Assassin 85 (Original Version)

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

Assassin 85 (Original Version)

Post by Andrew »

I thought it would be a good idea to set up a separate thread for this puzzle, which Ruud has now posted in the Unsolvables.
CathyW wrote:I have 3 placements and am totally stuck.
That's how far I've managed to get.

I'm posting what I've done so far in the hope that someone will pick it up as a "tag" solving attempt. I get the impression from the posts by Afmob and Gary on the A85 thread that they have both got a bit further than I have.

Here are my opening steps. I'm using normal text because it's a separate thread and it's best for "tags".

Prelims

a) R1C34 = {79}, locked for R1
b) R12C5 = {14/23}
c) R1C67 = {38/56} (other combinations eliminated by step a)
d) R34C1 = {19/28/37/46}, no 5
e) R34C9 = {49/58/67}
f) R5C89 = {14/23}
g) R5C12 = {16/25} (cannot be {34} which clashes with R5C89)
h) R67C1 = {19/28/37/46}, no 5
i) R67C9 = {39/48/57}, no 1,2,6
j) R89C5 = {49/58/67}, no 1,2,3
k) R9C34 = {18/27/36/45}, no 9
l) R9C67 = {18/27/36/45}, no 9
m) 21(3) cage in N1 = {489/579/678}, no 1,2,3
n) 10(3) cage in N3 = {127/136/145/235}, no 8,9
o) 10(3) cage at R6C7 = {127/136/145/235}, no 8,9
p) 18(5) cage at R2C2 = {12348/12357/12456}, no 9

1. Killer pair 1,2 in R5C12 and R5C89, locked for R5

2. 21(3) cage in N1 = {489/678} (cannot be {579} because 7,9 only in R2C1), no 5, 8 locked for N1, clean-up: no 2 in R4C1
2a. 7,9 only in R2C1 -> R2C1 = {79}
2b. 8 locked in R1C12, locked for R1, clean-up: no 3 in R1C67

3. Naked pair {56} in R1C67, locked for R1, clean-up: no 7 in R2C1 (step 2) -> R2C1 = 9, R1C34 = [79], clean-up: no 1 in R34C1, no 3 in R4C1, no 1 in R67C1, no 2 in R9C4
3a. Naked pair {48} in R1C12, locked for R1 and N1, clean-up: no 1 in R2C5, no 6 in R4C1

4. 10(3) cage in N3 = {127/136/235} (cannot be {145} because 4,5 only in R2C9), no 4
4a. 5,6,7 only in R2C9 -> R2C9 = {567}

5. 45 rule on C1 3 outies R159C2 = 18 = {189/459/468} (cannot be {279/369/567} because R1C2 only contains 4,8, cannot be {378} because no 3,7,8 in R5C2), no 2,3,7, clean-up: no 5 in R5C1
5a. 1 of {189} must be in R5C2 -> no 1 in R9C2
5b. 5 of {459} must be in R5C2 -> no 5 in R9C2
5c. 6 of {468} must be in R5C2 -> no 6 in R9C2

6. 45 rule on C9 3 outies R159C8 = 12 = {129/138/147/237/246/345} (cannot be {156} because 5,6 only in R9C8)
6a. 5,6,7,8,9 only in R9C8 -> R9C8 = {56789}

7. 45 rule on C6789 4 innies R456C6 + R5C7 = 29
7a. Max R456C6 = 24 -> min R5C7 = 5
7b. Max R5C7 = 9 -> min R456C6 = 20, no 1,2

8. Killer triple 4,7,8 in R1C1, R4C1 and R67C1, locked for C1

9. 5 in C1 locked in R89C1, locked for N7, clean-up: no 4 in R9C4
9a. 15(3) cage in N7 = 5{19/28/46}, no 3

10. 45 rule on R9 3 outies R8C159 = 18 = {189/279/369/459/468/567} (cannot be {378} because no 3,7,8 in R8C1)
10a. 1,2 of {189/279} must be in R8C1 -> no 1,2 in R8C9

I've looked at the 5-cell cages and the outies from R12, R89, C12 and C89 but so far haven't found anything useful from them.

