SudokuSolver Forum

A forum for Sudoku enthusiasts to share puzzles, techniques and software
It is currently Sat Apr 27, 2024 7:52 am

All times are UTC




Post new topic Reply to topic  [ 4 posts ] 
Author Message
 Post subject: Assassin 445
PostPosted: Wed Nov 01, 2023 8:38 am 
Offline
Grand Master
Grand Master

Joined: Wed Apr 16, 2008 1:16 am
Posts: 1044
Location: Sydney, Australia
Attachment:
a445.png
a445.png [ 74.92 KiB | Viewed 2281 times ]
Note: cells with no cage can have repeats as per standard sudoku rules.

Assassin 445
Long time since I made a "zero" killer. Jsudoku has a terrible time with this puzzle but SudokuSolver gives it 1.55. I found it a pretty quick one but with enough interesting steps to make it worthy.

triple click code:
3x3::k:3584:3584:3584:6145:3842:3843:2820:4613:4613:5382:3335:3335:6145:3842:3843:2820:9736:4613:5382:3335:6145:6145:3842:3843:9736:9736:9736:5382:3337:6145:2314:2314:9736:9736:4875:4875:3337:3337:1292:1292:5133:9736:9736:4875:0000:5135:5135:5135:5135:5133:5133:5648:0000:0000:2322:4115:7956:7956:7956:7956:5648:0000:0000:2322:4115:5649:5649:3086:7956:5648:0000:3605:2322:5649:4115:3086:3086:7956:5648:0000:3605:
solution:
+-------+-------+-------+
| 8 2 4 | 1 6 3 | 5 9 7 |
| 5 3 1 | 9 7 8 | 6 4 2 |
| 7 9 6 | 5 2 4 | 1 3 8 |
+-------+-------+-------+
| 9 5 3 | 8 1 6 | 2 7 4 |
| 1 7 2 | 3 4 5 | 9 8 6 |
| 6 4 8 | 2 9 7 | 3 5 1 |
+-------+-------+-------+
| 4 1 5 | 6 8 9 | 7 2 3 |
| 2 8 9 | 7 3 1 | 4 6 5 |
| 3 6 7 | 4 5 2 | 8 1 9 |
+-------+-------+-------+
Cheers
Ed


Top
 Profile  
Reply with quote  
 Post subject: Re: Assassin 445
PostPosted: Sat Nov 04, 2023 4:10 am 
Offline
Grand Master
Grand Master

Joined: Wed Apr 23, 2008 6:04 pm
Posts: 1895
Location: Lethbridge, Alberta, Canada
Thanks Ed for a challenging Assassin. My solving path had an easy start, an easy finish but a pretty hard middle. I hope someone else managed to find a better way to solve the middle.

Which both Ed and wellbeback did. I've added a note about what I missed; if I'd found it, my solving path could have been like theirs, rather than harder and quite different.

Here's how I solved Assassin 445:
Prelims

a) R12C7 = {29/38/47/56}, no 1
b) R4C45 = {18/28/36/45}, no 9
c) R5C34 = {14/23}
d) R89C9 = {59/68}
e) 21(3) cage at R2C1 = {489/579/678}, no 1,2,3
f) 19(3) cage at R4C8 = {289/379/469/478/568}, no 1
g) 20(3) cage at R5C5 = {389/479/569/578}, no 1,2
h) 9(3) cage at R7C1 = {126/135/234}, no 7,8,9
i) 22(3) cage at R8C3 = {589/679}
j) 38(8) cage at R2C8 = {12345689}, no 7

