New Killers in The Times

For fans of Killer Sudoku, Samurai Sudoku and other variants

New Killers in The Times

Postby PaulIQ164 » Fri Mar 24, 2006 12:28 am

I notice The Times have recently changed their Killer supplier from AIIA Corporation or whoever it was to Puzzler Media. These new ones are far to difficult for a bit of sudoku fun in the middle of the day for me. I don't suppose I'll get much sympathy around these parts, but does anyone else find the new ones markedly harder?
PaulIQ164
 
Posts: 533
Joined: 16 July 2005

Postby jimbob » Fri Mar 24, 2006 8:51 am

They are harder (particularly towards the end of the week). In my opinion this is an improvement - the old ones were far too easy.
jimbob
 
Posts: 47
Joined: 07 March 2006

Postby HATMAN » Fri Mar 24, 2006 10:10 am

They are certainly more reasonable - I no longer have to do them in KiMo mode to make them interesting.

Note however, that they do not even attempt to be symmetrical. We had a long discussion on DJApe’s forum on the need for killers to be symmetrical. The conclusion was that while it is perfectly acceptable for hand-crafted (difficult) puzzles to be asymmetric, a professional puzzle ought to maintain symmetry.
HATMAN
 
Posts: 203
Joined: 25 February 2006

Postby PaulIQ164 » Thu Apr 13, 2006 6:15 pm

If anyone has any tips on how to proceed in today's Times Killer, beyond the 4 in r4c9, and the 9,1 and 4 in r6c3, r7c2 and r7c3 respectively, they'd be gratefully recieved.
PaulIQ164
 
Posts: 533
Joined: 16 July 2005

Postby Ruud » Thu Apr 13, 2006 7:14 pm

Hi Paul,

Can you post the puzzle here, or at least a link to it? I can't seem to find it anywhere.

TIA,

Ruud.
Ruud
 
Posts: 664
Joined: 28 October 2005

Postby Jean-Christophe » Thu Apr 13, 2006 7:31 pm

Will have to wait tomorow when The Times will post it on their site.
I'll have a look:D
Jean-Christophe
 
Posts: 149
Joined: 22 January 2006

Postby CathyW » Thu Apr 13, 2006 8:04 pm

I'm glad it's not just me! Definitely a tough one.

Image
Picture isn't too clear here but if you click on the link it's better and hopefully you can read the numbers.

Having recreated in DJApe's Perfect Sudoku and checked it out, it does have a unique solution but I can't see how to progress it.

The text file for anyone else using PS is (you'll have to remove the line breaks:
3x3::k:4097:4097:7426:7426:7426:7426:3846:4871:4871:
4097:2314:2314:3853:3853:7426:3846:4871:4871:
5140:5140:5140:3853:3853:5399:3846:3865:3865:
5140:4892:4134:4134:4134:5399:5399:5399:3865:
4892:4892:4134:5159:3377:5417:5417:5417:5417:
2093:2093:3384:5159:3377:2364:2364:5694:5417:
5942:2093:3384:5159:6467:6467:2364:5694:5694:
5942:4416:4416:5159:6467:6467:2127:2127:2640:
5942:4416:4416:2635:2635:3149:3149:2127:2640:
CathyW
 
Posts: 316
Joined: 20 June 2005

Postby Bigtone53 » Thu Apr 13, 2006 8:16 pm

If anyone has any tips on how to proceed in today's Times Killer, beyond the 4 in r4c9, and the 9,1 and 4 in r6c3, r7c2 and r7c3 respectively, they'd be gratefully recieved.


This was a hard one.

Consider where the 6 is in (3,3)

Then investigate the possibilities for the 25 shape in (3,2)

A worthy puzzle
Bigtone53
 
Posts: 413
Joined: 19 September 2005

Postby CathyW » Thu Apr 13, 2006 8:33 pm

Thanks - should have spotted the locked 6s in row 7 box 9! 9 is also locked to r7c89 so r7c1 must be 8.

Still seems to be a lot of possibilities for the 25(4) in box 8 - I assume that's what you mean by 3,2 - so I'm still stuck! It can't be {1789} but how did you narrow down the remaining possible combinations?
Last edited by CathyW on Thu Apr 13, 2006 4:55 pm, edited 1 time in total.
CathyW
 
Posts: 316
Joined: 20 June 2005

Postby Bigtone53 » Thu Apr 13, 2006 8:46 pm

how did you narrow down the remaining possible combinations?


In row 7, there are only 4 possibilities for the top line in box 8. This restricts the possibilities for the 25 box there as two of them have to be part of a 25 total. Working with this and the possibilities for the 10 thing in box 8 clarifies which way around the two candidates for the 12 in row 9 go, It all falls into place after that,

Sorry about being enigmatic but I am working from memory, having thrown the whole thing in the bin when I finally got there after way more than the allotted time.
Bigtone53
 
Posts: 413
Joined: 19 September 2005

Postby CathyW » Thu Apr 13, 2006 9:12 pm

[quote="Bigtone53] It all falls into place after that.[/quote]
So it does:D Many thanks. Worthy indeed - Perfect Sudoku (v4) rates it as Insane, the hardest level!
CathyW
 
Posts: 316
Joined: 20 June 2005

Postby Jean-Christophe » Fri Apr 14, 2006 3:14 pm

I agree, pretty hard one. Usually Times killer are quite easy, but they raised the level with this very interesting one.

