A Kanji/Hanji/Chinese Sudoku puzzle

For fans of Killer Sudoku, Samurai Sudoku and other variants

A Kanji/Hanji/Chinese Sudoku puzzle

Postby udosuk » Mon Jul 17, 2006 1:45 pm

This puzzle is adopted from one of Pyrrhon's failed attempts...

Image

Could be of interest to anyone who likes Hanji/Chinese characters...

PS: It could assist a little if you know the meaning of the characters...

Here is the list of the characters in Chinese font (if your computer is compatible with the language):

前列在陣者鬥兵皆臨 (not in order)...
Last edited by udosuk on Sun Jul 23, 2006 3:37 am, edited 1 time in total.
udosuk
 
Posts: 2698
Joined: 17 July 2005

Postby nj3h » Tue Jul 18, 2006 3:40 pm

Wouldn't the puzzle with the pencil marks entered be really festive?
nj3h
 
Posts: 47
Joined: 07 July 2005

Postby tso » Tue Jul 18, 2006 5:31 pm

Translation of single characters out of context can be iffy.

Chinese to English translation from Babelfish:

前 -- Front
列 -- Row
在 -- In
陣 -- ??
者 -- ??
鬥 -- Fights
兵 -- Soldier
皆 -- All
臨 -- Near

Japanese to English translation from Babelfish:

前 -- Before
列 -- Line
在 -- Resident in
陣 -- Position
者 -- Person
鬥 -- ??
兵 -- Soldier
皆 -- Everyone
臨 -- Overlooking
tso
 
Posts: 798
Joined: 22 June 2005

Postby udosuk » Wed Jul 19, 2006 2:44 am

前 -- Front
列 -- Arrangement / List / Line / Row
在 -- In
陣 -- Array
者 -- Person
鬥 -- Fighting
兵 -- Soldiers
皆 -- All / Everyone
臨 -- Approaching / Arriving / Near

If you understand Chinese/Japanese, you should notice the middle column (c5) contains a famous sentence reading from top to bottom... Knowing this sentence, the puzzle can be solved with singles only, without the need of candidate marking.

If you don't understand the language, you'd have to replace the symbols by digits or positioned dots and the puzzle would be a bit more difficult...
udosuk
 
Posts: 2698
Joined: 17 July 2005

Postby udosuk » Thu Jul 20, 2006 3:38 am

I should have said "Kanji" (the Japanese term for "Chinese characters") instead of "Hanji" (the Korean term)... In Chinese it's spelt "Hanzi" (漢字)...

It's all very confusing... (Confuciusing?):)
udosuk
 
Posts: 2698
Joined: 17 July 2005

Postby Pyrrhon » Fri Jul 28, 2006 7:21 pm

I guess I haven't solved the translation act. How works the shorter way?

Pyrrhon
Pyrrhon
 
Posts: 240
Joined: 26 April 2006

Postby udosuk » Fri Jul 28, 2006 9:12 pm

Okay, the 9 characters, when arranged in a particular order, will form a famous Buddhist/Zen/Ninja incantation phrase... Putting this phrase on the middle column (c5) from top to bottom will reduce the puzzle to singles only...

There is info in "another place" about where to find this phrase...:D
udosuk
 
Posts: 2698
Joined: 17 July 2005

Postby udosuk » Sat Jul 29, 2006 7:36 am

This is the puzzle in numerical format:
Code: Select all
.........
1.47.83.5
.2..3..9.
7.83.95.6
.........
24.86.97.
.62.73.49
..3..4..2
.........

After basic moves (singles/pairs/locked candidates) you need a complex turbot fish/multiple colouring, a naked quad/hidden pair, an x-wing, and finally an xy-wing to solve it... But of course all these are not necessary if you know the middle column from the characters...
udosuk
 
Posts: 2698
Joined: 17 July 2005

Postby Pyrrhon » Sat Jul 29, 2006 10:33 am

You can solve it with naked and hidden singles, one box-line interaction, two naked paira and one xyz-wing (the almost naked pair is in B8 (R9C5, R7C4) and the helping cell is R1C5).

All the other techniques you used are not necessary.

Pyrrhon
Pyrrhon
 
Posts: 240
Joined: 26 April 2006

Postby udosuk » Sat Jul 29, 2006 10:53 am

Pyrrhon wrote:You can solve it with naked and hidden singles, one box-line interaction, two naked paira and one xyz-wing (the almost naked pair is in B8 (R9C5, R7C4) and the helping cell is R1C5).

All the other techniques you used are not necessary.