1a. 45 rule on C1234567 3 outies R2C8 + R3C89 = 15, 45 rule on N3 4 innies R2C8 + R3C789 = 16 -> R3C7 = 1
1b. 45 rule on C7 2 remaining innies R45C7 = 11 = {29/38/56}, no 4
1c. 45 rule on C123456 2 innies R45C6 = 11 = {29/38/56}, no 4
1d. 38(8) cage at R2C8 = {12345689}, 4 locked for N3, clean-up: no 7 in R12C7
1e. 45 rule on N12 two outies R4C13 = 12 = {48/57}/[93], no 1,2,6, no 9 in R4C3
1f. 45 rule on N2 two outies R34C3 = 9 = [27/45/54/63], clean-up: no 4 in R4C1
1g. 45 rule on N124 two outies R56C4 = 5 = {14/23}
1h. Whatever is in R5C3 must be in R6C4 and in N6 in R4C789, no 1 in R4C789 -> no 1 in R5C3, no 1 in R6C4, clean-up: no 4 in R5C4
1i. 45 rule on R12345 2 innies R5C59 = 10 = {37/46}/[82/91], no 5, no 8,9 in R5C9
1j. 1 in N5 only in R4C45 = {18} or in R56C4 = [14] -> no 4 in R4C45 (locking-out cages), clean-up: no 5 in R4C45
[I ought also to have looked at 1 in N5 only in R4C45 = {18} or R5C34 -> R4C13 = [57/75/93] (cannot be {48}), no 4,8. Then I could have looked at combinations for the 15(3) cages in N2 including R123C4 = 15.]
1k. 22(3) cage at R8C3 = {589/679}, CPE no 9 in R8C2
1l. 45 rule on N7 1 outie R8C4 = 1 innie R7C3 + 2, no 1,2,8,9 in R7C3

2a. R4C13 = 12 (step 1f) = [57/75/84/93], R5C59 = 10 (step 1i)
2b. R5C34 = 5, R56C4 = 5 (step 1g) -> R5C3 + R6C123 = 20(4) cage at R6C1
2c. 20(4) cage = R5C3 + R6C123 = {1289/1379/1469/1478/2369/2468/2567} (cannot be {2378/2459/3458/3467} which clash with R4C13, cannot be {1568} because R6C4 only contains 2,3,4)
2d. Consider combinations for R56C4 = [14]/{23}
R56C4 = [14] => R4C2 = 1 (hidden single in R4) => 20(4) cage = R5C3 + R6C123 = {2369/2468/2567}
or R56C4 = {23}, 1 then only in R4C45 = {18}, locked for R4 => 20(4) cage = R5C3 + R6C123 = {1289/2369/2468/2567} (cannot be {1379} which clashes with R4C13 = [57/75/93] cannot be {1469/1478} because R6C4 then only contains 2,3)
-> 20(4) cage = R5C3 + R6C123 = {1289/2369/2468/2567}, 2 locked for R6 and N4
[Alternatively the eliminations of {1469/1478} can be made in a similar way to step 1h, cannot contain both of 1,4.]
2e. Consider placement for 1 in N5
1 in N5 only in R4C45 = {18}, 8 locked for R4
or in R56C4 = [14] => 20(4) cage = {268}4, 8 locked for N4)
-> R4C13 = [57/75/93], no 4,8, clean-up: no 5 in R3C3 (step 1f)
2f. Consider combinations for R45C7 (step 1b) = {29/38/56}
R45C7 = {29} blocks R5C3459 (step 1i) = {23}[91] => 1 in N6 only in R6C89, locked for R6
or R45C7 = {38}, locked for N6 and 3 in R6 in 20(4) cage = {2369}
or R45C7 = {38}, locked for N6 and 3 in R6 in 20(3) cage at R5C5, locked for N5 => R56C4 = [14] => 20(4) cage = {2468}
or R45C7 = {56} => R45C6 = {29/38} => R56C4 = [14] (cannot be {23} which then clashes with R56C4) => 20(4) cage = {2468}
-> 20(4) cage = R5C3 + R6C123 = {2369/2468/2567}, no 1, 6 locked for R6 and N4
2g. 1 in R6 only in R6C89, locked for N6, clean-up: no 9 in R5C5 (step 1i)
2h. Consider placement for 4 in N5
4 in 20(3) cage = {479}, 9 locked for R6
or R6C4 = 4 => 20(4) cage = {268}4
-> 20(4) cage = R5C3 + R6C123 = {2468/2567}, no 3,9, clean-up: no 2 in R5C4, no 3 in R5C3
2i. Consider permutations for R56C4 = [14/32]
R56C4 = [14] => 20(4) cage = {268}4
or R56C4 = [32] => R45C7 = {56}, 6 locked for N5, R5C3 = 2, R5C59 = [46] => 20(3) cage = 4{79}, 7 locked for R6
-> 20(4) cage = R5C3 + R6C123 = {2468}, 4,8 locked for R6 and N4
2j. 20(3) cage at R5C5 = {389/479/569/578}
2k. 4,6,8 only in R5C5 -> R5C5 = {468}, clean-up: no 3,7 in R5C9
2l. Killer pair 2,4 in R5C3 and R5C59, locked for R5, clean-up: no 9 in R4C6, no 9 in R4C7
2m. 4 in R4 only in R4C89, locked for N6, clean-up: no 6 in R5C5
2n. 19(3) cage at R4C8 = {469/478}, no 2,3,5
2o. 6 in N5 only in R4C45 = {36} or R45C6 (step 1c) = {56} -> no 3 in R45C6 (locking-out cages), clean-up: no 8 in R45C6
2p. 38(8) cage at R2C8 = {12345689}, CPE no 8 in R12C7, clean-up: no 3 in R12C7
2q. 6 in N5 only in R4C45 + R45C6, CPE no 6 in R4C7, clean-up: no 5 in R5C7
2r. 4 in C7 only in R789C7, locked for N9
2s. Consider combinations for 19(3) cage = {469/478}
19(3) cage = {469}, 6,9 locked for N6, R5C9 = 2 => R5C5 = 8, R45C7 = [83], R5C4 = 1, R4C23 = [13] (hidden pair in N4) => R4C1 = 9
or 19(3) cage = {478} => 9 in R4 only in R4C12
-> 9 in R4C12, locked for R4 and N4
[The last hard step, maybe the hardest step]
2t. 13(3) cage at R4C2 = {157} (cannot be {139} = 9{13} which clashes with R5C4), locked for N4 -> R4C13 = [93], R3C3 = 6, clean-up: no 6 in R4C45, no 8 in R5C7, no 5,8 in R8C4 (step 1l)
2u. 6 in N5 only in R45C6 = {56}, locked for C6 and 38(7) cage at R2C8, 5 locked for N5
2v. 9 in N5 only in R6C56, locked for R6
2w. Killer pair 2,8 in R4C45 and R4C7, locked for R4
2x. 19(3) cage = {469/478} -> R5C8 = {89}

