Assassin 53

Our weekly <a href="http://www.sudocue.net/weeklykiller.php">Killer Sudokus</a> should not be taken too lightly. Don't turn your back on them.
Andrew
Grandmaster
Grandmaster
Posts: 300
Joined: Fri Aug 11, 2006 4:48 am
Location: Lethbridge, Alberta

Post by Andrew »

Initially I'd got stuck on Assassin 53V2. Then after completing V1 and working through Cathy's and Para's V1 walkthroughs I started V2 again.

Here is my walkthrough

1. R2C12 = {18/27/36/45}, no 9

2. R23C5 = {59/68}

3. R2C89 = {39/48/57}, no 1,2,6

4. R34C1 = {16/25/34}, no 7,8,9

5. R34C6 = {16/25/34}, no 7,8,9

6. R34C9 = {18/27/36/45}, no 9

7. R67C1 = {18/27/36/45}, no 9

8. R67C4 = {89}, locked for C4

9. R67C9 = {13}, locked for C9, clean-up: no 9 in R2C8, no 6,8 in R34C9

10. R78C5 = {18/27/36/45}, no 9

11. R8C12 = {16/25/34}, no 7,8,9

12. R8C89 = {89}, locked for R8 and N9, clean-up: no 1 in R7C5

13. 27(4) cage at R2C3 = 9{378/468/567}, no 1,2

14. 45 rule on C5 2 innies R19C5 = 5 = {14/23}

15. 45 rule on N1 2 outies R4C12 = 12 = [39/48/57], clean-up: no 1,5,6 in R3C1

16. 45 rule on N3 2 outies R4C89 = 9 = {27/45} -> no 5,7 in R4C12 (clashes with R4C89), clean-up: no 2 in R3C1
[The elimination of 5,7 from R4C12, which is what I'd missed earlier :oops: , is I think the key move for V2. I assume this is the move that Para referred to at the end of his walkthrough. Without it I'd done about 30 moves and was stuck. After this elimination and the resulting naked pair it went very smoothly.]

17. Naked pair {34} in R34C1, locked for C1, clean-up: no 5,6 in R2C2, no 5,6 in R67C1, no 3,4 in R8C2

18. 45 rule on N7 2 outies R6C12 = 12 = [75/84], clean-up: no 7,8 in R7C1

19. Killer pair 1/2 in R7C1 and R8C12, locked for N7

20. 45 rule on N9 2 outies R6C89 = 9 = [63/81]

21. 45 rule on N2 2 innies R3C46 = 9 = {36/45}/[72], clean-up: no 6 in R4C6

22. 45 rule on N8 2 innies R7C46 = 15 = [87/96]

23. 45 rule on R1234 2 innies R4C57 = 13 = {49/58/67}, no 1,2,3

24. 45 rule on R6789 2 innies R6C35 = 10 = {19/28/37/46}, no 5

25. 45 rule on R4, R4C12 = 12 (step 15), R4C57 = 13 (step 23) and R4C89 = 9 (step 16) -> R4C346 = 11 = {128/137/146/236/245}, no 9

26. 45 rule on R6, R6C12 = 12 (step 18), R6C35 = 10 (step 24) and R6C89 = 9 (step 20) -> R6C467 = 14 with R6C4 = {89} = 8{15}/8{24}/9{14}/9{23}, no 6,7,8,9 in
R6C67
[Alternatively this result comes from combinations for 12(3) cage at R6C6]

27. 45 rule on C123 1 outie R5C4 – 1 = 1 innie R4C3 -> no 7,8 in R4C3, no 1 in R5C4

28. 45 rule on C789 1 outie R5C6 – 2 = 1 innie R6C7 -> R5C6 = {34567}

29. 45 rule on R1 2 outies R2C46 = 9 -> no 1,9 in R2C6

30. 45 rule on R9 2 outies R8C46 = 9 -> {27/36/45}, no 1

31. 45 rule on R12, 3 innies R2C357 = 15
31a. Min R2C35 = 8 -> max R2C7 = 7

32. R9C789 = {467} (only remaining combination), locked for R9 and N9, clean-up: no 1 in R1C5 (step 14)

33. R9C123 = {358} (only remaining combination), locked for R9 and N7, clean-up: no 2 in R1C5 (step 14), no 2 in R8C12

34. Naked pair {12} in R9C45, locked for R9 and N8 -> R9C6 = 9, R7C4 = 8, R6C4 = 9, R7C6 = 7 (step 22), clean-up: no 2 in R2C4 (step 29), no 7,8 in R78C5

35. Killer pair 5/6 in R23C5 and R78C5, locked for C5

36. Naked pair {16} in R8C12, locked for R8 and N7 -> R7C1 = 2, R6C1 = 7, clean-up: no 2,7 in R2C2, no 3 in R7C5

37. R7C5 = 6 (hidden single in R7) -> R8C5 = 3, R1C5 = 4, R9C5 = 1 (step 14), R9C4 = 2, clean-up: no 8 in R23C5, no 5 in R2C46 (step 29), no 5 in R3C46 (step 21), no 2,3 in R4C6, no 1 in R4C3 (step 27), no 5 in R6C7 (step 28)

38. Naked pair {49} in R7C23, locked for 25(4) cage -> R6C2 = 5, R8C3 = 7, clean-up: no 6 in R5C4 (step 27)

39. R8C7 = 2 (hidden single in R8), clean-up: no 4 in R5C6 (step 28)

40. Naked pair {59} in R23C5, locked for N2

41. Naked triple {278} in R456C5, locked for N5, clean-up: no 6 in R4C3 (step 27)

42. R7C6 = 7 -> R6C67 = 5 = {14}, no 3, clean-up: no 5 in R5C6 (step 28)

43. Naked pair {14} in R6C67, locked for R6 -> R67C9 = [31], R6C8 = 6 (cage sum)

44. 1 in N4 locked in R5C123, locked for R5

45. 1 in R4 locked in R4C46, locked for N5 -> R6C67 = [41], R8C46 = [45], R4C6 = 1, R3C6 = 6, R3C4 = 3 (step 21), R5C6 = 3, R5C4 = 5, R4C4 = 6, R4C3 = 4 (cage sum), R34C1 = [43], R7C23 = [49], clean-up: no 5 in R2C1, no 5 in R34C9

