X-Treme Dec 29

Welcome to the world of the unknown. Each day you need to solve <a href="http://www.sudocue.net/xfile.php">two mysteries</a>.<br>Take the X-R-Size or solve the X-Treme problem. Stuck? Found a nice solution? Tell us about it.
Post Reply
emm
Hooked
Hooked
Posts: 34
Joined: Wed Oct 11, 2006 6:01 pm

X-Treme Dec 29

Post by emm »

I’ve just looked at the solution to see where I went wrong on Dec 29 XT.

This is what I did - Type 3 UR -78- -> r5c4 cannot be 5 leaving 8.

Code: Select all

  *--------------------------------------------------------------------*
 | 678    389    469    | 2      1      5      | 4679   389    4678   |
 | 128    238    159    | 6      4      7      | 19     158    38     |
 | 1567   47     146    | 3      8      9      | 167    125    4567   |
 |----------------------+----------------------+----------------------|
 | 4      1      78*    | 578*   35     2      | 358    6      9      |
 | 3      6      2      | 58     9      4      | 58     7      1      |
 | 9      5      78*    | 178*   36     16     | 38     4      2      |
 |----------------------+----------------------+----------------------|
 | 27     89     3      | 15#    56     16     | 247    289    478    |
 | 1567   47     1456   | 9      2      8      | 167    135    3567   |
 | 1568   289    1569   | 4      7      3      | 1269   189    568    |
 *--------------------------------------------------------------------*
However the solution has r5c4 = 5 so … you can’t use URs on diagonals??
Ruud
Site Owner
Site Owner
Posts: 601
Joined: Fri Dec 30, 2005 10:21 pm

Post by Ruud »

The intended deadly pattern has a single cell in each diagonal, which means that one of the diagonals has a 7 outside the pattern and the other has an 8. These force the pattern to a single solution.

In Sudoku-X, it is not possible to form URs with cells that belong to one of the diagonals, but they could be used in larger deadly patterns (BUG Lite), like this example:

Code: Select all

X . .|. X .|. . .
. . .|. . .|. . .
. . X|. X .|. . .
-----+-----+-----
. . .|. . .|. . .
X . X|. . .|. . .
. . .|. . .|. . .
-----+-----+-----
. . .|. . .|. . .
. . .|. . .|. . .
. . .|. . .|. . .
But these are very rare, I think.
emm
Hooked
Hooked
Posts: 34
Joined: Wed Oct 11, 2006 6:01 pm

Post by emm »

Thanks Ruud.

Can you tell me how to get the rating on these puzzles. If I put them into SudoCue and click analyzer they all say easy :? I know that's not right!! My screen doesn't seem to show a rating in the bottom RH corner as it supposedly should.

Funny that our New Year is already in and you guys are just winding up!

Happy New Year already!
emm
Hooked
Hooked
Posts: 34
Joined: Wed Oct 11, 2006 6:01 pm

Post by emm »

... in the meantime ... I'll rate them myself

X-Treme Advanced

A = 'Aven't got a clue!
B = By golly that took ages!
C = Could it be I'm improving?
D = Doddle

1.1.07 = D

Image
emm
Hooked
Hooked
Posts: 34
Joined: Wed Oct 11, 2006 6:01 pm

Post by emm »

2.1.07 = C

dum de dum dum de dum
sudokuEd
Grandmaster
Grandmaster
Posts: 257
Joined: Mon Jun 19, 2006 11:06 am
Location: Sydney Australia

Post by sudokuEd »

Still stuck on the December 29 X-treme. emm, FYI, SudoCue rates it as "Unfair 6951" - or "EA" on your rating scale. Thankfully not eA! eA would have to be Oct 23 X-treme I'd think :D

Anyway, this is where I've been able to get to for Dec 29 X-treme. Help appreciated.

Code: Select all

.---------------.---------------.---------------.
| 68   389  469 | 2    1    5   | 4679 389  4678|
| 128  23   159 | 6    4    7   | 19   158  38  |
| 1567 47   146 | 3    8    9   | 16   2    456 |
&#58;---------------+---------------+---------------&#58;
| 4    1    78  | 578  35   2   | 358  6    9   |
| 3    6    2   | 58   9    4   | 58   7    1   |
| 9    5    78  | 178  36   16  | 38   4    2   |
&#58;---------------+---------------+---------------&#58;
| 27   89   3   | 15   56   16  | 247  89   47  |
| 167  47   1456| 9    2    8   | 167  135  356 |
| 1568 289  1569| 4    7    3   | 1269 189  568 |
'---------------'---------------'---------------'
Ruud
Site Owner
Site Owner
Posts: 601
Joined: Fri Dec 30, 2005 10:21 pm

