SudokuSolver Forum

A forum for Sudoku enthusiasts to share puzzles, techniques and software
It is currently Sun Apr 28, 2024 1:25 pm

All times are UTC




Post new topic Reply to topic  [ 4 posts ] 
Author Message
PostPosted: Thu Jul 05, 2012 5:30 pm 
Offline
Grand Master
Grand Master

Joined: Wed Apr 30, 2008 9:45 pm
Posts: 694
Location: Saudi Arabia
Human Solvable 11 Diamond Killer 2

OK I know no-one has done any software to solve Diamond Killers, but if/when they do they will have trouble with this one.

1/2 cells contribute half to the cage total.

No number may repeat in a cage - even when halves. This is slightly more information than standard killers. The largest cage of pure halves is 24(8/2).

I challenge you to solve it without the HS bit - I have not.


Image


Uploaded with ImageShack.us


Top
 Profile  
Reply with quote  
PostPosted: Mon Jul 09, 2012 10:10 pm 
Offline
Grand Master
Grand Master

Joined: Wed Apr 23, 2008 6:04 pm
Posts: 1895
Location: Lethbridge, Alberta, Canada
I've just come across this new puzzle by using our daughter's laptop while we are on holiday.

It looks an interesting puzzle. I won't make any serious attempt to solve it until we get home in a week and a half.

One question now. Am I right in thinking that the half contributions to cage totals can be n.5 as well as n.0? From a quick glance I think that some half contributions will have to be n.5, unless I'm not thinking carefully enough.

Andrew


Top
 Profile  
Reply with quote  
PostPosted: Thu Jul 12, 2012 12:26 am 
Offline
Grand Master
Grand Master

Joined: Tue Jun 16, 2009 9:31 pm
Posts: 282
Location: California, out of London
Absolutely Andrew! In fact, since the corner nonets have 7 half-cells, some of the half-cells must be odd numbers. However, so long as there are an even number of odd-number half-cells in a cage, the total will still be a whole number.

Thanks HATMAN for another fine puzzle. I did find this easier than your previous diagonal killer. I'm not sure what the "Human-Solvable" element is to which you refer. Some possibilities that the design imposes are:

Hidden Text:
There is a very restricted set of cells in each corner nonet to which the cells sticking out from the nonet can go. E.g., Whatever is in r9c6 can only go in r7c7 or r8c9 in n9.

The two whole cells in each corner nonet have a fixed difference. So if you know one you also know the other.


Top
 Profile  
Reply with quote  
PostPosted: Wed Sep 26, 2012 12:51 am 
Offline
Grand Master
Grand Master

Joined: Wed Apr 23, 2008 6:04 pm
Posts: 1895
Location: Lethbridge, Alberta, Canada
Thanks HATMAN for another enjoyable puzzle. Having set up the diagram back in July, while we were driving across Canada from Lethbridge, Alberta to St. John's, Newfoundland, I only came back to this puzzle last week.

Thanks wellbeback for the interesting observations in your post. The first one seems to be essential to solving this puzzle fairly directly.

Here is my walkthrough for Human Solvable 11 Diamond Killer 2:
If I have made any errors, including typos, or anything needs clarification, please send me a PM.

This puzzle has the condition that half cells cannot contribute the same amount to a cage as a full cell.

1. 4(3) cage at R4C2 can only be 2 + 0.5 + 1.5 (cannot be 1 + 0.5 + 2.5 because 1 would be repeated in the cage) -> R4C3 = 2, R4C2 + R5C3 = {13}

Each of the four corner nonets has two half outies. Each of these outie cells “sees” all except two cells inside that nonet. Thus R1C4 “sees” all cells in N1 except for R2C1 and R3C3 and its complementary outie R4C1 “sees” all cells in N1 except for R1C2 and R3C3.

Each of the three-cell cages contains two half-cells and one full cell. Therefore to make an integer total the two half-cells must contain two even numbers or two odd numbers.