46. Naked pair {27} in R34C9, locked for C9, clean-up: no 5 in R2C8

47. 2 in R4 locked in R4C89 = {27} (step 16), locked for R4 and N6 -> R456C5 = [872], R4C2 = 9, R4C7 = 5, R6C3 = 8, R3C3 = 5, R23C5 = [59], R7C78 = [35], R9C3 = 3

48. R3C7 = 8 (hidden single in R3)

49. Naked pair {78} in R3C27, locked for R3 -> R3C9 = 2, R4C89 = [27]

50. 27(4) cage at R2C3 (step 13) = {5679} (only remaining combination) -> R2C3 = 6, R3C2 = 7, R3C7 = 8, R2C7 = 4 (cage sum), R2C9 = 9, R2C8 = 3, R1C789 = [675], R8C89 = [98], R5C789 = [984], R9C789 = [746]

51. R2C12 = {18}, locked for R2 and N1

and the rest is naked singles
Para
Yokozuna
Yokozuna
Posts: 384
Joined: Wed Nov 08, 2006 7:42 pm
Location: The Netherlands

Post by Para »

Hi all

This is the work i did on 53 V0.1 so far. Nothing spectacular yet. Maybe anyone can find a way in.

Opening 53V0.1

1. 28(4) in R2C8 = {4789/5689}: no 1,2,3; 89 locked in 28(4) cage -->> R1C9: no 8,9

2. 12(4) in R6C9 = {1236/1245}: no 7,8,9; 12 locked in 21(4) cage -->> R9C9: no 1,2

3. 8(3) in R9C1 = {125/134}: no 6,7,8,9; 1 locked for R9 and N7

4. 45 on R1: 2 outies: R2C46 = 6 = {15/24}: no 3,6,7,8,9
4a. 23(5) cage needs {15} or {24} -->> 23(5) = {12479/12569/12578/13568/14567/23459/23468}

5. 45 on R9: 2 outies: R8C46 = 12 = {39/48/57}: no 1,2,6
5a. 31(5) in R8C4 = {25789/34789/35689/45679}: 9 locked for N8

6. 45 on N1: 2 outies: R4C12 = 9 = {18/27/36/45}: no 95 on N2

7. 45 on N2: 1 outie – 1 innie : R5C6 - R3C4 = 3 -->> R3C4: no 7,8,9; R4C6: no 1,2,3
7a. 45 on N2: 3 outies: R4C346 = 15 = {159/168/249/367/348/357/456}: no {258} clashes with step 7

8. 45 on N3: 2 outies: R4C89 = 10 = [19/28/37/46/64] -->> R4C8: no 5,7,8,9; R4C9: no 5

9. 45 on N7: 2 outies: R6C12 = 7 = {16/25/34}: no 7,8,9

10. 45 on N8: 3 outies: R6C467 = 20 = {389/479/569/578}: no 1,2
10a. 45 on N8: 1 outie – 1 innie: R5C4 – R7C6 = 6 -->> R6C4 ={789}; R7C6 = {123}

11. 45 on N9: 2 outies: R6C89 = 5 = {14/23}: no 5,6,7,8,9
11a. R6C12: no {34} : clashes with R6C89
11b. Killer Pair {12} in R6C12 + R6C89 -->> locked for R6
11c. R6C467: no {569} clashes with R6C12; R6C467 = {389/479/578}: no 6

12. 45 on R1234: 2 innies: R4C57 = 11 = {29/38/47/56} : no 1

13. 45 on R6789: 2 innies: R6C35 = 13 = {49/58/67} : no 3

14. 45 on C123: 3 outies: R345C4 = 10 = {127/136/145/235}: no 8,9
14a. 45 on C123: 1 innie – 1 outie: R4C3 – R5C4 = 2 -->> R4C3: no 1,2

15. 45 on C789: 1 innie = 1 outie: R6C7 = R5C6-->> R5C6: no 1,2,6
15a. 45 on C789: 3 outies: R567C6 = 14

16. 14(3) in R6C6: no {248}: clashes with step 11c

17. 12(3) in R4C5: min R6C5 = 4 -->> max R45C5 = 8: no 8,9; min R46C5 = 6 -->> max R5C5 = 6: no 7
17a. Clean up: R4C7: no 2,3 (step 12)

18. 12(4) in R6C9 needs one of {56} in N9, so no combinations with both {56} in 20(4) in R6C8 or 18(3) in R9C7
18a. 20(4): no {1568/2567}
18b. 18(3) = {279/369/378/468}: no {567} and no {459} clashes with 8(3) in R9C1 -->> no 5
18c. 8(3) in R9C1 = {125} or {134}; When {134} -> 18(3) = {279} -->> 2 locked for R9: R9C456: no 2

19. 20(4) in R6C4 has 1-6 only in N8.
19a. Only place for 9 is R6C4. R6C4 = 9 -> R7C6 = 3(step 10a). So 20(4) can’t contain both a 3 and 9
19b. 31(5) = {34789/35689/45679}: needs at least one of {34/35/36/45/46} so 20(4) can’t have any combinations containing both digits.
19c. 20(4) = {1289/1478/1568/2378/2567} = [9]{128}/[8]{147/156}/ [7]{238/256}

20. 30(5) in R4C7 can’t have both {24} or {34} because of R6C89. R5C6 doesn’t see R6C89 but R6C7 = R5C5(step 15) so it sees R6C89 through that. -->> 30(5) = {15789/25689/35679/45678}

20 steps but nowhere close to something good.

Here's a marks pic.

