Times Super fiendish 1357

Post the puzzle or solving technique that's causing you trouble and someone will help

Times Super fiendish 1357

Postby richardm » Thu Jun 07, 2007 10:01 pm

I'd be interested in knowing how others solve this one. I used two advanced techniques after whicih the puzzle fell out very quickly, But it took me a while to spot these techniques. So rather than say what they were let me quote the puzzle:
Code: Select all
8.5 .4. ..1
.7. 1.. .6.
... ..2 ..3

..1 .8. .4.
4.. 7.9 ..6
.3. .6. 7..

1.. 8.. ...
.4. ..7 .9.
3.. .2. 5.4


cheers, richardm
richardm
 
Posts: 53
Joined: 27 December 2006

Re: Times Super fiendish 1357

Postby re'born » Thu Jun 07, 2007 11:40 pm

richardm wrote:I'd be interested in knowing how others solve this one. I used two advanced techniques after whicih the puzzle fell out very quickly, But it took me a while to spot these techniques. So rather than say what they were let me quote the puzzle:
Code: Select all
8.5 .4. ..1
.7. 1.. .6.
... ..2 ..3

..1 .8. .4.
4.. 7.9 ..6
.3. .6. 7..

1.. 8.. ...
.4. ..7 .9.
3.. .2. 5.4


cheers, richardm



The basics take you here:
Code: Select all
 *--------------------------------------------------------------------*
 | 8      29     5      | 36     4      36     | 29     7      1      |
 | 29     7      3      | 1      59%    8      | 4      6      259-   |
 | 6      1      4      | 59     7      2      | 89     58     3      |
 |----------------------+----------------------+----------------------|
 | 7      6      1      | 2      8      35     | 39     4      59%    |
 | 4      258*   28     | 7      13     9      | 1238   12358* 6      |
 | 259    3      289    | 4      6      15     | 7      1258   258    |
 |----------------------+----------------------+----------------------|
 | 1      259*   269    | 8      359*   4      | 236    23     7      |
 | 25     4      268    | 35     135    7      | 12368  9      28     |
 | 3      89     7      | 69     2      16     | 5      18     4      |
 *--------------------------------------------------------------------*


where an extended version of the Y-wing trick implies r2c9<>9, solving the puzzle.

9-[r4c9]-5-[r5c8]=5=[r5c2]-5-[r7c2]=5=[r7c5]-5-[r2c5]-9

(of course, my first time through it was an x-wing and a swordfish, but I like one move solutions)

Edit: Bah, my swordfish on 9 will solve the puzzle by itself because it kills r2c9=9 as well. Forget the Y-wing extension.
re'born
 
Posts: 551
Joined: 31 May 2007

Postby udosuk » Fri Jun 08, 2007 5:44 am

In case you're wondering what re'born meant, after 15 hidden singles:
Code: Select all
 *--------------------------------------------------------------------*
 | 8      29     5      | 369    4      36     | 29     7      1      |
 |*29     7      3      | 1     *59     8      | 4      6     -259    |
 | 6      1      4      | 59     7      2      | 89     58     3      |
 |----------------------+----------------------+----------------------|
 | 7      6      1      | 2      8      35     | 39     4      59     |
 | 4      258    28     | 7      135    9      | 1238   12358  6      |
 |*259    3     *289    | 4      6      15     | 7      1258  -2589   |
 |----------------------+----------------------+----------------------|
 | 1      259   *269    | 8     *359    4      | 236    23     7      |
 | 25     4      268    | 356    135    7      | 12368  9      28     |
 | 3      89     7      | 69     2      16     | 5      18     4      |
 *--------------------------------------------------------------------*

Then a swordfish on 9 @ r267c135 => r26c9<>9 => r4c9=9

The rest are all naked singles.

I think this is one of the "textbook solutions" to this puzzle.:idea:

(The other one is the complimentary swordfish on 9 @ r139c247, but it doesn't eliminate the 9 in r2c9 directly.)
udosuk
 
Posts: 2698
Joined: 17 July 2005

Postby richardm » Fri Jun 08, 2007 4:45 pm

Thanks for that, I'll study your solution in more detail.