[Fairly straightforward from here]
3a. 15(3) cage at R1C6 = {249/348}, no 1,7, 4 locked for C6 and N2
3b. 1 in C6 only in R789C6, locked for N8
3c. 12(3) cage at R8C5 = {237/246/345}, no 8,9
3d. 24(5) cage at R1C4 = {13569/23568}, no 7, 5 locked for C4 and N2
3e. Killer pair 8,9 in 24(5) cage and 15(3) cage at R1C6, locked for N2
3f. 7 in N2 only in 15(3) cage at R1C5 = {267} (only remaining combination), locked for C5, 2 locked for N2, clean-up: no 8 in 24(5) cage, no 9 in 15(3) cage at R1C6, no 2,7 in R4C45
3g. Naked triple {159} in R123C4, 1,9 locked for C4 -> R4C45 = [81], R5C3456789 = [2346986], R4C67 = [62], R6C456 = [297], clean-up: no 7 in R7C3 (step 1l)
3h. Naked pair {56} in R12C7, locked for C7 and N3 -> R6C7 = 3
3i. Naked triple {478} in R789C7, 7,8 locked for N9
3j. Naked pair {59} in R89C9, locked for C9 and N9 -> R6C89 = [51]
3h. Naked pair {35} in R89C5, locked for N8, R9C4 = 4 (cage sum)
3i. R7C5 = 8, R789C6 = {129} = 12 -> R7C34 = 11 = [56] (cannot be [47] which clashes with R7C7)
3j. R8C4 = 7 -> R8C3 + R9C2 = 15 = [96], R89C9 = [59], R89C5 = [35]
3k. 9(3) cage at R7C1 = {234} (only remaining combination), locked for C1 and N7
3l. R4C1 = 9 -> R23C1 = 12 = {57}, locked for N1, 7 locked for C1 -> R5C12 = [17], R78C2 = [18], R6C2 = 4
3m. R1C1 = 8 -> R1C23 = 6 = [24]
3n. R1C48 = [19] (hidden pair in R1) -> R12C9 = 9 = [72]

and the rest is naked singles.