Let's see how much further we can get before we need to resort to chains or hypotheticals. From the ratings that Afmob posted, I expect we will need to use them later in this puzzle.
Last edited by Andrew on Tue Jan 15, 2008 12:57 am, edited 1 time in total.
mhparker
Grandmaster
Grandmaster
Posts: 345
Joined: Sat Jan 20, 2007 10:47 pm
Location: Germany

Re: Assassin 85 (Original Version)

Post by mhparker »

Andrew wrote:Let's see how much further we can get before we need to resort to chains or hypotheticals.
Probably not very far... ](*,)

In the meantime, here's another couple of steps:

11. 45 rule on C1234: 3 outies R5C567 = 2 innies R46C4 + 18
11a. Max. R5C567 = 24 -> max. R46C4 = 6
11b. -> no 6,7,8 in R46C4
11c. Min. R46C4 = 3 -> min. R5C567 = 21
11d. -> no 3 in R5C56

12. 45 rule on C9: 5 innies R12589C9 = 20 = {12359/12368/12458/12467}
(Note: {13457/23456} blocked because these innies must contain the {12} of C9)
12a. 5 of {12359/12458} must go in R2C9
12b. -> no 5 in R89C9

Grid state after step 12:

Code: Select all

.-----------------------.-----------------------.-----------.-----------------------.-----------------------.
| 48          48        | 7           9         | 123       | 56          56        | 123         123       |
|           .-----------'-----------------------&#58;           &#58;-----------------------'-----------.           |
| 9         | 12356       12356       12345678  | 234       | 12345678    12345678    12345678  | 567       |
&#58;-----------&#58;           .-----------------------+-----------+-----------------------.           &#58;-----------&#58;
| 236       | 12356     | 12356       12345678  | 12345678  | 12345678    123456789 | 123456789 | 456789    |
|           |           |           .-----------'           '-----------.           |           |           |
| 478       | 12345678  | 12345689  | 12345       123456789   3456789   | 123456789 | 123456789 | 456789    |
&#58;-----------'-----------+-----------'-----------------------------------'-----------+-----------'-----------&#58;
| 126         156       | 345689      345678      456789      456789      56789     | 1234        1234      |
&#58;-----------.-----------+-----------.-----------------------------------.-----------+-----------.-----------&#58;
| 234678    | 123456789 | 12345689  | 12345       123456789   3456789   | 1234567   | 123456789 | 345789    |
|           |           |           '-----------.           .-----------'           |           |           |
| 234678    | 12346789  | 1234689     12345678  | 123456789 | 1234567     1234567   | 123456789 | 345789    |
&#58;-----------&#58;           '-----------------------+-----------+-----------------------'           &#58;-----------&#58;
| 1256      | 12346789    1234689     12345678  | 456789    | 123456789   123456789   123456789 | 346789    |
|           '-----------.-----------------------&#58;           &#58;-----------------------.-----------'           |
| 1256        489       | 123468      135678    | 456789    | 12345678    12345678  | 56789       12346789  |
'-----------------------'-----------------------'-----------'-----------------------'-----------------------'
Cheers,
Mike
Afmob
Expert
Expert
Posts: 103
Joined: Sat Sep 22, 2007 5:36 pm
Location: MV, Germany

Post by Afmob »

13. 17(3) @ R6C3: R7C4 <> 1 because R67C3 <> 7

14. Innies R9 = 27(5) = 9{1278/1368/1458/2367/2457/3456} because 9 is locked there and other combos blocked by Killer triples (146,235) of combined 9(2) cages
14a. Innies R9 = 27(5): R9C9 <> 6 because 3 only possible there

15. R8C3 <> 9 because it sees all 9 of N4

16. Innies C6789 = 29(1+3): R5C7 <> 5 because R456C6 would be {789}
-> not possible since 33(5) must have at least 1 of (789) @ R5C45

I hope that step 16 is not regarded as too "contradiction chain"-like otherwise I remove it until we can't get any further without using chains.

By the way, I have a small contradiction chain that eliminates 3 from R8C9 (by using Outies of R9) which removes 9 from R9C9, otherwise I haven't got further than our tag walkthrough.
Last edited by Afmob on Wed Feb 06, 2008 8:29 am, edited 1 time in total.
mhparker
Grandmaster
Grandmaster
Posts: 345
Joined: Sat Jan 20, 2007 10:47 pm
Location: Germany

Post by mhparker »

Great work, Afmob! Every candidate counts...