Code: Select all

.-----------------------------------.-----------------------------------.-----------------------------------.
| 123456789   123456789   123456789 | 123456789   123456789   123456789 | 123456789   123456789   1234567   |
&#58;-----------------------.-----------&#58;           .-----------.           &#58;-----------.-----------------------&#58;
| 123456789   123456789 | 123456789 | 1245      | 123456789 | 1245      | 123456789 | 456789      456789    |
|           .-----------'           &#58;-----------&#58;           '-----------&#58;           '-----------.           |
| 123456789 | 123456789   123456789 | 123456    | 123456789   123456789 | 123456789   123456789 | 456789    |
|           |           .-----------'           &#58;-----------.           &#58;-----------.           |           |
| 12345678  | 12345678  | 3456789     1234567   | 234567    | 456789    | 456789    | 12346     | 46789     |
&#58;-----------'-----------'-----------------------&#58;           &#58;-----------'           '-----------'-----------&#58;
| 123456789   123456789   123456789   1234567   | 123456    | 345789      123456789   123456789   123456789 |
&#58;-----------.-----------.           .-----------&#58;           &#58;-----------------------.-----------.-----------&#58;
| 1256      | 1256      | 456789    | 789       | 456789    | 345789      345789    | 1234      | 1234      |
|           |           '-----------&#58;           '-----------&#58;           .-----------'           |           |
| 23456789  | 23456789    23456789  | 12345678    12345678  | 123       | 123456789   123456789 | 123456    |
|           '-----------.           &#58;-----------.           &#58;-----------&#58;           .-----------'           |
| 23456789    23456789  | 23456789  | 345789    | 12345678  | 345789    | 123456789 | 123456      123456    |
&#58;-----------------------'-----------&#58;           '-----------'           &#58;-----------'-----------------------&#58;
| 12345       12345       12345     | 3456789     3456789     3456789   | 2346789     2346789     346789    |
'-----------------------------------'-----------------------------------'-----------------------------------'
Andrew wrote:16. 45 rule on N3 2 outies R4C89 = 9 = {27/45} -> no 5,7 in R4C12 (clashes with R4C89), clean-up: no 2 in R3C1
[The elimination of 5,7 from R4C12, which is what I'd missed earlier :oops: , is I think the key move for V2. I assume this is the move that Para referred to at the end of his walkthrough. Without it I'd done about 30 moves and was stuck. After this elimination and the resulting naked pair it went very smoothly.]
Yes, that's the one. It is all the help sumocue needed to solve the puzzle.

greetings

Para
mhparker
Grandmaster
Grandmaster
Posts: 345
Joined: Sat Jan 20, 2007 10:47 pm
Location: Germany

Post by mhparker »

Andrew wrote:16. 45 rule on N3 2 outies R4C89 = 9 = {27/45} -> no 5,7 in R4C12 (clashes with R4C89), clean-up: no 2 in R3C1
[The elimination of 5,7 from R4C12, which is what I'd missed earlier :oops: , is I think the key move for V2. I assume this is the move that Para referred to at the end of his walkthrough. Without it I'd done about 30 moves and was stuck. After this elimination and the resulting naked pair it went very smoothly.]
Para wrote:Yes, that's the one. It is all the help sumocue needed to solve the puzzle.
Well, that's true, but the current version of Sumocue also has a more glaring bug, which also caused it to fail to solve the puzzle:

It for some reason overlooked the C5 innies (R19C5 = 5/2). When Sumocue got stuck (with all candidates still in R1C5, and R9C5 = {12389}), I had to manually clear out the {89} from R9C5 and {156789} from R1C5. This then caused it to correctly get R9C5 = 1, but it still didn't get R1C5 = 4! Manually setting R1C5 to 4 was then enough for Sumocue to solve the puzzle.

I think I've seen this bug before, and I think it was the C5 innies that time as well. Hopefully, Ruud will see this post and take a look.
Cheers,
Mike
mhparker
Grandmaster
Grandmaster
Posts: 345
Joined: Sat Jan 20, 2007 10:47 pm
Location: Germany

Post by mhparker »

sudokuEd wrote:But no one has posted about Jean-Christophe's V2.5 or V0.1. I don't like puzzles having no response. But it shows that we probably had too many versions this week and in hindsight, didn't need those two.
I agree. In retrospect, we didn't need the V2.5. I suspect what happened is that JC first of all tried to use an identical cage pattern to the original. But, as he said at the time, the cage pattern made it impossible to produce a much more difficult puzzle. I can confirm this, because I also experimented with generating several random V2's based directly on this design. Despite lengthy efforts, my program couldn't come up with a single one that Sumocue couldn't easily solve. Seems like JC had the same problems, explaining why he shortly afterwards took a different approach and posted a second puzzle (the V3), with some cages having been merged.

As for the V0.1, I don't know why JC posted it so early after Ed announced the breakthrough move for the V3, before we were even through with the V3. But I sense some informal competition between man and machine (i.e., JSudoku) here. I hope not. That would be like comparing apples and oranges.

BTW, I'm not sure how seriously JC intended us to take the V0.1. After all, why call it a V0.1 and not a V4, if not to stress its inofficiality? Well, whatever, at least Para's made a start with it now.
sudokuEd wrote:I really value the (mainly) hand-solving ethic of this forum, including that we are mostly working on the same puzzles. Too many versions ... doesn't fit in with this ethos in my opinion. We could burn out or lose too much of the group feeling.
Agree again. Too much competition between us and JSudoku (or any other computer-based solver for that matter) can only be detrimental to the popularity of this forum.
sudokuEd wrote:I'm not suggesting putting a number figure on how many versions/puzzles per week. But am asking that we think about the (slower?) pace of this forum before posting a puzzle.
Andrew wrote:Since we don't want to discourage any forum member from posting variants, I would suggest that anyone except Ruud should limit themselves to one variant each week.
On reflection, I don't think a limit is a good idea, since there may be bone-fide reasons for somebody to post multiple puzzles in a particular week (e.g., to train particular skills). But, in general, one or two variants a week may be a good guideline. Also, it's a good idea to wait a while until the previous puzzles have been solved, unless the new puzzle is intended to appeal to different forum members. For example, posting a very difficult V2 up front when Ruud has only produced a single, relatively straightforward Assassin, would be considered OK.