What I did was used a jellyfish on 5s and swordfish on 9s - if these are the right terms. I.e. I found 4 rows with 5 in <= 4 common places and 3 cols with 9 in <= 3 common places.
Is this the same as your y-wing?

Richard
Last edited by richardm on Mon Jun 11, 2007 6:33 pm, edited 1 time in total.
richardm
 
Posts: 53
Joined: 27 December 2006

Postby re'born » Fri Jun 08, 2007 5:05 pm

richardm wrote:Thanks for that, I'll study your solution in more detail.

What I did was used a jellyfish on 4s and swordfish on 9s - if these are the right terms. I.e. I found 4 rows with 4 in <= 4 common places and 3 cols with 9 in <= 3 common places.
Is this the same as your y-wing?

Richard


If you use forget about the jellyfish, your swordfish still solves the puzzle. But neither the jellyfish or the swordfish have anything to do with the Y-wing (except for common exclusion of 9 from r2c9). The Y-wing takes two bivalued cells with the same values (say a and b) and then connects them with alternating weak and strong links on one of the digits, say a. The deduction then is that you can eliminate b from any cell that sees the two bivalued endpoint cells. In my case the endpoints are r2c5<59> and r4c9<59> with a = 5, b = 9. Look at the *'d cells in my grid above to see my path connecting the endpoints. As r2c9 sees both of the endpoint cells, it cannot contain b = 9.
re'born
 
Posts: 551
Joined: 31 May 2007

Postby richardm » Mon Jun 11, 2007 10:52 pm

Ok, I understand your y-wing - a weak form of remote pairs.
My method, which I quoted incorrectly above, used the complementary swordfish on r139 which udosuk quoted. Having decided to work with that, I then happend to spot the 4-fish in the 5 right away. Not the most elegant of solutions.

Richardm
richardm
 
Posts: 53
Joined: 27 December 2006

re: The Times #1357

Postby Pat » Tue Jun 12, 2007 11:16 am

The Times (2007.Jun.1) wrote:
      #1357 - Super Fiendish

      [ 28 clues ]
Code: Select all
 8 . 5 | . 4 . | . . 1
 . 7 . | 1 . . | . 6 .
 . . . | . . 2 | . . 3
-------+-------+------
 . . 1 | . 8 . | . 4 .
 4 . . | 7 . 9 | . . 6
 . 3 . | . 6 . | 7 . .
-------+-------+------
 1 . . | 8 . . | . . .
 . 4 . | . . 7 | . 9 .
 3 . . | . 2 . | 5 . 4


udosuk wrote:after 15 hidden singles:

Code: Select all
 8 . 5 | . 4 . | . 7 1
 . 7 3 | 1 . 8 | 4 6 .
 6 1 4 | . 7 2 | . . 3
-------+-------+------
 7 6 1 | 2 8 . | . 4 .
 4 . . | 7 . 9 | . . 6
 . 3 . | 4 6 . | 7 . .
-------+-------+------
 1 . . | 8 . 4 | . . 7
 . 4 . | . . 7 | . 9 .
 3 . 7 | . 2 . | 5 . 4


Then a swordfish on 9 c135\r267 => r26c9<>9 => hidden single 9 in c9

The rest are all hidden singles.

this is one of the "textbook solutions" to this puzzle.

The other one is the complementary swordfish on 9 r139\c247, but it doesn't eliminate the 9 in r2c9 directly.

hey udosuk -- the 2nd Swordfish immediately gives a "hidden single" 9 in r4
    trying to classify this situation:

    the digit (9) is missing 7 times,
    so we expect a Swordfish to be complemented by a Jellyfish,
    but here the Jellyfish degenerates giving an immediate "hidden single"
    -- this makes the puzzle somewhat easier,
    i have 2 ways to catch a Swordfish, either by rows or by columns

    7 = 3 + 1 + 3
User avatar
Pat
 
Posts: 4056
Joined: 18 July 2005

Re: re: The Times #1357

Postby ronk » Tue Jun 12, 2007 11:45 am

Pat wrote:but here the Jellyfish degenerates giving an immediate "hidden single"

An immediate assignment by way of a degenerate fish ... is certainly valid and is also a very interesting POV.
ronk
2012 Supporter
 
Posts: 4764
Joined: 02 November 2005
Location: Southeastern USA

Re: re: The Times #1357

Postby udosuk » Tue Jun 12, 2007 1:28 pm

Pat wrote:hey udosuk -- the 2nd Swordfish immediately gives a "hidden single" 9 in r4...