17. 11(2) at R1C67 requires 1 of the 2 6s of C67
17a. ->h29(3+1) at R456C6+R5C7 (step 16) cannot contain both 6s of C67
17b. -> {689}[6] combination blocked
17c. -> no 6 in R5C7

New grid state after step 17:

Code: Select all

.-----------------------.-----------------------.-----------.-----------------------.-----------------------.
| 48          48        | 7           9         | 123       | 56          56        | 123         123       |
|           .-----------'-----------------------&#58;           &#58;-----------------------'-----------.           |
| 9         | 12356       12356       12345678  | 234       | 12345678    12345678    12345678  | 567       |
&#58;-----------&#58;           .-----------------------+-----------+-----------------------.           &#58;-----------&#58;
| 236       | 12356     | 12356       12345678  | 12345678  | 12345678    123456789 | 123456789 | 456789    |
|           |           |           .-----------'           '-----------.           |           |           |
| 478       | 12345678  | 12345689  | 12345       123456789   3456789   | 123456789 | 123456789 | 456789    |
&#58;-----------'-----------+-----------'-----------------------------------'-----------+-----------'-----------&#58;
| 126         156       | 345689      345678      456789      456789      789       | 1234        1234      |
&#58;-----------.-----------+-----------.-----------------------------------.-----------+-----------.-----------&#58;
| 234678    | 123456789 | 12345689  | 12345       123456789   3456789   | 1234567   | 123456789 | 345789    |
|           |           |           '-----------.           .-----------'           |           |           |
| 234678    | 12346789  | 1234689     2345678   | 123456789 | 1234567     1234567   | 123456789 | 345789    |
&#58;-----------&#58;           '-----------------------+-----------+-----------------------'           &#58;-----------&#58;
| 1256      | 12346789    123468      12345678  | 456789    | 123456789   123456789   123456789 | 346789    |
|           '-----------.-----------------------&#58;           &#58;-----------------------.-----------'           |
| 1256        489       | 123468      135678    | 456789    | 12345678    12345678  | 56789       1234789   |
'-----------------------'-----------------------'-----------'-----------------------'-----------------------'
At this rate (1 candidate eliminated per move), we should be finished in another 403 steps at the latest... :wink:
Cheers,
Mike
Nasenbaer
Master
Master
Posts: 167
Joined: Thu Jul 20, 2006 7:19 am
Location: Fellbach (Deutschland)

Post by Nasenbaer »

We're speeding up, I'm removing 2 :!: candidates in one move! :wink:

18. r3c9 = 6 -> r4c9 = 7 -> r2c9 = 5 -> can't be done because of r1c7
--> no 6 in r3c9 and no 7 in r4c9
Andrew
Grandmaster
Grandmaster
Posts: 300
Joined: Fri Aug 11, 2006 4:48 am
Location: Lethbridge, Alberta

Post by Andrew »

Good to see that we now have 3 new contributors to the "tag" that I started. I won't say my "tag" since they have an existence of their own once they have been started.

After steps 16 and 17 we appear to have reached what Gary mentioned on the A85 thread with R5C7 reduced to {789}.

While steps 16 and 18 are both contradiction chains, they are so short and localised that I, at least, consider them acceptable for what is clearly a very difficult puzzle.
Afmob
Expert
Expert
Posts: 103
Joined: Sat Sep 22, 2007 5:36 pm
Location: MV, Germany

Post by Afmob »

Time for some additional moves but this time I use chains.

19. Outies R9 = 18(3) <> 3 because
- i) Outies R9 = [693] -> 13(2) = [94] -> R9C5 = 4
- ii) Outies R9 = [693] -> 15(3) = [654] (step 9a) -> R9C5 <> 4
19a. 17(3) @ R9: R9C9 <> 9 because R8C9 + R9C8 >= 9

20. Innies+Outies N1: 11 = R2C4+R4C12 - R3C3
-> Max. R2C4+R4C12 = 17
20a. R2C4 @ 18(5) <> 8 since R4C2 @ 18(5) would be 4 (18(5) = {12348} if R2C4 = 8)
-> R2C4+R4C12 >= 8 + 4 + 7 (R4C1) >= 19
20b. 8 locked in 24(5) @ R2 -> R34C8 <> 8
20c. 24(5) = 8{1249/1267/1357/1456/2347/2356}
Last edited by Afmob on Wed Jan 16, 2008 6:02 am, edited 1 time in total.
Nasenbaer
Master
Master
Posts: 167
Joined: Thu Jul 20, 2006 7:19 am
Location: Fellbach (Deutschland)

Post by Nasenbaer »

Your step 19 gave me the idea to think about the 3 in r9c9. I found a way to remove it, but it involves chains so I don't know if we want to do it. I write it down as step 21 and you can decide if you want to keep it. But if we don't keep it the following steps would have to be discarded too.