2. 45 rule on N7 2 half-outies R6C1 + R9C4 = 7 -> R6C1 + R9C4 = 14 = {59/68}
2a. R6C1 + R9C4 “see” all cells in N7 except for R7C3, R8C1 (for R9C4) and R9C2 (for R6C1)
2b. R8C1/R9C2 cannot be 8,9 to make 5(3) cage in R8C1
2c. 5(3) cage in R8C1 can only be 2.5 + 0.5 + 2 or 2.5 + 1 + 1.5 (cannot be 6 in R8C1/R9C2 contributing 3 because the remaining cells would each contribute 1, which is specifically forbidden) -> R8C1/R9C2 = 5, R9C1 = {12} and the other of R8C1/R9C2 = {13}, 5(3) cage contains both of 1,5 and one of 2,3
2d. R8C1/R9C2 = 5 -> R6C1 + R9C4 = {59} -> R7C3 = 9 (because 9 cannot be in R8C1/R9C2)
2e. 5 must be in R68C1, locked for C1
2f. 5 must be in R9C24, locked for C9

3. R7C3 = 9 -> half-cells R7C2 + R8C3 = 5
3a. For 21(7) cage at R6C1, half-cells R6C1 + R9C4 = 7, half-cells R7C2 + R8C3 = 5 -> remaining half-cells R7C1 + R8C2 + R9C3 = 9
3b. For 12(5) cage at R7C1, half-cells R7C1 + R8C2 + R9C3 = 9 -> remaining half-cells R8C1 + R9C2 = 3 -> R9C1 = 2, R8C1 + R9C2 = 6 = {15}

4. For 7(3) cage at R9C4, R9C4 is odd -> R9C6 must be odd
45 rule on N9 2 half-outies R6C9 + R9C6 = 5 -> R6C9 + R9C4 = 10 = {19/37}
R9C4 = {59} -> cage contribution = 2.5 or 4.5 -> cage = 4.5 + 1 + 1.5 (cannot be 2.5 + 1 + 3.5 because R9C45 = [51] clashes with R9C2, cannot be 2.5 + 3 + 1.5 because 3 would be repeated in the cage, cannot be {2.5 + 4 + 0.5 because R9C46 = [51] clashes with R9C2) -> R9C456 = [913], R6C9 = 7, R6C1 = 5, R8C1 = 1, R9C2 = 5

5. For 9(3) cage at R4C9, R6C9 is odd -> R4C9 must be odd
5a. 45 rule on N3 2 half-outies R1C6 + R4C9 = 3 -> R1C6 + R4C9 = 6 = {15}
5b. 9(3) cage = 0.5 + 5 + 3.5 or 2.5 + 3 + 3.5 = [157/537], 5 locked for C9 and N6

6. For 9(3) cage at R1C4, R1C6 is odd -> R1C4 must be odd
6a. 45 rule on N1 2 half-outies R1C4 + R4C1 = 6 -> R1C4 + R4C1 = {39/57} = [39/57]
6b. 9(3) cage at R1C4 = 1.5 + 7 + 0.5 or 2.5 + 6 + 0.5 (cannot be 1.5 + 5 + 2.5 because 5 would be repeated in the cage) = [371/561] -> R1C6 = 1, R4C9 = 5, R5C9 = 3, R5C3 = 1, R4C2 = 3
6c. 14(3) cage at R4C1 = 3.5 + 8 + 2.5 or 4.5 + 7 + 2.5 -> R456C1 = [785/975], 7 locked for C1 and N4

7. 3 in R6 must be in 5(3) cage at R5C4, cannot be 0.5 + 3 + 1.5 or 1.5 + 3 + 0.5 (because 3 would be repeated in the cage) -> R6C5 = 3, R5C4 must be odd -> 5(3) cage = 2.5 + 1 + 1.5 -> R56C4 = [51], R1C4 = 3, R1C5 = 7 (step 6b), R4C1 = 9 (step 6a), R5C1 = 7 (step 6c)

8. R4C1 = 9 “sees” all cells in N1 except for R1C2 and R3C3, R3C3 cannot be 9 -> R1C2 = 9
8a. R1C2 is odd -> R2C1 must be odd = 3, 12(3) cage at R1C1 = 6 + 4.5 + 1.5 -> R1C1 = 6

9. R1C2 + R2C1 = [93] = 12 -> half-cells R1C2 + R2C1 = 6 -> half-cells R1C3 + R2C2 + R3C1 = 7
9a. R1C4 + R4C1 = [39] = 12 -> half-cells R1C4 + R4C1 = 6, half-cells R1C3 + R2C2 + R3C1 = 7 -> half-cells R2C3 + R3C2 = 4 -> R3C3 = 5
9b. Half-cells R2C3 + R3C2 = 4 -> R2C3 + R3C2 = 8 = {17} -> R2C3 = 7, R3C2 = 1
9c. R2C2 = 2 (hidden single in N1)