On a final note, many thanks to JC for an excellent A53V3. Keep 'em coming!
Cheers,
Mike
Andrew
Grandmaster
Grandmaster
Posts: 300
Joined: Fri Aug 11, 2006 4:48 am
Location: Lethbridge, Alberta

Post by Andrew »

Andrew wrote:Since we don't want to discourage any forum member from posting variants, I would suggest that anyone except Ruud should limit themselves to one variant each week.
mhparker wrote:On reflection, I don't think a limit is a good idea, since there may be bone-fide reasons for somebody to post multiple puzzles in a particular week (e.g., to train particular skills). But, in general, one or two variants a week may be a good guideline. Also, it's a good idea to wait a while until the previous puzzles have been solved, unless the new puzzle is intended to appeal to different forum members. For example, posting a very difficult V2 up front when Ruud has only produced a single, relatively straightforward Assassin, would be considered OK.
Maybe I didn't make myself clear. What I meant was that, in general, forum members should voluntarily limit themselves unless there is a particular reason for more than one.
mhparker wrote:On a final note, many thanks to JC for an excellent A53V3. Keep 'em coming!
I'll support Mike on that!
mhparker
Grandmaster
Grandmaster
Posts: 345
Joined: Sat Jan 20, 2007 10:47 pm
Location: Germany

Post by mhparker »

Para wrote:20 steps but nowhere close to something good.
Hi Para,

Thanks for getting the ball rolling.

Surprise, surprise! - I got stuck around this point in the game, too. Had to use a hypothetical to continue, as shown below:


Assassin 53V0.1 Walkthrough (continued)

21. Outies-innies R5: R4C7+R6C3 = R5C5 + 12
21a. R4C7 and R6C3 cannot both be 9, as shown by following grouped x-cyle:
(9)r4c7-r45c6=r6c456-r6c3
21b. -> max. R4C7+R6C3 = 17
21c. -> max. R5C5 = 5
21d. -> no 6 in R5C5

22. R5C6 = R6C7 (step 15)
22a. R6C467 = 20/3 (step 10)
22b. -> R6C46+R5C6 = h20(3)n5
22c. -> R45C4+R4C6 = h13(3)n5

23. Following hypothetical removes {46} from R4C89:
23a. r4c89={46} -> r6c89={23} -> r6c12={16} -> r45c4=[16] (hsingles r4/r5)
23b. -> contradiction, because h13(3)n5 (step 22c) cannot be {166}
23c. -> no {46} in h10(2)r4 in R4C89

24. 6 in N6 locked in 30(5) at R4C7 = {25689/35679/45678} (no 1)

25. CPE: R8C8 sees all 1's in N6
25a. -> no 1 in R8C8
25b. -> 1 in 12(4) at R6C9 now locked in C9
25c. -> no 1 in R1C9

26. 28(4) at R2C8 must contain 1 of {45} and 1 of {46} (not distinct) (step 1)
26a. {456} in 28(4) now only available in N3
26b. -> 12(3) at R1C7 cannot contain both of {45} or both of {46}
26c. -> {246} and {345} both blocked
26d. -> 12(3) at R1C7 = {129/138/147/237/156} = {(1/7)..}
26e. -> 16(3) at R1C1 cannot contain both of {17}
26f. -> {178} blocked
26g. -> 16(3) at R1C1 = {169/259/349/268/358/367/457}
26h. Similarly, h17(3) at R1C456 also cannot contain both of {17} (step 26d)
26i. -> {179} blocked
26j. -> no 1 in R1C456

27. 15(4) at R2C7 cannot contain both of {46} due to 28(4) at R2C8 (steps 26, 26a)
27a. -> {2346} blocked
27b. -> 15(4) = {1(239/248/257/347/356)}
27c. CPE: R1C8 sees all 1's in 15(4) at R2C7
27d. -> no 1 in R1C8
27e. Cleanup: no 8,9 in R1C7 (would require 1 in R1C89 - unavailable)

28. Only combo with 1 for 16(3) at R1C1 is {169} (step 26g)
28a. -> 12(3) at R1C7 = {237} (only non-conflicting combination)
28b. but this is impossible, because it would leave both of {45} in R1C456 -> clash w/ R2C46
28c. -> {169} blocked for 16(3) at R1C1
28d. -> no 1 in R1C123

29. Hidden single in R1 at R1C7 = 1
29a. Cleanup: no 2,3 in R1C8

30. Hidden single in 15(4) at R2C7 (step 27b) -> R4C8 = 1
30a. -> R4C9 = 9 (step 8)
30b. Cleanup: no 9 in R5C6 (step 15); no 4 in R6C89 (step 11), no 8 in R4C12 (step 6),
no 7 in R5C4 (step 14a), no 2 in R4C5 (step 12), no 6 in R3C4 (step 7)
30c. 8 in 28(4) at R2C8 now locked in N3 -> no 8 elsewhere in N3
30d. Cleanup: no 3 in R1C9 ({138} blocked for 12(3) because none of these digits in R1C8)

31. R6C89 = {23}, locked for R6 and N6
31a. Cleanup: no 5 in R6C12 (step 9), no 3 in R5C6 (step 15)

32. R6C12 = {16}, locked for R6 and N4
32a. Cleanup: no 3 in R4C12 (step 6), no 4 in R5C4 (step 14a), no 7 in R6C35 (step 13)

33. 9 in R5 locked in N4 -> not elsewhere in N4
33a. Cleanup: no 4 in R6C5 (step 13)

34. 12(3) at R4C5: no 9 in R6C5 (would require either of {12} in R4C5 - unavailable)
34a. -> R6C35 (step 13) = {58}, locked for R6
34b. Cleanup: no 5,8 in R5C6 (step 15), no 2 in R7C6 (step 10a)

35. 2 in C6 locked in N2 -> not elsewhere in N2
35a. Cleanup: no 5 in R4C6 (step 7), no 4 in R2C6 (step 4)

36. h20(3) at R6C467 (step 10) = {479}
36a. 4 locked in R6C67
36b. but R5C6 = R6C7 (step 15)
36c. -> 4 locked in R56C6 for C6 and N5
36d. Cleanup: no 1 in R3C4 (step 7), no 7 in R4C7 (step 12)

37. 12(3) at R3C4 = {(27|45)3} (no 6,8)

Maybe you or someone else can pick it up again from here?
Cheers,
Mike
mhparker
Grandmaster
Grandmaster
Posts: 345
Joined: Sat Jan 20, 2007 10:47 pm
Location: Germany