Here is my walkthrough. Probably not the shortest way to go.

PS I'm using the killer lingo we use on other sites too. See :
http://sudoku.apinc.org/?page_id=3

< SPOILER >

Step 1
Outies of N3 -> R4C9 = 4, R3C89 = 11
Innies of N7 -> R7C23 = 5
R7C3 belongs to cage 13/2 -> >= 4
-> R7C23 = [14], R6C3 = 9
-> R6C12 = 7 = {25|34} (can't be {16} because numbers can't be repeated in cages)

Step 2
Cage 23/3 in N7 = {689} (naked triplet within C1, N7)
-> Cage 17/4 in N7 = {2357}

Step 3
45 on N14 -> R145C3 = 10/3 -> they're can't be a 8 in these
-> 8 of C3 in R23C3 -> 8 of N4 in R45C2
-> Cage 19/3 in N4 must have a 8 -> {478|568} = {8(4|5)...}
The notation {8(4|5)...} means must have a 8 and also either 4 or 5 then something I don't mind.
At step 1 we found that R6C12 = 7 = {25|34} = {(4|5)...}
So it also must have either 4 or 5.
This is a form of naked pair specific to killer sudoku.
Each of the 2 cages must have either 4 or 5
Althought we don't know yet which is which, no other cell in N4 may have {45}

Step 4
45 on N4 -> R4C1+R45C3 = 10/3
Since it can't have {45} -> {127|136}
-> R4C1 = {1237}

Step 5
45 on N1 -> R1C3 = R4C1 -> R1C3 = {1237} (ie it can't be 8)
R2C3 can't be 8 because R2C2 can't be 1
-> R3C3 = 8
-> R3C89 = {29|56}

Step 6
45 on N2 -> R3C6 = R1C3 + 1 -> R3C6 = {234} (can't be 8) -> R1C3 = R4C1 = {123}
45 on R1..4 -> R5C123 = 14/3
45 on N4 -> R4C123 = 15/3

Step 7
Cage 19/3 in N4 = {478|568}
R4C2 can't be 5 because R5C1 can't have 6 nor 8.
-> R4C2 = {678}
R4C123 = 15/3. Many choices are not compatibles the various constraints.
-> R4C123 = {168|267} = {6(18|27)}
-> 6 in R4C123, nowhere else in R4, N4
-> R4C1 = {12} -> R1C3 = {12}, R3C6 = {23}
R5C123 = 14/3. Many choices are not compatibles...
We are left with {158|248|257|347}
{248} can't work with R6C12 (which must have either 2 or 4)
{347} dito (R6C12 must have either 3 or 4)
-> R5C123 = {158|257} = {5(18|27)} (must have a 5)
-> R5C1 = 5, R6C12 = {34}
-> Cage 19/4 in N4 = {568} -> R45C2 = [68]
-> R4C123 = {267} -> R4C13 = [27], R5C3 = 1
-> R1C3 = 2 -> R3C6 = 3

From here on it's much more easy
...
< /SPOILER >
Jean-Christophe
 
Posts: 149
Joined: 22 January 2006

Postby HATMAN » Fri Apr 14, 2006 9:30 pm

That one was difficult - very hard to get started. 45mins? - nearer to two hours for me.

My approach was similar to JC in breaking into N4. I also found the "45" on N8 giving a 22/3 outie useful.
HATMAN
 
Posts: 203
Joined: 25 February 2006

Postby Crazy Girl » Sat Apr 15, 2006 1:38 pm

the way i cracked it was to look at N9 and the 10[2] cage / 8[3] cage.

the 8[3] had to contain 4 so 8={134} which made 10[2]= {28}.

using the fact that the outties of N8=22 and the 22[3] cage in N69 contains 9 one can deduce that the 12[2] cage in R9 is {57}.

Then went on to look at the Innes of N8, 19[3] in N4 and innes / outties of N1 and N2. and deduce the 9[2] in N1

The puzzle definelty required a computer to track all the candidates for each cell, so the 'tough' rating should of been upgraded in my opinion, as previous tough rated puzzles didn't require computer use to solve them.:D
Crazy Girl
 
Posts: 189
Joined: 08 November 2005

Postby PaulIQ164 » Sat Apr 15, 2006 10:37 pm

Good work people. If I'm remembering correctly, we cracked this one after I posted (more specifically, my Dad did it). The Deadly the next day was messed up. I'd say that if a puzzle needs a computer to solve, it's too difficult to go in a newspaper (though they'd be a cool extra to put on the website).
PaulIQ164
 
Posts: 533
Joined: 16 July 2005

Next

Return to Sudoku variants