SudoCue Users Forum Index SudoCue Users
A forum for users of the SudoCue programs and the services of SudoCue.Net
 
 FAQFAQ   SearchSearch   UsergroupsUsergroups   RegisterRegister 
 ProfileProfile   Log in to check your private messagesLog in to check your private messages   Log inLog in 

(unofficial) Assassin 95

 
Post new topic   Reply to topic    SudoCue Users Forum Index -> Weekly Assassins
View previous topic :: View next topic  
Author Message
sudokuEd
Grandmaster
Grandmaster


Joined: 19 Jun 2006
Posts: 257
Location: Sydney Australia

PostPosted: Thu Mar 20, 2008 10:36 pm    Post subject: (unofficial) Assassin 95 Reply with quote

Nice tradition Mike has started with unofficial Assassins (uA) Very Happy .

I really enjoyed getting this new one ready. This puzzle has a ridiculously low SS(v3)score of 1.05. But it took about 6 sessions to finally find the 2 key moves that SudokuSolver must find easily to give it such a low score![edit: I don't know how SS actually does this one - will check at the end of the uA week] Indeed, they are not hard moves, just very-hard-for-me to find. So, I give it a rating of 1.25.

Those using SudokuSolver v3 can import the background graphic via the url shown. BTW - the Good Friday parallels are purely co-incidental.


(unofficial) Assassin 95 (uA95)





.
Code:
-----.--.--.-----.--------.
|14   |18|19|11   |18      |
|  .--'  |  |  .--'-----.--:
|  |     |  |  |18      |6 |
:--'--.--:  :--'-----.--:  |
|17   |7 |  |19      |14|  |
:--.  |  |  :-----.--'  :--:
|28|  |  |  |8    |     |18|
|  '--+--'--'-----'--.--'  |
|     |20            |     |
|  .--'--.-----.--.--+--.  |
|  |8    |11   |22|13|19|  |
:--:  .--'-----:  |  |  '--:
|9 |  |16      |  |  |     |
|  :--'-----.--:  :--'--.--:
|  |16      |20|  |12   |14|
:--'-----.--'  |  |  .--'  |
|10      |     |  |  |     |
'--------'-----'--'--'-----'


Text code:
3x3::k:3585:3585:4610:4867:2820:2820:4613:4613:4613:3585:4610:4610:4867:2820:4614:4614:4614:1543:4360:4360:1801:4867:4874:4874:4874:3595:1543:7180:4360:1801:4867:2061:2061:3595:3595:4622:7180:7180:5135:5135:5135:5135:5135:4622:4622:7180:2064:2064:2833:2833:5650:3347:4884:4622:2325:2064:4118:4118:4118:5650:3347:4884:4884:2325:4119:4119:4119:5144:5650:3097:3097:3610:2587:2587:2587:5144:5144:5650:3097:3610:3610:

Cheers
Ed
Back to top
View user's profile Send private message
Afmob
Expert
Expert


Joined: 22 Sep 2007
Posts: 103
Location: MV, Germany

PostPosted: Fri Mar 21, 2008 3:30 pm    Post subject: Reply with quote

I agree with Ed that this Killer should be rated above 1.0. It took me some time to find step 5a (though the step itself is not difficult) and after that the puzzle was pretty much solved. Another interesting move was step 4a which helped me to shorten the end game a bit.

By the way, I finally got the status "Addict" which more fits I think Wink.

UA 95 Walkthrough:

1. C123
a) 28(4) = 89{47/56} -> 8,9 locked for N4
b) Outies N1 = 11(2) = [56/65/74]
c) 7(2): R3C3 <> 4,5,6
d) Hidden Triple (123) in R5C3+R6C23 for N4 -> no other candidates
e) 8(3) must have 4 xor 5 and it's only possible @ R7C2 -> R7C2 = (45)
f) 8(3) = 1{25/34} -> 1 locked for R6+N4
g) Naked triple (123) locked in R356C3 for C3
h) 9(2) <> 4,5 because {45} blocked by R7C2 = (45)
i) 10(3) <> 4 because {145} blocked by R7C2 = (45)
j) 10(3) must have 5,6 xor 7 and R9C3 = (567) -> R9C12 <> 5,6,7