Yes, I mentioned that they're both "textbook solutions".

The 1st swordfish (c135/r267) solves the puzzle by leaving a hidden 9 on c9 (r4c9). I used it there because it immediately gives the elimination r2c9<>9 as mentioned by re'born.

The 2nd swordfish (r139/c247) solves the puzzle by leaving a hidden 9 on r4 (r4c9) (which will indirectly eliminate 9 from r2c9). It also leaves a naked single r4c7=3, so in a sense it's a quicker solution. However, it doesn't immediately give the elimination r2c9<>9. So I didn't use it to demonstrate what re'born meant when he said "my swordfish on 9 will solve the puzzle by itself because it kills r2c9=9 as well".
udosuk
 
Posts: 2698
Joined: 17 July 2005

Re: re: The Times #1357

Postby ronk » Tue Jun 12, 2007 2:13 pm

udosuk wrote:
Pat wrote:hey udosuk -- the 2nd Swordfish immediately gives a "hidden single" 9 in r4...

Yes, I mentioned that they're both "textbook solutions".

I think Pat quoted hidden single because she was looking at a jellyfish.
Code: Select all
 . *9  . |  .  .  . | *9  .  .
 9  .  . |  .  9  . |  .  .  9
 .  .  . | *9  .  . | *9  .  .
---------+----------+----------
 .  .  . |  .  .  . | *9  .+*9
 .  .  . |  .  .  . |  .  .  .
 9  .  9 |  .  .  . |  .  .  .
---------+----------+----------
 . -9  9 |  .  9  . |  .  .  .
 .  .  . |  .  .  . |  .  .  .
 . *9  . | *9  .  . |  .  .  .

 sashimi jellyfish r1349\c2479, implies r4c9=9 and r7c2<>9

The only candidate of the jellyfish in c9 is in r4, so it may be placed immediately ... without an intermediate elimination.

It's a valid POV which IMO has little chance of catching on, but it is interesting.
ronk
2012 Supporter
 
Posts: 4764
Joined: 02 November 2005
Location: Southeastern USA

Postby daj95376 » Tue Jun 12, 2007 3:39 pm

Wow, talk about covering a puzzle from all angles. I don't have anything new to add, just a personal preference after reading the above.

The Swordfish c135\r267 will lead to a hidden <9> in [r4c9], and this will lead to a cascade of naked singles starting with [r4c7]=3.

The Swordfish r139\c247 will directly lead to [r4c7]=3 and the cascade of singles.

The degenerate Jellyfish assignment of [r4c9]=9 is just a hidden single that follows from either Swordfish. I don't consider it to be a viable argument of its own.
daj95376
2014 Supporter
 
Posts: 2624
Joined: 15 May 2006

Postby udosuk » Wed Jun 13, 2007 3:23 am

daj95376 wrote:The degenerate Jellyfish assignment of [r4c9]=9 is just a hidden single that follows from either Swordfish. I don't consider it to be a viable argument of its own.

Agree with you here.:)
udosuk
 
Posts: 2698
Joined: 17 July 2005

re(2): The Times #1357

Postby Pat » Wed Jun 13, 2007 9:21 am

udosuk wrote:The 2nd swordfish (r139/c247) -- However, it doesn't immediately give the elimination r2c9<>9. So I didn't use it to demonstrate what re'born meant when he said "my swordfish on 9 will solve the puzzle by itself because it kills r2c9=9 as well".

thanks, udosuk, you are so very patient (as always)!

i had mis-interpreted your post --
my interest was only in the Swordfish
(and not in explaining re'born's post).
    by the way, i did it on paper, the way i like, not writing lists of possibilities in the cells,
    hence my great preference for "hidden" stuff.
    i saw the 2 Swordfish,
    each of them sufficient for cracking the puzzle by giving the next "hidden single".
User avatar
Pat
 
Posts: 4056
Joined: 18 July 2005


Return to Help with puzzles and solving techniques