Post by mhparker »

BTW, here's a new A53V0.1 marks pic after step 37:

Code: Select all

.-----------------------------------.-----------------------------------.-----------------------------------.
| 23456789    23456789    23456789  | 3456789     3456789     2356789   | 1           45679       24567     |
&#58;-----------------------.-----------&#58;           .-----------.           &#58;-----------.-----------------------&#58;
| 123456789   123456789 | 123456789 | 145       | 13456789  | 125       | 2345679   | 45678       45678     |
|           .-----------'           &#58;-----------&#58;           '-----------&#58;           '-----------.           |
| 123456789 | 123456789   123456789 | 345       | 13456789    12356789  | 2345679     2345679   | 45678     |
|           |           .-----------'           &#58;-----------.           &#58;-----------.           |           |
| 2457      | 2457      | 3457        2357      | 3567      | 678       | 4568      | 1         | 9         |
&#58;-----------'-----------'-----------------------&#58;           &#58;-----------'           '-----------'-----------&#58;
| 2345789     2345789     2345789     12356     | 1235      | 47          45678       45678       45678     |
&#58;-----------.-----------.           .-----------&#58;           &#58;-----------------------.-----------.-----------&#58;
| 16        | 16        | 58        | 79        | 58        | 479         47        | 23        | 23        |
|           |           '-----------&#58;           '-----------&#58;           .-----------'           |           |
| 23456789  | 23456789    23456789  | 12345678    12345678  | 13        | 23456789    23456789  | 123456    |
|           '-----------.           &#58;-----------.           &#58;-----------&#58;           .-----------'           |
| 23456789    23456789  | 23456789  | 345789    | 12345678  | 35789     | 23456789  | 23456       123456    |
&#58;-----------------------'-----------&#58;           '-----------'           &#58;-----------'-----------------------&#58;
| 12345       12345       12345     | 3456789     3456789     356789    | 2346789     2346789     34678     |
'-----------------------------------'-----------------------------------'-----------------------------------'
Cheers,
Mike
Para
Yokozuna
Yokozuna
Posts: 384
Joined: Wed Nov 08, 2006 7:42 pm
Location: The Netherlands

Post by Para »

Very nice work here's some quick steps.

38. First some clean up: R8C4: no 8 (step 5); R4C4: no 7(only place for 2 in 12(3) cage in R3C4); R5C4: no 6 (step 14a)

39. 30(5) cage in R4C7 = {45678}; 7 in 30(5) cage locked for R5.

40. 12(3) cage in R4C5 needs one of {58} in R6C5 -->> 12(3) = [318/615] -->> R4C5 = {36}; R5C5 = 1
40a.R8C9 = 1(hidden single)
40b. Clean up: R4C3: no 3(step 14a); R4C7: no 4,6(step 12)

41. 3 in 12(3) cage for R3C4 locked for C4
41a. Clean up: R4C3: no 5; R8C6: no 9(step 5)
41b. R3C4: no 4 (R4C3 needs 4 or 7 so they can't be anywhere else)
41c. Clean up: R4C6: no 7(step 7)

42. Killer Pair {47} in R4C12 + R4C3 -->> locked for R4 and N4

43. Naked Triple {235} in R345C4 -->> locked for C4
43a. Clean up: R2C6: no 1; R8C6: no 7

44. Follow up on step 19c: 20(4) in R6C4 = [9]{128}/7{238/256} -->> R7C45 + R8C5: no 4,7

45. 4,7 in N8 locked in 31(5) in R8C4 -->> 31(5) = {34789/45679}
45a. When {45679}-->> R8C46 = [75]; When {34789} no 5 -->> R9C56: no 5

46. 5 in R9 locked in 8(3) in R9C1 -->> 8(3) = {125} -->> locked for R9 and N7

47. 23(4) in R6C2: no 2 or 5 and needs one of {16} -->> 23(4) = {1679}
47a. R6C2 = 1(only place for 1); R6C1 = 6; {679} locked in 23(4) cage for N7

48. 23(5) in R1C4 can't have both {35} and needs a 5 when there is a 1(only 1 is in R2C4 which implicates a 5 in R3C6)
48a. 23(5) = {12569/12578/14567/23468}
48b. Killer Pair {35} in 23(5) cage and R3C4 -->> locked for N2
48c. When {12569/12578/14567}: R2C46 = {15}: R1C56: no 5; When {23468}: no 5 -->> R1C56: no 5

This my progress so far.

[edit] Some extra steps. Now it's bedtime.

49. R8C46: no {48}: clashes with R8C12

50. 4 in N8 locked for R9

51. 18(3) in R9C7 = {369/378} -->> 3 locked for R9 and N9

52. 20(4) in R6C8 can’t have both {46} due to the 12(4) cage in R6C9: {2468} blocked

53. Pointing Naked Pair {23} in R6C89: R8C8: no 2
53a. 2 in 12(4) cage locked for C9
53b. Clean up: R1C8: no 9

54. 9 in N3 locked in 15(4) cage -->> 15(4) cage = {239}

55. 12(3) in R1C7 = 1{47/56}
55a. Can’t have both {45}.{46},{57} or {67} in other cages in R1
55b. 16(3) in R1C1: {367/457} -->> 16(3) = {259/268/349/358}: no 7
55c. Hidden 17(3) cage in R1C456: no {467} -->> 17(3) = {269/278/368}: no 4

56. Naked pair {79} in R68C4 -->> locked for C4
56a. 16(3) in R1C1 = {259/349/358}: {268} blocked by R1C4 -->> no 6


Marks pic.