2. R6789
a) Innies+Outies N7: -10 = R8C4 - R7C23
-> R8C4 = (1234) because R7C23 <= 14
-> R7C3 <> 4,5 because R7C23 must at least be 11
b) Outies N9 = 13(2) <> 2,3
c) Innies = 10(2) = [46/64/73/82]
d) Using Outies N9 = 13(2): R6C236 = 11(3) = 1{28/37} -> R6C6 = (78)
e) 11(2) <> {38} since it's a Killer pair of R6C236
f) Outies R89 = 13(2+1): R7C16 <> 6,7,8,9 because R6C6 >= 7
g) Naked triple (123) locked in R7C1+R9C12 for N7

3. R1234
a) Innies = 11(2): R4C9 <> 1,8,9
b) Using Outies N1 = 11(2): R4C478 = 15(3): R4C4 <> 1 because R4C78 @ 14(3) <= 13
c) R4C478 = 15(3) <> 7 because 7{26/35} blocked by Killer pairs (57,67) of Outies N1
d) 17(3): R3C12 <> 5 because there is no combo with {56} or {57}

4. R789 !
a) ! Innies N9 = 19(3) must have 2 of (6789), only other place possible in R7
is 16(3) -> 16(3) must have 2 of (6789)
b) 16(3) = {169/178/268/367} <> 4,5
c) Innies N9 = 19(3) <> 5 because {568} blocked by Killer pair (68) of 16(3) @ R7
d) 13(2): R6C7 <> 8
e) Outies N9 = 13(2): R6C8 <> 5

5. R456 !
a) ! Hidden Killer pair (89) in R4C1 for R4 since R4C478 can only have one of (89) -> R4C1 = (89)
b) Innies R1234 = 11(2) = [83/92]
c) Innies R6789 = 10(2) <> 2,8 since [82] is a Killer pair of Innies R1234
d) 11(2) <> {47} because it's a Killer pair of Innies R6789
e) 8(2) <> {26} since it's a Killer pair of 11(2)
f) Outies N9 = 13(2) <> {67} since it's a Killer pair of Innies R6789
g) 13(2) <> 6,7
h) 7 locked in R5C789 for R5
i) 28(4): R6C1 <> 4 because 7 only possible there

6. R456 !
a) Innies R6789 = 10(2): R6C9 <> 6
b) 4 locked in R6C789 for N6
c) 20(5) = 12{359/368/467} since other combos blocked by Killer pair (45) of 28(4)
-> 1,2 locked for R5
d) ! 18(4) <> {3456} because it's blocked by Killer pair (56) of 20(5)
e) 18(4) = 2{349/358/367/457} -> R4C9 = 2
f) Innies R1234 = [92] -> R4C1 = 9
g) 28(4) = 89{47/56} -> 8 locked for R5
h) 18(4) = 23{49/67} since {2457} blocked by Killer pair (45) of Outies N9
-> 3 locked for N6
i) R4C478 = 15(3) <> 5 because {456} blocked by Killer pair (46) of Outies N1

7. C789
a) 6(2) = {15} locked for C9+N3
b) 14(3) @ N3 = {167} since R4C78 = (168) and R3C8 <> 5 -> R3C8 = 7, {16} locked for R4+N6
c) 18(4) = {2349} -> R6C9 = 4, {39} locked for R5+N6
d) 13(2) = [58] -> R6C7 = 5, R7C7 = 8
e) 18(3) @ R1C7 = 6{39/48} -> 6 locked for R1+N3
f) R6C8 = 8 -> 19(3) = 8[29/47]

8. R456
a) 11(2) = {29} locked for R6+N5
b) 8(3) = {134} -> R7C2 = 4
c) Outies N1 = 11(2) = [74] -> R4C2 = 7, R4C3 = 4 -> R3C3 = 3
d) R6C1 = 6

9. R789
a) 19(3) = {289} -> R7C8 = 2, R7C9 = 9
b) 16(3) @ R7 = {367} -> 3 locked for R7+N8
c) 9(2) = [18] -> R7C1 = 1, R8C1 = 8
d) Hidden Single: R9C1 = 3 @ C1 -> R9C3 = 5, R9C2 = 2
e) 16(3) @ R8 = {169} -> R8C4 = 1, {69} locked for R8+N7

