2,1}. As there are 5 cells, any placement eliminating all of the 4's or all of the 5's must be false. Hence, r7c9, r9c3 <> 5, and the puzzle is reduced to singles.
"Obviousness is always the enemy to correctness."-Bertrand Russell
Post Reply