Code: Select all

.-----------------------------------.-----------------------------------.-----------------------------------.
| 234589      234589      234589    | 68          36789       236789    | 1           4567        4567      |
&#58;-----------------------.-----------&#58;           .-----------.           &#58;-----------.-----------------------&#58;
| 12345789    23456789  | 123456789 | 14        | 46789     | 25        | 239       | 45678       45678     |
|           .-----------'           &#58;-----------&#58;           '-----------&#58;           '-----------.           |
| 12345789  | 23456789    123456789 | 35        | 46789       126789    | 239         239       | 45678     |
|           |           .-----------'           &#58;-----------.           &#58;-----------.           |           |
| 2457      | 2457      | 47          235       | 36        | 68        | 58        | 1         | 9         |
&#58;-----------'-----------'-----------------------&#58;           &#58;-----------'           '-----------'-----------&#58;
| 23589       23589       23589       25        | 1         | 47          45678       45678       45678     |
&#58;-----------.-----------.           .-----------&#58;           &#58;-----------------------.-----------.-----------&#58;
| 6         | 1         | 58        | 79        | 58        | 479         47        | 23        | 23        |
|           |           '-----------&#58;           '-----------&#58;           .-----------'           |           |
| 348       | 679         679       | 168         23568     | 13        | 2456789     2456789   | 2456      |
|           '-----------.           &#58;-----------.           &#58;-----------&#58;           .-----------'           |
| 348         348       | 679       | 79        | 23568     | 35        | 2456789   | 456         1         |
&#58;-----------------------'-----------&#58;           '-----------'           &#58;-----------'-----------------------&#58;
| 125         25          125       | 468         46789       6789      | 36789       36789       3678      |
'-----------------------------------'-----------------------------------'-----------------------------------'
greetings

Para
mhparker
Grandmaster
Grandmaster
Posts: 345
Joined: Sat Jan 20, 2007 10:47 pm
Location: Germany

Post by mhparker »

Hi Para,

Here's the next batch of moves for the V0.1. BTW, liked the clever logic you used in step 48.

We should have seen the next move ages ago...

57. CPE: R5C4 can see all 5's in R6
57a. -> no 5 in R5C4
57b. -> R5C4 = 2
57c. -> R4C3 = 4 (step 14a)
57d. Cleanup: no 5 in R4C12

58. Pointing naked pair {27} in R4C12 -> no 2,7 in R2C2

59. R46C5 must have one of {68}
59a. -> R46C5 and R4C6 form complex naked pointing pair on {68}
59b. -> no 6,8 in R23C5

60. 25(4) at R2C5 must have one of {68} due to R4C6, {35} unavailable
60a. {2689} blocked due to R23C5 (which only have 1 of {2689} between them)
60b. -> 25(4) at R2C5 = {(19/46)78} (no 2)
60c. -> 7 locked in R23C5+R3C6 for N2, 8 locked in R34C6 for C6

61. 7 in R1 locked in C89 -> R1C89 = {47}, locked for R1 and N3

62. 5 in R1 locked in N1 -> not elsewhere in N1

63. 6 in R1 locked in N2 -> not elsewhere in N2

64. 7 in C1 locked in 18(4) at R2C1 = {7..}, 5 unavailable
64a. -> 18(4) = {1278/1467/2367} (no 9)
64b. if {1278}, then 8 must go in R2C2 (since it doesn't have {12} and 7 is locked in C1)
64c. if {1467} or {2367}, then 6 must go in R2C2 (only place for 6 in cage)
64d. -> R2C2 = {68}, no 8 in R23C1

65. Naked triple on {568} in R2 at R2C289 -> no 5,6,8 elsewhere in R2

This triggers off quite a few singles. Seems like a good place to do a handover...
Cheers,
Mike
mhparker
Grandmaster
Grandmaster
Posts: 345
Joined: Sat Jan 20, 2007 10:47 pm
Location: Germany

Post by mhparker »

OK, I'll do the easy bit as well... :)

66. Naked single (NS) at R2C6 = 2
66a. -> R2C4 = 4 (step 4)

67. Hidden single (HS) in C4 at R7C4 = 1

68. NS at R7C6 = 3
68a. -> R6C4 = 9 (step 10a)

69. NS at R8C46 = [75]

70. HS in C5 at R9C5 = 4
70a. -> Split 15(2) cage at R9C46 = [69] (only remaining permutation)

71. NS at R1C46 = [86]
71a. -> R1C5 = 3 (cage remainder)

72. NS at R3C4 = 5
72a. -> R4C4 = 3

73. NS at R4C56 = [68]
73a. -> R4C7 = 5

74. NS at R6C5 = 5
74a. -> R6C3 = 8

75. HS in C6 at R3C6 = 1

76. 16(3) at R1C1 = {259}, locked for N1

77. 18(3) at R9C7 = {378}, locked for N9

78. 8 in C1 locked in N7 -> not elsewhere in N7

79. 20(4) at R2C3 must have 1 of {27} due to R4C2, with {59} unavailable
79a. -> {1478/2378/2468/3467}
79b. {1478} blocked by R3C3+R4C2
79c. {2468} blocked by R2C3
79d. 20(4) at R2C3 = {2378/3467} (no 1) = {(4/8)..}
79e. -> 3 locked in R2C3+R3C23 for N1
79f. {48} only in R3C2
79g. -> R3C2 = {48}

80. HS in C3 at R9C3 = 1

81. HS in C1 at R2C1 = 1

82. HS in C3 at R1C3 = 2

83. HS in C3 at R5C3 = 5

84. 3 in C3 locked in N1 -> not elsewhere in N1

85. 9 in C3 locked in N7 -> not elsewhere in N7

86. 20(4) at R6C8 must have 1 of {23}, w/ {178} unavailable
86a. -> {2369/2459} = {(4/6)..}
86b. {46} only in R7C78+R8C7
86c. -> R7C78+R8C7 and R8C8 form killer pair on {46} in N9
86d. -> no 4,6 elsewhere in N9
86e. 5 in 20(4) only in R7C8
86f. -> no 4 in R7C8

It's getting more difficult again now... :shock:
Cheers,
Mike
mhparker
Grandmaster
Grandmaster
Posts: 345
Joined: Sat Jan 20, 2007 10:47 pm
Location: Germany

Post by mhparker »

Here's the A53V0.1 candidate grid after step 86f:

Code: Select all

.--------------------.--------------------.--------------------.
| 59     59     2    | 8      3      6    | 1      47     47   |
&#58;-------------.------&#58;      .------.      &#58;------.-------------&#58;
| 1      68   | 37   | 4    | 79   | 2    | 39   | 568    568  |
|      .------'      &#58;------&#58;      '------&#58;      '------.      |
| 47   | 48     367  | 5    | 79     1    | 239    239  | 68   |
|      |      .------'      &#58;------.      &#58;------.      |      |
| 27   | 27   | 4      3    | 6    | 8    | 5    | 1    | 9    |
&#58;------'------'-------------&#58;      &#58;------'      '------'------&#58;
| 39     39     5      2    | 1    | 47     4678   4678   4678 |
&#58;------.------.      .------&#58;      &#58;-------------.------.------&#58;
| 6    | 1    | 8    | 9    | 5    | 47     47   | 23   | 23   |
|      |      '------&#58;      '------&#58;      .------'      |      |
| 48   | 67     679  | 1      28   | 3    | 2469   2569 | 25   |
|      '------.      &#58;------.      &#58;------&#58;      .------'      |
| 348    34   | 69   | 7    | 28   | 5    | 2469 | 46     1    |
&#58;-------------'------&#58;      '------'      &#58;------'-------------&#58;
| 25     25     1    | 6      4      9    | 378    378    378  |
'--------------------'--------------------'--------------------'
Maybe somebody wants to assist in looking for a chain here?

Cathy? Para? Ed? - Anyone's more than welcome to take over from here! :D
Cheers,
Mike
Para
Yokozuna
Yokozuna
Posts: 384
Joined: Wed Nov 08, 2006 7:42 pm
Location: The Netherlands

Post by Para »

Just some more steps.

87. 20(4) in R6C8 = {2369/2459}: {29} locked in 20(4) cage.
87a. 20(4) needs one of {35} in R67C8 and can't have both {29} in R78C7 because of R23C7 (needs 2 of {239}).
87b. 20(4) needs one of {29} in R67C8 and one in R78C7.
87c. Killer Triple {239} in R23C7 + R78C7 -->> locked for C7
87d. When {2369} R6C8 = 3, so R7C8 = {29}(step 87b); When {2459} R7C8 = 5 -->> R7C8: no 6

88. 3 in C7 locked for N3

This one to solve it.

89. "Killer" XY-chain from R7C1
89a. R7C1 = 4: R8C12 <> 4
89b. R7C1 = 8 -->> R7C5 = 2 -->> R7C9 = 5 -->> R8C8 = 4: R8C12 <> 4
89c. Conclusion: R8C12 <> 4

Now all naked singles :D

90. R8C2 = 3; R78C1 = [48]; R78C5 = [82]; R3C1 = 7; R4C12 = [27]
90a. R5C12 = [39]; R9C12 = [52]; R1C12 = [95]; R2C2 = 8; R2C3 = 3
90b. R3C23 = [46]; R8C3 = 9; R7C23 = [67]; R23C5 = [79]; R2C7 = 9
90c. R3C78 = [32]; R3C9 = 8; R6C89 = [32]; R7C7 = 2; R7C89 = [95]
90d. R8C78 = [64]; R1C89 = [74]; R2C89 = [56]; R9C789 = [783]
90e. R6C67 = [74]; R5C6789 = [4867]

And we are done.

greetings

Para
Para
Yokozuna
Yokozuna
Posts: 384
Joined: Wed Nov 08, 2006 7:42 pm
Location: The Netherlands

Post by Para »

Hi all

As Ed remarked in his post 53V2.5 doesn't have a walk-through yet. Here it is.

Pic A53V2.5:

Image
Walk-through Assassin 53V2.5

1. R2C12 = {49/58/67}: no 1,2,3

2. R23C5, R67C9 and R78C5 = {19/28/37/46}: no 5

3. R2C89 = {14/23}: no 5,6,7,8,9

4. R34C1 = {39/48/57}: no 1,2,6

5. 21(3) at R3C4 = {489/579/678}: no 1,2,3

6. R34C6 and R8C89 = {59/68}: no 1,2,3,4,7

7. R34C9 = {17/26/35}: no 4,8,9

8. R67C1 = {29/38/47/56}: no 1

9. R67C4 = {15/24}: no 3,6,7,8,9

10. R8C12 = {18/27/36/45}: no 9

11. 45 on R1: 2 outies: R2C46 = 8 = {17/26/35}: no 4,8,9

12. 45 on R12: 3 innies: R2C357 = 19 = {289/379/469/578}({478} blocked by R2C12): no 1
12a. Clean up: R3C5: no 9

13. 45 on R1234: 2 innies: R4C57 = 7 = {16/25/34}: no 7,8,9

14. 45 on R9: 2 outies: R8C46 = 7 = {16/25/34}= {3|5|6..}: no 7,8,9
14a. R8C89 = {5|6..} and R8C46 = {3|5|6..} -->> R8C12: {36} blocked(R8C4689 needs at least one of {36}) -->> R8C12 = {18/27/45}: no 3,6 = {2|5|8..}
14b. R8C89 = {5|8..} and R8C12 = {2|5|8..} -->> R8C46 = {25} blocked(R8C1289 nees at least one of {25}) -->> R8C46 = {16/34} = {4|6..}: no 2,5
14c. R78C5 = {19/28/37}: {46} blocked by R8C46: no 4,6
14d. R8C46 = {4|6..} and R8C89 = {5|6..} -->> R8C12 = {18/27}:{45} blocked by R8C4689(needs at least one of {45}): no 4,5

15. 45 on R6789: 2 innies: R6C35 = 11 = {29/38/47/56}: no 1

16. 45 on N1: 2 outies: R4C12 = 11 = [92]/{38}/{47}/[56] -->> R4C2: no 1,5,9

17. 45 on N2: 3 outies: R4C346 = 22 = {589/679}: no 4; 9 locked for R4
17a. Clean up: R3C1: no 3; R4C2: no 2
17b. R4C12 = [56] blocked by R4C346 -->> R4C1: no 5; R4C2: no 6
17c. R4C12 = {38/47} = {7|8..}/{3|4..}
17d. Killer Pair {78} in R4C12 + R4C346 -->> locked for R4
17e. R4C57 = {16/25}: {34} blocked by R4C12: no 3,4