10. R123
a) 17(3) = 7[28/46] -> R3C2 = (68)
b) Hidden Single: R4C4 = 8 @ R4 -> 19(4) = 28{36/45} -> 2 locked for C4+N2
c) Killer pair (46) of 19(4) blocks {146} of 11(3)
d) 11(3) = {137} locked for N2
e) 19(4) = {2458} -> {45} locked for C4+N2
f) 18(3) @ R2C6 = 6{39/48} -> R2C6 = 6
g) 18(3) @ R2C6 = {369} -> {39} locked for R2+N3

11. N89
a) Hidden Single: R8C8 = 5 @ N9 -> 12(3) = {345} -> R8C7 = 3, R9C7 = 4
b) 20(3) = {479} -> R8C5 = 4, {79} locked for R9+N8

12. Rest is singles.

Rating: (Easy) 1.25.


Last edited by Afmob on Mon Mar 24, 2008 8:21 am; edited 1 time in total
Back to top
View user's profile Send private message
gary w
Addict
Addict


Joined: 07 Sep 2007
Posts: 84
Location: south wales

PostPosted: Sun Mar 23, 2008 12:10 am    Post subject: Reply with quote

I'ld rate this as 1.25 at least..or maybe I'm out of practice!?
The innies of n14-> r5c3=2/3,r6c23=1 2/3 so r7c2=4/5,hence r6c6=7/8.With r7c126=10 combo work in n7 places 4 in either r7c2 or r8c23.It is then fairly straightforward to eliminate 3 from r4c3 as this would place a 3 at both r4c4 and r4c8.

Regards

Gary
Back to top
View user's profile Send private message
Andrew
Grandmaster
Grandmaster


Joined: 11 Aug 2006
Posts: 300
Location: Lethbridge, Alberta

PostPosted: Mon Mar 24, 2008 11:14 pm    Post subject: Reply with quote

Thanks for a fun puzzle Ed! Very Happy

Afmob wrote:
By the way, I finally got the status "Addict" which more fits I think Wink.

Congratulations! It won't take you too long to reach Expert and Master, which are also appropriate!

Afmob's steps 2d and 3b are the sort of steps that are so easy not to spot; I missed them. Using an innie pair and an outie pair to do a 45 isn't a particularly common step. Steps 4a and 5c were also interesting. With those steps, together with Afmob saying that step 5a was difficult to spot, I felt that it should have been rated 1.25 rather than (Easy) 1.25.

Gary's R7C126 = 10 is useful although both Afmob and I got the same cage eliminations using different steps; however the only possible combinations in R7C126 do lock 5 for R7 very quickly.

I used one, fairly easy to spot, contradiction move so I'll rate this puzzle 1.25.

Ed commented that SS had two key moves that gave this puzzle a low rating. I'll be interested to see what they were. I don't think I found either of them.

Here is my walkthrough

Prelims

a) R23C9 = {15/24}
b) R34C3 = {16/25/34}, no 7,8,9
c) R4C56 = {17/26/35}, no 4,8,9
d) R6C45 = {29/38/47/56}, no 1
e) R67C7 = {49/58/67}, no 1,2,3
f) R78C1 = {18/27/36/45}, no 9
g) 11(3) cage in N2 = {128/137/146/236/245}, no 9
h) R3C567 = {289/379/469/478/568}, no 1
i) 8(3) cage at R6C2 = 1{25/34}, CPE no 1 in R45C2
j) 19(3) cage at R6C8 = {289/379/469/478/568}, no 1
k) 20(3) cage in N8 = {389/479/569/578}, no 1,2
l) R9C123 = {127/136/145/235}, no 8,9
m) 28(4) cage in N4 = {4789/5689}, 8,9 locked for N4

1. 45 rule on N1 2 outies R4C23 = 11 = {56}/[74], clean-up: no 4,5,6 in R3C3
1a. 17(3) cage at R3C1 = {179/269/278/359/368/458/467}
1b. 5 of {359/458} must be in R4C2 -> no 5 in R3C12

2. Killer pair 4,5 in R4C23 and 28(5) cage, locked for N4
2a. Killer pair 6,7 in R4C23 and 28(5) cage, locked for N4
[Alternatively 45 rule on N14 3 innies R5C3 + R6C23 = 6 = {123} or another alternative hidden triple {123} in R5C3 + R6C23. Take your pick! Wink ]