Post by Ruud »

sudokuEd wrote:Still stuck on the December 29 X-treme.
I have been experimenting with a new Almost Locked Candidates move. You gave me a marvelous example:

Code: Select all

.---------------.---------------.---------------.
| 68   389  469 | 2    1    5   |-4679 389  4678|
| 128  23   159 | 6    4    7   |$19   158  38  |
| 1567 47   146 | 3    8    9   |$16   2    456 |
&#58;---------------+---------------+---------------&#58;
| 4    1    78  | 578  35   2   | 358  6    9   |
| 3    6    2   | 58   9    4   | 58   7    1   |
| 9    5    78  | 178  36   16  | 38   4    2   |
&#58;---------------+---------------+---------------&#58;
| 27   89   3   | 15   56   16  | 247 *89   47  |
| 167  47   1456| 9    2    8   | 167 *135 *356 |
| 1568 289  1569| 4    7    3   | 1269*189 *568 |
'---------------'---------------'---------------'
I love the logic behind this move :-k

There is an ALS in box 9, 5 starred cells with candidates digits 135689. Of these digits, 358 appear nowhere else in the box, and 169 only appear in the intersection with column 7, so this intersection must contain one of these digits 169.
Further in column 7, there is an ALS marked with $ for digits 169. This set requires 2 of the digits 169.
As a result, digits 169 cannot appear anywhere else in column 7, allowing us to eliminate candidates 6 & 9 from r1c7.

Any questions? 8-)
“If the human brain were so simple that we could understand it, we would be so simple that we couldn't.” - Emerson M Pugh
sudokuEd
Grandmaster
Grandmaster
Posts: 257
Joined: Mon Jun 19, 2006 11:06 am
Location: Sydney Australia

Post by sudokuEd »

emm on Oct 23 thread wrote:Ruud said that his XY ALS was more appropriate than the 2XY example in your thread. Maybe yours is the adult version. Very Happy
If mine was the adult version, Ruud has finally shown us the X-rated adult version. That is a really nice hidden triple, ALSxyz.

Now you've inspired me to find my own moves 8-)

1. 2 9's in n3
1a. 9 in r1c8 -> r2c1279 = {1238} -> no 8 in r2c8
1b. 9 in r2c7 -> r1c8 & r2c9 = {38} -> no 8 in r2c8.
1c. -> no 8 in r2c8.
1d. -> naked triple {159} r2 -> no 1 in r2c1

2. 2 8's in r2 -> a strong links X-wing thingo move eliminates 8 from r5c4
2a. 8 in r2c1 -> 8 in D\ must be in r4c4 or r9c9 and in D/ in r1c9 or r6c4. They can't both be in c9 -> an 8 must be in r4c4 or r6c4 -> no 8 in r5c4
2b. 8 in r2c9, exactly the mirror situation occurs for c1 -> 8 must be in r4c4 or r6c4 -> no 8 in r5c4
2c. -> no 8 in r5c4
2d. -> r5c7 = 8

then, its slow and steady progress to the end.

Thanks for a ripper Ruud. Been really enjoying these X-tremes. Seems to be one of these RA rated XT's a week :D
Last edited by sudokuEd on Wed Jan 03, 2007 10:35 pm, edited 2 times in total.
emm
Hooked
Hooked
Posts: 34
Joined: Wed Oct 11, 2006 6:01 pm

Post by emm »

V nice move Ruud - I'm finding the ALS really hard to find - apart from little XYwing ones. Are there any clues for spotting them?

PS: 3.1.07 XT = B (finally did it with UR). Any thoughts on how I can rate these?

PPS: I like my new shade of indigo!! :D
Para
Yokozuna
Yokozuna
Posts: 384
Joined: Wed Nov 08, 2006 7:42 pm
Location: The Netherlands

Post by Para »

sudokuEd wrote: If mine was the adult version, Ruud has finally shown us the X-rated adult version. That is a really nice hidden triple, ALSxyz.

Now you've inspired me to find my own moves 8-)

1. 2 9's in n3
1a. 9 in r1c8 -> r2c1279 = {1238} -> no 8 in r2c8
1b. 9 in r2c7 -> r1c8 & r2c9 = {38} -> no 8 in r2c8.
1c. -> no 8 in r2c8.
1d. -> naked triple {159} r2 -> no 1 in r2c1

2. 2 8's in r2 -> a strong links X-wing thingo move eliminates 8 from r5c4
2a. 8 in r2c1 -> 8 in D\ must be in r4c4 or r9c9 and in D/ in r1c9 or r6c4. They can't both be in c9 -> an 8 must be in r4c4 or r6c4-> no 8 in r5c4
2b. 8 in r2c9, exactly the mirror situation occurs for c1 -> 8 must be in r4c4 or r6c4-> no 8 in r5c4
2c. -> no 8 in r5c4
2d. -> r5c7 = 8

then, its slow and steady progress to the end.