10. 1 in R4 only in 8(3) cage at R4C7, R4C8 and R5C7 must both be even or both odd; cannot be odd = {19} because 0.5 + 3 + 4.5 clashes with R5C9 -> R4C7 = 1, half-cells R4C8 + R5C7 = 7 -> R4C8 + R5C7 = 14 = {68}

11. R9C6 = 3 “sees” all cells in N9 except for R7C7 and R9C8, 3 blocked from R9C8 -> R7C7 = 3, R8C3 = 3 (hidden single in N7)
11a. 14(3) cage at R7C2 = 3.5 + 9 + 1.5 -> R7C2 = 7

12. R6C9 = 7 “sees” all vacant cells in N9 except for R9C8 -> R9C8 = 7, R3C7 = 7 (hidden single in C7)

13. R1C6 = 1 “sees” all vacant cells in N3 except for R2C9 -> R2C9 = 1
13a. R4C9 = 5 “sees” all vacant cells in N3 except for R1C8 -> R1C8 = 5
13b. 7(3) cage at R1C8 = 2.5 + 4 + 0.5 -> R1C9 = 4

14. R3C1 = 4 (hidden single in N1), R1C3 = 8 (final cell in N1), R1C7 = 2 (final cell in R1), R7C1 = 8 (final cell in C1)

15. R3C7 = 7 -> half-cells R2C7 + R3C8 = 6 -> R2C7 + R3C8 = 12 = {39} -> R2C7 = 9, R3C8 = 3

16. R9C8 is odd -> R8C9 must be odd -> R8C9 = 9
16a. 16(3) cage at R8C9 = 4.5 + 3.5 + 8 -> R9C9 = 8
16b. R2C8 = 8 (hidden single in N3), R3C9 = 6 (final cell in N3), R7C9 = 2 (final cell in C9)
16c. R4C8 + R5C7 (step 10) = {68} -> R4C8 = 6, R5C7 = 8

17. R7C7 = 3 -> half-cells R7C8 + R8C7 = 3 -> R7C8 + R8C7 = 6 = {15} -> R7C8 = 1, R8C7 = 5, R9C7 = 6 (hidden single in N9), R8C8 = 4 (final cell in N9), R9C3 = 4 (final cell in R9), R8C2 = 6 (final cell in N7), R6C2 = 8 (hidden single in C2), R5C2 = 4 (final cell in C2), R6C3 = 6 (final cell in N4), R6C7 = 4 (final cell in C7)

18. 9 in R3 must be in 15(3) cage at R2C6, R2C6 and R3C5 must both be even or both odd -> 15(3) cage = 2 + 4 + 9 or 2.5 + 4.5 + 8 -> R2C6 + R3C56 = [489/598]
18a. R3C4 = 2 (hidden single in R3)

19. For 10(3) cage at R7C4, R7C4 and R8C5 must both be even (because there are no remaining odd candidates) -> R8C4 = 7 (cannot be 8 because R7C4 + R8C5 would require odd candidates), half-cells R7C4 + R8C5 = 3 -> R7C4 + R8C5 = 6 = {24} -> R7C4 = 4, R8C5 = 2, R8C6 = 8 (final cell in R8)

20. R8C6 = 8 -> R2C6 + R3C56 (step 18) = [489] -> R2C6 = 4, R3C56 = [89]

[And the rest are hidden singles and final cells …]
21. R2C5 = 5, R2C4 = 6, R4C4 = 8, R4C5 = 4, R4C6 = 7, R5C5 = 9, R6C8 = 9, R5C8 = 2, R5C6 = 6, R6C6 = 2, R7C5 = 6, R7C6 = 5

Solution:
6 9 8 3 7 1 2 5 4
3 2 7 6 5 4 9 8 1
4 1 5 2 8 9 7 3 6
9 3 2 8 4 7 1 6 5
7 4 1 5 9 6 8 2 3
5 8 6 1 3 2 4 9 7
8 7 9 4 6 5 3 1 2
1 6 3 7 2 8 5 4 9
2 5 4 9 1 3 6 7 8

Rating Comment:
I won't give a specific rating for this puzzle. It would be somewhere in the 1.5 range because of the use of "sees all except" steps. Maybe they are what HATMAN refers to as the HS bit, but I think software solvers can do these type of steps.


Top
 Profile  
Reply with quote  
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 4 posts ] 

All times are UTC


Who is online

Users browsing this forum: No registered users and 58 guests


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
Powered by phpBB® Forum Software © phpBB Group