3. 8(3) cage at R6C2 = 1{25/34}
3a. 4,5 only in R7C2 -> R7C2 = {45}
3b. 1 locked in R6C23, locked for R6 and N4
3c. R78C1 = {18/27/36} (cannot be {45} which clashes with R7C2), no 4,5
[I should also have spotted that R7C2 = {45} eliminates 4 from R9C123]

4. Naked triple {123} in R356C3, locked for C3

5. 45 rule on N9 2 outies R6C78 = 13 = {49/58/67}, no 2,3

6. 45 rule on N7 2 innies R7C23 = 1 outie R8C4 + 10, max R7C23 = 14 -> max R8C4 = 4
6a. Max R8C34 = 13 -> min R8C2 = 3

7. 45 rule on R1234 2 innies R4C19 = 11 = {47/56}/[83/92], no 1,8,9 in R4C9

8. 45 rule on R6789 2 innies R6C19 = 10 = {46}/[73/82], no 5,9, no 7,8 in R6C9

9. 45 rule on N23 1 outie R4C4 = 1 innie R3C8 + 1, no 1 in R4C4, no 9 in R3C8

10. 45 rule on N78 1 outie R6C6 = 1 innie R7C2 + 3 -> R6C6 = {78}

11. Hidden killer pair 5,9 in R6C45 and R6C78. Neither can have both 5 and 9 -> R6C45 = {29/56}, R6C78 = {49/58}, clean-up: no 6,7 in R7C7
11a. R4C56 = {17/35} (cannot be {26} which clashes with R6C45), no 2,6

12. 45 rule on R89 2 outies R67C6 = 1 innie R8C1 + 4
12a. Min R67C6 = 8 -> no 1,2,3 in R8C1, clean-up: no 6,7,8 in R7C1
12b. Min R8C1 = 6 -> min R67C6 = 10, no 1 in R7C6
12c. Max R8C1 = 8 -> max R67C6 = 12, no 6,7,8,9 in R7C6

13. 45 rule on N69 1 outie R3C8 = 1 innie R5C7, no 9 in R5C7

14. 45 rule on C123 2 innies R57C3 = 1 outie R8C4 + 8, min R57C3 = 9 -> min R7C3 = 6

15. 45 rule on N5 5 innies R4C4 + R5C456 + R6C6 = 26 and must contain 4 = {14678/23489/34568} (cannot be {14579/24569/24578} which clash with R6C45)

16. Killer pair 5,7 in R4C23 and R4C56, locked for R4, clean-up: no 4,6 in R3C8 (step 9), no 4,6 in R4C19 (step 7), no 4,6 in R5C7 (step 13)

17. 7 in N6 locked in R5C789, locked for R5

18. R6C6 = R7C2 + 3 (step 10)
18a. 45 rule on N14 R7C2 = R5C3 + 2
18b. -> R6C6 = R5C3 + 5

19. R4C4 + R5C456 + R6C6 (step 15) = {14678/23489/34568}
19a. 8 of {23489} must be in R6C6 => R5C3 => 3 (step 18b) => R4C4 = 3 => R3C8 = 2 (step 9) => R5C7 = 2 (step 13) -> R4C4 + R5C456 + R6C6 cannot be {23489}
19b. R4C4 + R5C456 + R6C6 = {14678/34568}, no 2,9, 6 locked for N5, clean-up: no 1,8 in R3C8 (step 9), no 1,8 in R5C7 (step 13), no 5 in R6C45

20. Naked pair {29} in R6C45, locked for R6, clean-up: no 8 in R6C1 (step 8), no 4 in R6C78, no 4,9 in R7C7
20a. Naked pair {58} in R6C78, locked for R6 and N6 -> R6C6 = 7, R7C2 = 4 (step 10), clean-up: no 5 in R3C8 (step 13), no 6 in R4C4 (step 9), no 1 in R4C56, no 3 in R6C9 (step 8)

21. R4C2 = 7 (hidden single in R4), R4C3 = 4 (step 1), R3C3 = 3, R5C3 = 2, R6C23 = [31], R6C19 = [64]; clean-up: no 2 in R23C9, no 3 in R7C1