Pyrrhon

I couldn't see the xyz-wing...:(
Code: Select all
 *-----------------------------------------------------------*
 | 3     58    67    | 149  *19    56    | 2     18    1478  |
 | 1     9     4     | 7     2     8     | 3     6     5     |
 | 58    2     67    | 14    3     56    | 478   9     1478  |
 |-------------------+-------------------+-------------------|
 | 7     1     8     | 3     4     9     | 5     2     6     |
 | 6     3     9     | 2     5     7     | 48    18    148   |
 | 2     4     5     | 8     6     1     | 9     7     3     |
 |-------------------+-------------------+-------------------|
 | 58    6     2     |*15    7     3     | 18    4     9     |
 | 9     78    3     | 16    18    4     | 1678  5     2     |
 | 4     578   1     | 569  *89    2     | 678   3     78    |
 *-----------------------------------------------------------*

Could you explain a little please?:?: Thanks!
udosuk
 
Posts: 2698
Joined: 17 July 2005

Postby Pyrrhon » Sat Jul 29, 2006 12:52 pm

You are right and we can take the turbot chain with strong links R7C1-R7C7 in R7, R3C1-R1C2 in B1, and R1C8-R5C8 in C8 0> R5C7 <> 8 and a x-wing of 8 with strong links in R3 R7 and weak links in C1 C7. Another way is the simple forcing loop

R1C5 -> R1C8 -> R1C2 -> R9C2 -> R9C5 -> R1C5

It follows that:

R1C45 <> 1 because of R1C58
R1C9 <> 8 because of R1C82
R9C79 <> 8 because of R9C25
Pyrrhon
 
Posts: 240
Joined: 26 April 2006

Postby udosuk » Sat Jul 29, 2006 1:35 pm

Awww...:( My head is blowing up... I can't follow a bit of that "simple forcing loop"...:!: But one thing is certain, in the solution r1c5=1 so your assertion that "r1c45<>1" is clearly false...

Are you deliberately writing false things to mess with my brain?:D
udosuk
 
Posts: 2698
Joined: 17 July 2005

Postby Pyrrhon » Sat Jul 29, 2006 1:48 pm

I need better glasses (its to hot in Germany): From the forcing loop follows that R1C5=1 or R1C8 = 1 and so R1C4 <> 1 and R1C9 <> 1.
Pyrrhon
 
Posts: 240
Joined: 26 April 2006

Postby udosuk » Sun Jul 30, 2006 5:12 am

Let's recap... This is the state you can achieve with basic techniques:
Code: Select all
 3     58    67    | 149   19    56    | 2     18    1478
 1     9     4     | 7     2     8     | 3     6     5   
 58    2     67    | 14    3     56    | 478   9     1478
-------------------+-------------------+------------------
 7     1     8     | 3     4     9     | 5     2     6   
 6     3     9     | 2     5     7     | 48    18    148 
 2     4     5     | 8     6     1     | 9     7     3   
-------------------+-------------------+------------------
*58    6     2     |*15    7     3     | 18    4     9   
 9    -78    3     | 16   *18    4     | 1678  5     2   
 4     578   1     | 569   89    2     | 678   3     78   

There is a simple xy-wing in r7c14 and r8c25 => r8c2<>8, therefore r8c2=7...

Pyrrhon's "simple forcing loop" works like this:
Code: Select all
 3    *58    67    | 149  *19    56    | 2    *18    1478
 1     9     4     | 7     2     8     | 3     6     5   
 58    2     67    | 14    3     56    | 478   9     1478
-------------------+-------------------+------------------
 7     1     8     | 3     4     9     | 5     2     6   
 6     3     9     | 2     5     7     | 48    18    148 
 2     4     5     | 8     6     1     | 9     7     3   
-------------------+-------------------+------------------
 58    6     2     | 15    7     3     | 18    4     9   
 9     7     3     | 16    18    4     | 168   5     2   
 4    *58    1     | 569  *89    2     | 678   3     78   

[r1c5]-1-[r1c8]-8-[r1c2]=58=[r9c2]-8-[r9c5]=9=[r1c5]

It follows that:
r1c49=1|8 (together with r1c8) will force r1c5=9 & r1c2=5 => r9c2=8 => r9c5=9, so r1c4<>1, r1c9<>1|8
r9c79=8 will force r9c5=9 & r9c2=5 => r1c2=8 => r1c8=1 => r1c5=9, so r9c79<>8

Alternatively, you can work on the following turbot chain:
Code: Select all
 3    *58    67    | 149   19    56    | 2    *18    1478
 1     9     4     | 7     2     8     | 3     6     5   
*58    2     67    | 14    3     56    | 478   9     1478
-------------------+-------------------+------------------
 7     1     8     | 3     4     9     | 5     2     6   
 6     3     9     | 2     5     7     |-48   *18    148 
 2     4     5     | 8     6     1     | 9     7     3   
-------------------+-------------------+------------------
*58    6     2     | 15    7     3     |*18    4     9   
 9     7     3     | 16    18    4     | 168   5     2   
 4     58    1     | 569   89    2     | 678   3     78   

[r5c7]-8-[r7c7]=8=[r7c1]=8=[r3c1]=8=[r1c2]-8-[r1c8]=8=[r5c8]-8-[r5c7]

It follows directly that r5c7<>8 and thus r5c7=4... And then a naked quad on r3 and an x-wing on 8 solves it...
udosuk
 
Posts: 2698
Joined: 17 July 2005


Return to Sudoku variants