Thanks for a ripper Ruud. Been really enjoying these X-tremes. Seems to be one of these RA rated XT's a week :D
Step one can be recreated by the XYZ-wing using R17C8 and R2C9.

Code: Select all

.---------------.---------------.---------------.
| 68   389  469 | 2    1    5   | 4679 389* 4678|
| 128  23   159 | 6    4    7   | 19   15-8 38* |
| 1567 47   146 | 3    8    9   | 16   2    456 |
&#58;---------------+---------------+---------------&#58;
| 4    1    78  | 578  35   2   | 358  6    9   |
| 3    6    2   | 58   9    4   | 58   7    1   |
| 9    5    78  | 178  36   16  | 38   4    2   |
&#58;---------------+---------------+---------------&#58;
| 27   89   3   | 15   56   16  | 247  89*  47  |
| 167  47   1456| 9    2    8   | 167  135  356 |
| 1568 289  1569| 4    7    3   | 1269 189  568 |
'---------------'---------------'---------------'
Para

ps
Nice elimination with the 8 in R5C4. After that it goes swiftly. Ruud's elimination of the 6 and 9 is a bit too big for me but it isn't needed. Eliminated them another way after a few more steps.
Para
Yokozuna
Yokozuna
Posts: 384
Joined: Wed Nov 08, 2006 7:42 pm
Location: The Netherlands

Post by Para »

sudokuEd wrote: If mine was the adult version, Ruud has finally shown us the X-rated adult version. That is a really nice hidden triple, ALSxyz.

Now you've inspired me to find my own moves 8-)

1. 2 9's in n3
1a. 9 in r1c8 -> r2c1279 = {1238} -> no 8 in r2c8
1b. 9 in r2c7 -> r1c8 & r2c9 = {38} -> no 8 in r2c8.
1c. -> no 8 in r2c8.
1d. -> naked triple {159} r2 -> no 1 in r2c1

2. 2 8's in r2 -> a strong links X-wing thingo move eliminates 8 from r5c4
2a. 8 in r2c1 -> 8 in D\ must be in r4c4 or r9c9 and in D/ in r1c9 or r6c4. They can't both be in c9 -> an 8 must be in r4c4 or r6c4-> no 8 in r5c4
2b. 8 in r2c9, exactly the mirror situation occurs for c1 -> 8 must be in r4c4 or r6c4-> no 8 in r5c4
2c. -> no 8 in r5c4
2d. -> r5c7 = 8

then, its slow and steady progress to the end.

Thanks for a ripper Ruud. Been really enjoying these X-tremes. Seems to be one of these RA rated XT's a week :D
I tried to solve it without this elimination and came up with another way to get past this point.
This is another way to get past step 2.
Coloring for digit 7 eliminates the 7 in R6C4. This give a 7 in R4C4 and R6C3 and an 8 in R4C3.

Code: Select all

.---------------.---------------.---------------.
| 68   389  469 | 2    1    5   | 4679 389  467A8|
| 128  23   159 | 6    4    7   | 19   15   38  |
| 1567 47   146 | 3    8    9   | 16   2    456 |
&#58;---------------+---------------+---------------&#58;
| 4    1    78  | 57B8 35   2   | 358  6    9   |
| 3    6    2   | 58   9    4   | 58   7    1   |
| 9    5    78  | 1-78 36   16  | 38   4    2   |
&#58;---------------+---------------+---------------&#58;
| 27   89   3   | 15   56   16  | 247  89   47B |
| 167  47   1456| 9    2    8   | 167  135  356 |
| 1568 289  1569| 4    7    3   | 1269 189  568 |
'---------------'---------------'---------------'
Then you get this. R1C1 and R9C9 are the only 8's on D\ and so there can't be any 8's in R1C9 and R9C1 which leaves a single 8 in R6C4 on the D/. Then you are able to fill boxes 5, 6 and 8.

Code: Select all

.---------------.---------------.---------------.
| 68*  389  469 | 2    1    5   | 4679 389  467-8|
| 128  23   159 | 6    4    7   | 19   15   38  |
| 1567 47   146 | 3    8    9   | 16   2    456 |
&#58;---------------+---------------+---------------&#58;
| 4    1    8   | 7    35   2   | 35   6    9   |
| 3    6    2   | 58   9    4   | 58   7    1   |
| 9    5    7   | 18!  36   16  | 38   4    2   |
&#58;---------------+---------------+---------------&#58;
| 27   89   3   | 15   56   16  | 24   89   47  |
| 167  47   1456| 9    2    8   | 167  135  356 |
| 156-8 289 1569| 4    7    3   | 1269 189  568*|
'---------------'---------------'---------------'
I think i am going to try some more X-files puzzles. This one kinda inspired me. :)

Para
Ruud
Site Owner
Site Owner
Posts: 601
Joined: Fri Dec 30, 2005 10:21 pm

Post by Ruud »

indigo girl, rating these puzzles is difficult because not all solving techniques recognize the diagonals. SudoCue often uses a more advanced technique like [[Multi-Colors]] when an easier pattern using the diagonals is present. When I solve these puzzles manually, I often find such shortcuts. Also, all the puzzles were rated using an earlier version of SudoCue which did not yet support the latest semi-advanced techniques.