22. Naked pair {35} in R4C56, locked for R4 and N5 -> R4C4 = 8, R3C8 = 7 (step 9), R5C7 = 7 (step 13), R4C19 = [92]
22a. Naked pair {16} in R4C78, locked for N6

23. R9C1 = 3 (hidden single in C1), R9C23 = 7 = [16/25]

24. Naked pair {15} in R23C9, locked for C9 and N3

25. R4C4 = 8 -> R123C4 = 11
25a. 45 rule on N2 3 innies R2C6 + R3C56 = 23 = {689}, locked for N2

26. R3C567 = {289/469} -> R3C7 = {24}, 9 locked in R3C56, locked for R3 and N2

27. R4C2 = 7 -> R3C12 = 10 = {28}/[46], no 1
27a. Killer pair 2,4 in R3C12 and R3C7, locked for R3

28. R123C4 = 11 (step 25) = {137/245}
28a. R3C4 = {15} -> no 1,5 in R12C4

29. R2C678 = {369/468}, no 2, 6 locked for R2

30. R1C789 = {369/468}, no 2, 6 locked for R1 and N3 -> R2C6 = 6 (step 29)

31. R3C7 = 2 (hidden single in N3), clean-up: no 8 in R3C12 (step 27) -> R3C12 = [46]

32. 18(3) cage in N1 = {189/279}, no 5, 9 locked for N1
32a. 1,2 only in R2C2 -> R2C2 = {12}
32b. 9 locked in R12C3, locked for C3

33. Naked pair {12} in R29C2, locked for C2
33a. Naked pair {58} in R15C2, locked for C2 -> R8C2 = 9
33b. R8C34 = 7 = [52/61]
33c. Naked pair {56} in R89C3, locked for C3

34. R7C345 = {178/268/358/367} (cannot be {169/259} because R7C3 only contains 7,8), no 9

35. 9 in R7 locked in R7C89, locked for N9
35a. R6C8 = [829] (only remaining permutation), R5C89 = [93], R67C7 = [58], R7C3 = 7, R78C1 = [18], R5C12 = [58], R9C23 = [25] (step 23), R12C2 = [51], R23C9 = [51], R3C4 = 5, R8C34 = [61] (step 33b)
35b. R3C4 = 5 -> R12C4 = {24} (step 28), locked for C4 and N2 -> R5C4 = 6, R6C45 = [92], R7C4 = 3, R7C56 = [65], R9C4 = 7, R8C56 = [42], R5C56 = [14], R4C56 = [53], R1C6 = 1, R8C78 = [35], R89C9 = [76], R1C9 = 8, R12C3 = [98]

36. R1C789 (step 30) = {468}, 4,6 locked for R1 and N3

and the rest is naked singles and a couple of cage sums

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


Last edited by Andrew on Thu Mar 27, 2008 9:51 pm; edited 2 times in total
Back to top
View user's profile Send private message
mhparker
Grandmaster
Grandmaster


Joined: 20 Jan 2007
Posts: 345
Location: Germany

PostPosted: Tue Mar 25, 2008 12:40 pm    Post subject: Reply with quote

Andrew wrote:
Thanks for a fun puzzle Ed! Very Happy

Indeed. Thanks, Ed!

Andrew wrote:
I used one, fairly easy to spot, contradiction move so I'll rate this puzzle 1.25.

I took a very similar route to Andrew up to and including his step 17, but managed to avoid a contradiction move as follows:

Grid state after Andrew's step 17:

Code:
.-----------------------.-----------.-----------.-----------------------.-----------------------------------.
| 123456789   123456789 | 456789    | 123456789 | 12345678    12345678  | 123456789   123456789   123456789 |
|           .-----------'           |           |           .-----------'-----------------------.-----------:
| 123456789 | 123456789   456789    | 123456789 | 12345678  | 123456789   123456789   123456789 | 1245      |
:-----------'-----------.-----------:           :-----------'-----------------------.-----------:           |
| 12346789    12346789  | 123       | 123456789 | 23456789    23456789    23456789  | 123578    | 1245      |
:-----------.           |           |           :-----------------------.-----------'           :-----------:
| 89        | 567       | 456       | 234689    | 1357        1357      | 1234689     1234689   | 23        |
|           '-----------+-----------'-----------'-----------------------'-----------.-----------'           |
| 45689       45689     | 23          12345689    12345689    12345689    123578    | 123456789   123456789 |
|           .-----------'-----------.-----------------------.-----------.-----------+-----------.           |
| 4678      | 123         123       | 2569        2569      | 78        | 4589      | 4589      | 2346      |
:-----------:           .-----------'-----------------------:           |           |           '-----------:
| 123       | 45        | 6789        123456789   123456789 | 2345      | 4589      | 23456789    23456789  |
|           :-----------'-----------------------.-----------:           :-----------'-----------.-----------:
| 678       | 3456789     456789      1234      | 3456789   | 123456789 | 123456789   123456789 | 123456789 |
:-----------'-----------------------.-----------'           |           |           .-----------'           |
| 1234567     1234567     4567      | 3456789     3456789   | 123456789 | 123456789 | 123456789   123456789 |
'-----------------------------------'-----------------------'-----------'-----------'-----------------------'


triple-click to see what I wrote:
18. 28(4) cage in N4 (prelim. m) = {4789/5689}
18a. 7 only available in R6C1
18b. -> no 4 in R6C1
18c. cleanup: no 6 in R6C9 (step 8)

19. 6(2) at R23C9 = {15}, locked for C9 and N3
(Note: {24} combo blocked by R46C9)
19a. no 1,5 in R5C7 (step 13)

20. 18(4) at R4C9: max. R46C9 = 7 -> min. R5C89 = 11
20a. -> no 1 in R5C8

21. 1 in N6 locked in 14(3) at R3C8 = {167} (last combo)
21a. -> R3C8 = 7; R4C78 = {16}, locked for R4 and N6
...

Note: I decided to make this post because step 19 is another good example of the use of a 2-cell ALS blocker, as discussed in the YAK94 thread.


P.S. Despite avoiding a contradiction move, I would still go along with Andrew's rating of a typical 1.25, rather than an "easy 1.25".
_________________
Cheers,
Mike


Last edited by mhparker on Wed Mar 26, 2008 9:00 am; edited 1 time in total
Back to top
View user's profile Send private message
Andrew
Grandmaster
Grandmaster


Joined: 11 Aug 2006
Posts: 300
Location: Lethbridge, Alberta

PostPosted: Wed Mar 26, 2008 3:47 am    Post subject: Reply with quote

Thanks Mike! Very Happy

That was a neat move. Definitely worth one of Afmob's !s as it broke this puzzle open. I must learn to spot that sort of move. So easy to understand but not so easy to spot.

I had a feeling that there ought to be something better than my contradiction move but I guess I was still feeling frustrated by not being able to find the breakthrough in uA94. Then after posting my uA95 walkthrough I had another look at uA94 and managed to find the breakthrough. Maybe it was helped by the large glass of Talisker single malt whisky that I'd drunk while watching TV earlier in the evening! Wink
Back to top
View user's profile Send private message
Afmob
Expert
Expert


Joined: 22 Sep 2007
Posts: 103
Location: MV, Germany

PostPosted: Wed Mar 26, 2008 6:23 am    Post subject: Reply with quote

You are probably right about the rating of my wt. It's more a 1.25 than an easy 1.25. But I think if we combine Andrew's step 11 and 16 (which I didn't spot and they would have shortened my solution) with my step 5c we can come to a short and easy walkthrough.

I just feel that uA 95 belongs more to the lower end of the 1.25 range because compared to uA94 which is also of rating 1.25 (in my opinion), it is a bit easier.
Back to top
View user's profile Send private message
sudokuEd
Grandmaster
Grandmaster


Joined: 19 Jun 2006
Posts: 257
Location: Sydney Australia

PostPosted: Wed Mar 26, 2008 9:53 am    Post subject: Reply with quote

Afmob wrote:
if we combine Andrew's step 11 and 16..my step 5c we can come to a short and easy walkthrough
Afmob's step 5c was the key move for me and hence the double-headed axe gif . Fun one!