18. 45 on N2: 2 innies: R3C46 = 13 = [49/58/76/85]: R3C4: no 6,9

19. 45 on N3: 2 outies: R4C89 = 5 = [41]/{23}-->> R4C8 = {234}; R4C9 = {123}
19a. Clean up: R3C9 = {567}

20. 45 on N7: 2 outies: R6C12 = 11 = {29/38/47/56}: no 1
20a. 1 in N4 locked for R5

21. 45 on N8: 3 outies: R6C467 = 8 = {125/134}: no 6,7,8,9; 1 locked for R6
21a. Clean up: R7C9: no 9

22. 45 on N8: 2 innies: R7C46 = 12 = [48/57]: R7C4 = {45}; R7C6 = {78}
22a. Clean up: R6C4: no 4,5

23. 45 on N9: R6C89 = 15 = {69/78}: no 2,3,4,5
23a. Clean up: R7C9: no 6,7,8

24. 45 on C123: 1 innie and 1 outie: R4C3 = R5C4 + 4 -->> R4C3 = {6789}; R5C4 = {2345}

25. 45 on C5: 2 innies: R19C5 = {29/38/47/56}: no 1

26. 45 on C789: 1 innie and 1 outie: R5C6 = R6C7 + 5 -->> R5C6 = {6789}; R6C7 = {1234}

27. 45 on C789: 3 outies: R567C6 = 19 = [658/748/847/937]:[928] blocked by R34C6: no 1,2
27a. Combined with step 26: R6C67 = [51/43/34]: [42] blocked by step 21: R6C7: no 2
27b. Clean up: R5C6: no 7(step 26)

28. 45 on C123: 3 outies: R345C4 = {78}[2]/[593]/[863]/[764]: {269} blocked by R3C4, {458} blocked by R7C4 -->> R3C4 = {578}; R4C4 = {6789}; R5C4 = {234}
28a. Clean up: R4C3: no 9
28b. 9 in R4 locked for N5
28c. R567C6 = [658/847]: 8 locked for C6; R6C6: no 3
28d. Clean up: R6C7: no 4(step 26); R34C6: no 6

29. R34C6 = {59} -->> locked for C6
29a. R6C6 = 4; R57C6 = [87](step 28c); R6C7 = 3
29b. R7C4 = 5(step 22); R6C4 = 1; R4C89 = [41](step 19); R3C9 = 7
29c. R3C4 = 8

30. Whole lot of clean up: R4C12 = {38}(step 16) -->> locked for R4 and N4
30a. R4C57 = {25}(step 13) -->> locked for R4
30b. R346 = [59]
30c. R2C46: no 3(step 13); R6C12: no 7(step 20); R6C3: no 2,7 and R6C5: no 7(step 15)
30d. R6C8 = 7(hidden); R6C9 = 8(hidden); R7C9 = 2; R9C7 = 7(hidden)
30e. R8C4: no 3; R8C6: no 6(step 14); R7C5: no 3; R8C5: no 3,8
30f. R23C5: no 2; R2C5: no 3; R1C5: no 4,6; R9C5: no 3,6(step 25)
30g. R2C8: no 3

31. 13(2) at R2C12 = {58}: {49} blocked by R3C1, {67} blocked by R2C46(R2C46 = {26}/[71] = {6|7..}) -->> {58} locked for R2 and N1
31a. 8 in C3 locked for N7
31b. Clean up: R8C12 = {27} -->> locked for R8 and N7
31c. Clean up: R78C5 = {19} -->> locked for C5 and N8
31d. R8C46 = [43](step 14); R9C5 = 8(hidden); R1C5 = 3(step 25)
31e. R1C4 = 9(hidden); R5C4 = 3(hidden)
31f. Clean up: R2C5: no 7
31g. R2C4 = 7(hidden); R4C34 = [76]; R9C46 = [26]; R5C5 = 7(hidden)

32. 20(4) at R2C7 = 4{169}(last remaining combo) R2C7 + R3C78 = {169} -->> locked for N3
32a. R2C89 = [23]; R12C6 = [21]

33. 14(3) at R9C7 = 7[34](last remaining combo) -->> R9C89 = [34]
33a. R1C789 = [485]

34. R1C123 = {167} -->> locked for N1

35. 17(4) at R2C3 = {2348}(last remaining combo): no 9
35a. R2C3 = 4; R23C5 = [64]; R2C7 = 9; R34C1 = [93]; R4C2 = 8

36. R67C1 = [56](last remaining combo)

37. R8C89 = [59](last remaining combo)

And the rest is all naked singles.

greetings

Para
Last edited by Para on Sat Aug 04, 2007 10:14 pm, edited 1 time in total.
Andrew
Grandmaster
Grandmaster
Posts: 300
Joined: Fri Aug 11, 2006 4:48 am
Location: Lethbridge, Alberta

Post by Andrew »

When I saw that Para had posted a walkthrough for V2.5 I looked through the thread for that puzzle and eventually found it. No diagram, only a string of code which is meaningless to me.

I'm not suggesting that a diagram should be posted now for V2.5, just that in future when people post puzzles on this forum they post a diagram, or ask if someone can post one, not just post a string of code. I may be the only one of the regular message posters that doesn't use any killer software but there are probably others who read this forum that don't use such software. Also people who do use that software are more likely to be attracted to a puzzle if they see a diagram rather than just a string of code.
Para
Yokozuna
Yokozuna
Posts: 384
Joined: Wed Nov 08, 2006 7:42 pm
Location: The Netherlands

Post by Para »

I had the same when looking back to make a walk-through. I remembered that this puzzle had a lot of variants, so completely overlooked that one for a while, but finally found it.

So added pic to walk-through.

[edit] Now it looks kinda strange, as if i posted a puzzle with walk-through :?
Andrew wrote:I may be the only one of the regular message posters that doesn't use any killer software but there are probably others who read this forum that don't use such software.
I do solve a lot of killers and other puzzles without software. Just don't try to tackle the assassins without software.
I actually rather solve puzzles on paper than on the computer. Find it more relaxing that way. I have a very comfortable chair to work in with my laptop though. I sometimes find it also more satisfying to do a puzzle on paper/without PM. Because software tends to make things easier.

Para
Post Reply