Top
 Profile  
Reply with quote  
 Post subject: Re: Assassin 445
PostPosted: Tue Nov 07, 2023 7:31 pm 
Offline
Grand Master
Grand Master

Joined: Tue Jun 16, 2009 9:31 pm
Posts: 282
Location: California, out of London
Thanks again Ed! You always provide puzzles with interesting combinations!
My middle bit was quite different from Andrew's. Here's how I did it...
Assassin 445 WT:
1. Innies n2 = r123c4 = +15(3)
-> r34c3 = +9(2)
-> Remaining Outies n1 = r4c13 = +12(2)
-> Remaining Outies n4 = r56c4 = +5(2)
-> Remaining Innies n5 = r45c6 = +11(2)
-> Remaining Outies n3 = r45c7 = +11(2)
-> Remaining Innie c7 = r3c7 = 1
-> 1 in n6 in uncaged area in n6
Also 1 in n9 in uncaged area in n9

2. 38(8) has no 7
-> 4 in 38(8) in n3
-> 7 in n3 in 18(3)n3
-> Two values in r45c7 are in n3 in 18(3)n3
-> Two values in r45c6 are in n3 in 11(2)n3

3. Since r45c4 = +5(2) -> r5c3 = r6c4
-> Whatever that value is goes in n6 in r4c789
-> That value not 1
-> 5(2)r5 = [41] and r6c4 = 4 - or - 5(2)r5 = {23} and r45c4 = {32}
Combined cage H+11(2)r56c7 + 19(3)n6 = +30(5)
That cannot have two values = +5
-> (Since value from r5c3 is in r4c789) -> Value from r5c4 is in n6 in r6c789
-> Value from r5c4 is in n4 in r4c123

4. 1 in n5 only in r5c4 (-> 5(2)r5 = [41]) or in 9(2)n5 = {18}
Either way r4c13 = +12(2) cannot be {48}
-> r4c13 = [93] or {57}
-> r34c3 from [63], [27], or [45]

5! Each column in n2 is a +15(3)
Possible arrangements are {159}, {267}, {348} - or - {168}, {249}, {357}
Consider H+15(3)r123c4
It cannot be {357} since r4c3 from (357)
It cannot be {348} since one of (34) in r45c4
It cannot be {249} since one of (24) in r45c4
It cannot be {267} since that puts 5(2)r5 = [41] leaving no solution for r34c4 = +9(2)
-> H+15(3)r123c4 from {1(59|68)}
-> r45c4 = {23}
-> (HS 1 in n5) 9(2)n5 = [81]
-> H+15(3)r123c4 = {159}
-> r789c4 = {467}

6. Remaining 15(3)s in n2 are {267} and {348}
-> (Since one of (47) in r123c6) H+11(2)r45c6 = {56}
-> 15(3)r1c6 = {348} and 15(3)r1c5 = {267}
-> 20(3)n5 = [{49}7]
-> r789c6 = {129}
-> 12(3)n8 = <345>
-> r7c5 = 8
-> r78c4 = {67}
Also 11(2)n3 = {56}

7. Trying 5(2)r5 = [32] puts r4c13 = [57] and r4c2 = 2 (Step 3 last line) which leaves no solution for 13(3)n4
-> 5(2)r5 = [23]
-> r6c4 = 2
Also r34c3 = [63]
-> r4c1 = 9

8. Innies r12345 = r5c59 = +10(2) = [19] or [46]
-> 7 in n6 in 19(3)
-> (HS 2 in r4) r45c7 = [29]
-> r5c59 = [46]
-> r6c5 = 9 and r45c6 = [65]
Also 19(3)n6 = [{47}8]
-> 13(3)n4 = [5{17}]
-> r6c123 = {468} and r6c789 = [3{15}]
-> r789c7 = {478}
-> 14(2)n9 = {59}
-> 18(3)n3 = [9{27}]
-> r4c89 = [74]
-> r3c9 = 8 and r23c8 = {34}
Also -> 13(3)n1 = {139}
-> 15(3)r1c6 = [384]
-> 14(3)n1 = {248}
etc.