I felt a bit of disquiet about the number of elements (5-cell innies, several "45"s and back-door combinations) of Andrew's contradiction move. More of a 2.0+ rating move in my mind. So, am glad there are other ways to solve this puzzle. [BTW- love Mike's ALS(+1) block Cool ]

Now, two more ways from Andrew's step 17 marks (see Mike's post). First, adapted from the way that SudokuSolver v3 (scoring routine order) makes it's big breakthrough (edit: incorrect combo's in 11(2)n5 taken out: thanks Andrew) and which leads to the low SS(v3)score of 1.05.

Code:
18. "45" n69: 3 innies r4c78 + r5c7 = h14(3) = {149/158/167/248/257/347/356} ({239} blocked by r4c9)
18a. -> no 3 in r5c7 (missing 5 or 7 in r4c78)
18b. -> no 3 in r3c8 (i/o n69)
18c. -> no 4 in r4c4 (i/o n23)

19. 4 in n5 only in r5: 4 locked for r5

20. 28(4)n4 = 89{47/56}
20a. 4 in {4789} must be in r6c1 -> no 7 in r6c1
20b. -> no 3 in r6c9 (h10(2)r6c19)

21. hidden pair {13} in r6c23 -> r7c2 = 4, r5c3 = 2
21a. &  7 in r6c6 (i/0 n1478)


Second, another move I stumbled on playing with SudokuSolver which it called by another name, but looks like Y-wing to me.

Code:
18. 3s in r5c3 & r4c9 'see' all 3's in r6 -> they cannot both be 3
18b. -> no 2 in common peers in r5c789 since this would force both r5c3 & r4c9 = 3
18c. -> no 2 in in r3c8(i/o n69)
18d. -> no 3 in r4c4 (i/o n23)

Now Andrew's (first part of) step 19
19. R4C4 + R5C456 + R6C6 (step 15) = {14678/23489/34568}
19a. 8 of {23489} must be in R6C6 => R5C3 => 3 (step 18b)
19b. but this clashes with 3 in r5 of innies n5
19c. back to Andrew's 19b


Cheers
Ed


Last edited by sudokuEd on Thu Mar 27, 2008 11:06 am; edited 1 time in total
Back to top
View user's profile Send private message
gary w
Addict
Addict


Joined: 07 Sep 2007
Posts: 84
Location: south wales

PostPosted: Thu Mar 27, 2008 12:31 am    Post subject: Reply with quote

Thanks for pointing out the typo Andrew..I should have posted...can eliminate 3 from r5c3 as this places a 3 at both r4c4 and r8c4.

Regards

Gary
Back to top
View user's profile Send private message
Andrew
Grandmaster
Grandmaster


Joined: 11 Aug 2006
Posts: 300
Location: Lethbridge, Alberta

PostPosted: Thu Mar 27, 2008 1:31 am    Post subject: Reply with quote

Thanks Afmob for mentioning my steps 11 and 16. I was pleased when I found step 11, which I found the most enjoyable step for this puzzle. As you say combining those two steps with your step 5c would make a short walkthrough.

Ed's SS moves are interesting. The first one is straightforward once one has spotted the innies for N69. Whether that then gives a 1.05 rated solution depends on whether that starting position can be reached without my steps 11 and 16. With them I'm inclined to call it an Easy 1.25 solution.

The second SS solution is a tricky one to spot. Definitely a neat move but not part of a 1.05 solution, probably not even part of a 1.25 solution.

Ed wrote:
I felt a bit of disquiet about the number of elements (5-cell innies, several "45"s and back-door combinations) of Andrew's contradiction move. More of a 2.0+ rating move in my mind.

I'll interpret that as meaning that this type of move is best kept for a 2.0+ rated puzzle. As I said in an earlier message, I was feeling frustrated by uA94. Otherwise I would have spent longer looking for a better step. Of course it's definitely not a 2.0+ level step; it's a very short contradiction. It just looks a bit more complicated than it really is because of the way I linked round R4C4 -> R3C8 -> R5C7; if I'd set these up first as R4C4 = R5C7 + 1 it would have been simpler.
Back to top
View user's profile Send private message
Display posts from previous:   
Post new topic   Reply to topic    SudoCue Users Forum Index -> Weekly Assassins All times are GMT
Page 1 of 1

 
Jump to:  
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 vote in polls in this forum


Powered by phpBB © 2001, 2005 phpBB Group