In general: The diagonal constraints completely change the frequency in which the solving techniques occur. For instance, I found several [[Sue_de_Coq|Sue de Coq]] moves, sometimes with alternative ALC moves in Sudoku-X, which are very rare in regular Sudoku.

Anyhow, my X-Tremes collection is like a box of chocolates: you never know what you will get.

Ruud
“If the human brain were so simple that we could understand it, we would be so simple that we couldn't.” - Emerson M Pugh
sudokuEd
Grandmaster
Grandmaster
Posts: 257
Joined: Mon Jun 19, 2006 11:06 am
Location: Sydney Australia

Post by sudokuEd »

Para wrote:Step one can be recreated by the XYZ-wing using R17C8 and R2C9.
Thanks for pointing that one out Para. I havn't come across XYZ-wing - but it's not nearly as complicated as the name suggests. A simple pattern -> findable.
Para wrote:Ruud's elimination of the 6 and 9 is a bit too big for me
Too adult for me too at the moment - but its great fun to know about many different ways to solve the same puzzle. Very encouraging.
Para wrote:This is another way to get past step 2.
Coloring for digit 7 eliminates the 7 in R6C4.
Really like this move too - missed that. But again, very findable.
emm
Hooked
Hooked
Posts: 34
Joined: Wed Oct 11, 2006 6:01 pm

Post by emm »

Ruud - that's what I thought - but someone else I know rates them and I just wondered if I could too. No worries, not knowing is half the fun!

Here's my move to eliminate the 69 in N3. I wasn't going to show off but since Para did :D I'm going to - don't I just have to be like everyone else!!

Strong links on 7s N3 and N5 -> r7c7 <> 7
-> r4c4 = 7
-> hidden pr 47 on D/
-> naked pr 47 in c9
-> hidden pr 47 in N3
-> r1c7 <> 69 8-)

PS: 4.1.07 XT is a Caramello :D
Para
Yokozuna
Yokozuna
Posts: 384
Joined: Wed Nov 08, 2006 7:42 pm
Location: The Netherlands

Post by Para »

Ruud wrote:The intended deadly pattern has a single cell in each diagonal, which means that one of the diagonals has a 7 outside the pattern and the other has an 8. These force the pattern to a single solution.

In Sudoku-X, it is not possible to form URs with cells that belong to one of the diagonals, but they could be used in larger deadly patterns (BUG Lite), like this example:

Code: Select all

X . .|. X .|. . .
. . .|. . .|. . .
. . X|. X .|. . .
-----+-----+-----
. . .|. . .|. . .
X . X|. . .|. . .
. . .|. . .|. . .
-----+-----+-----
. . .|. . .|. . .
. . .|. . .|. . .
. . .|. . .|. . .
But these are very rare, I think.
I found one in december 16th X-treme

Code: Select all

. . .|. . 8|2 6 .
. . 5|9 . 2|4 . .
. 7 .|4 . .|. . .
-----+-----+-----
. 6 2|. 1 .|. . .
. . .|8 . .|. . .
4 5 .|. . .|. . .
-----+-----+-----
1 2 .|. . .|. . 6
7 . .|. . .|. . .
. . .|. . .|3 . .
Solved it till here and then one is there.

Code: Select all

.------------------.------------------.------------------.
| 39    349   349  | 1     5     8    | 2     6     7    |
| 6     1     5    | 9     7     2    | 4     8     3    |
| 2     7     8    | 4     3     6    | 1     59    59   |
&#58;------------------+------------------+------------------&#58;
| 8     6     2    | 35    1     34   | 579   34579 459  |
| 39    39    1379 | 8     6     3457 | 57    123-45* 12-45*|
| 4     5     137  | 2     9     37   | 6     13    8    |
&#58;------------------+------------------+------------------&#58;
| 1     2     349  | 35    8     359  | 579   579   6    |
| 7     349   34   | 6     24*   1359 | 8     24*   159  |
| 5     8     6    | 7     24*   19   | 3     19    24*  |
'------------------'------------------'------------------'
2 naked pairs 2-4 on R89C5 and R8C8/R9C9 and 2 locked in R5C89 eliminates the 4 from R5C89 because it would form a (non-)unique pattern otherwise.

This was just as example.

Para
Post Reply