Top
 Profile  
Reply with quote  
 Post subject: Re: Assassin 445
PostPosted: Fri Nov 10, 2023 8:33 am 
Offline
Grand Master
Grand Master

Joined: Wed Apr 16, 2008 1:16 am
Posts: 1044
Location: Sydney, Australia
Thanks for you kind observations wellbeback! We are so lucky to have two high quality software solvers so I don't have to put in too much effort to find interesting cage structures.

I was very close to wellbeback's solving path but didn't need his step 7. Found a different way (esp steps 21,22).
A445:
Preliminaries
Cage 5(2) n45 - cells only uses 1234
Cage 14(2) n9 - cells only uses 5689
Cage 9(2) n5 - cells do not use 9
Cage 11(2) n3 - cells do not use 1
Cage 22(3) n78 - cells do not use 1234
Cage 9(3) n7 - cells do not use 789
Cage 21(3) n14 - cells do not use 123
Cage 20(3) n5 - cells do not use 12
Cage 19(3) n6 - cells do not use 1
Cage 38(8) n356 - cells ={12345689}

Andrew did his start a more direct way

1. "45" on c7: 3 innies r345c7 = 12

2. "45" on whole grid -> 7 zero cells = 24
2a. "45" on n69 (remembering h12(3)c7) -> 1 outie r3c7 = 1
2b. -> r45c7 = 11 = {29/38/56}(no 4)

3. "45" on c123456: 2 innies r45c6 = 11 (no 4)

4. "45" on n124: 1 innie r5c3 = 1 outie r6c4
4a. but can't both be 1 since there is no 1 available for r4c789 (Empty Rectangle)
4b. -> are both (234)
4c. no 4 in r5c4
4d. also r56c4 = 5 = [14]/{23}

5. 1 in n5 only in 9(2) = {18} or in r5c4 -> r5c34 = [41]:
5a. note = 4 or 8

6. "45" on n12: 2 outies r4c13 = 12
6a. but {48} leaves no 1 for n5 (bit more direct than Andrew's 2e)
6b. = [93]/{57}

7. "45" on n2: 2 outies r34c3 = 9 = [27/45/63]

8. 3 innies n2, r123c4 = 15
8a. {267} blocked by [44] in r35c3 (step 4.)
8b. {249/348} blocked by r56c4 (step 4d)
8c. {357} blocked by r4c3 = (357)
8d. = {159/168/258/456}(no 3,7)

9. 3 in n2 must be in one of the other 15(3) = {348/357}
9a. -> {258/456} blocked from r123c4
9b. = {159/168}(no 2,4)
9c. 1 locked for n2 and c4

The original version of this puzzle was now completely cracked, hence, the zero killer version.
10. r56c4 = {23}: both locked for c4 and n5
10a. no 4 in r5c3

11. r4c5 = 1 (hsingle n5) -> r4c4 = 8
11a. -> r123c4 = {159} only (step 9b): 5 and 9 locked for n2, c4, and no 5 in r4c3 -> no 4 in r3c3
11b. no 7 in r4c1 (h12(2))

12. h11(2)r45c6 = {56} only: both locked for c6, n5 and 38(8)

13. h11(2)r45c7 sees all n3 apart from 18(2) so must repeat there along with 7 (no 5,6)

14. 11(2)n3 = {56} (hpair): both locked for c7

15. 5(2)r5c3 = {23}: both locked for r5

16. 22(3)r8c3 must have 6 or 7 for r8c4 = {679} only: 9 locked for n7, no 6,7 in r8c12

17. 15(3)r1c6 = {348} only: all locked for c6 and n2

18. 15(3)r1c5 = {267}: all locked c5

19. 20(3)n5 = {49}[7]: 4,9 locked for c5

20. 12(3)n8 = {345} only -> r9c4 = 4
20a. 3,5 locked for n8
20b. r7c5 = 8

21. killer pair 2,3 in r345c3: both locked for c3, 3 for n4

22. 13(3)n4: [2]{56} blocked by r5c6 = (56)
22a. [2]{47} blocked by r45c3 = {237}
22b. = {148/157}(no 2,6,9)
22c. 1 locked for n4 and r6

23. "45" on r12345: 2 innies r5c59 = 10 = [46] only

On from there
Cheers
Ed


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 43 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