21. r9c9 <> 3
if r9c9 = 3 then 9 is in r34c9 or r8c9 for c9 (can't be in r67c9 because {39} is blocked by 3 in r9c9)
a) r34c9 = {49} -> r67c9 = {57} -> r2c9 = 6 -> r1c89 = [31] -> r5c89 = [32] -> problem, two 3's in c8
b) r8c9 = 9 -> r34c9 = [76] ({58} blocked by 12(2)} -> r2c9 = 5 -> r1c89 = [32] -> r5c89 = [41] -> r5c12 = [25] -> r3c3 = 5 but can't place {157|256}
--> no 3 in r9c9 -> {359} {368} removed from 17(3)

22. r9: {36} locked in 9(2) @ r9c3 or 9(2) @ r9c6 -> no 6 in r9c158
22a. -> no 7 in r8c5
22b. -> cleanup for h18(3) @ r8c159: no 6 in r8c9
22c. -> cleanup for 17(3) @ r8c9: {269} {467} removed and no 7 in r9c9

23. c9: r2c9 <> 7
-> 6 is in r2c9 or in r4c9 which would set r3c9 = 7 --> r2c9 <> 7
23a. -> {56} locked in r1c7 and r2c9 for n3
23b. -> 10(3) @ r1c8 = 3{16|25} -> 3 locked in r1c89 for r1 and n3
23c. -> no 2 in r2c5
23d. -> no 8 in r4c9
23e. -> {23568} removed from 24(5) @ r2c6 (can't be placed)
Nasenbaer
Master
Master
Posts: 167
Joined: Thu Jul 20, 2006 7:19 am
Location: Fellbach (Deutschland)

Post by Nasenbaer »

Here's one more chain.

24. 17(3) @ r8c9 <> {458}
-- r89c9 = [48], r9c8 = 5 -> r234c9 = [576] -> r1c89 = [32] -> r12c5 = [14] -> r89c5 = [67] -> only r9c2 left for 4 and 9 in r9
-- r89c9 = [84], r9c8 = 5 -> r234c9 = [576] -> r1c89 = [32] -> r12c5 = [14] -> r5c89 = [41] -> r2c78 = {18} -> r3c8 = 9 -> problem, can't set r2c6 r4c8 to {24}
--> 17(3) @ r8c9 <> {458}
EDIT: r1c89 can't be [23] because r67c9 = {39} -> 3 blocked for c9
24a. -> no 4 in r8c9, no 4,8 in r9c9, no 5 in r9c8
24b. -> 7 locked in r8c9 and r9c8 for 17(3) and n9
24c. -> no 5 in r6c9
24d. no 2 in r9c6

25. r8c5 <> 8
-- r8: h18(3) @ r8c159 = [189] sets r9c1 = 5 but also r9c5 = 5
--> r8c5 <> 8, r9c5 <> 5
Last edited by Nasenbaer on Wed Jan 16, 2008 1:01 pm, edited 1 time in total.
Nasenbaer
Master
Master
Posts: 167
Joined: Thu Jul 20, 2006 7:19 am
Location: Fellbach (Deutschland)

Post by Nasenbaer »

Bedtime is long overdue so just two minor steps and :!: a first placement.

26. r8c9 <> 9
-- r8c9 = 9, r9c89 = [71] -> h18(3) @r8c159 = [549] (only possible combination with 9 in r8c9) -> r12c5 = [23] -> r1c89 = [13] -> r234c9 = [685] -> problem, can't place {48|57} in 12(2) @r67c9
--> r8c9 <> 9
26a. {459} removed from h18(3)

27. c9: no 4,8 in 12(2) @ r6c9
-- 9 in 13(2) = {49} -> r67c9 = [75]
-- 9 in 12(2) = {39}
--> no 4,8 in 12(2) @ r6c9

28. 17(3) @ r8c9: {179} not possible
-- r89c9 = [71], r9c8 = 9 -> h18(3) @ r159c2 = {468} -> r5c12 = [16] -> r5c89 = {23} -> naked pair {23} in r15c9 -> problem, can’t place any combos in 12(2) because {39} and [75] blocked.
--> 17(3) @ r8c9: {179} not possible
28a. -> 17(3) = {278} -> r9c9 = 2 -> 7,8 locked in r8c9 and r9c8 for n9
28b. -> no1,7 in r9c6, no 3 in r5c8, no 7 in r9c4
28c. -> {12578} removed from 23(5) @ r6c8

EDIT: Corrections in blue, unnecessary elimination in 28b. deleted.

It's definitely real hard work to crack this thing. 4 hours now and just a few steps. Oh well.

Sorry I didn't wait for your approval of the chain in step 21 but I think we have to either use chains or give Ruud the satisfaction of defeating us. Can't do that, can't we? :wink:
Last edited by Nasenbaer on Wed Jan 16, 2008 1:30 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 »

Great work guys! :D It's good to see progress even though it does involve some pretty complicated contradiction chains but I think that's probably the only way this puzzle is going to get broken into.

Once this "tag" has been finished, it will be interesting to see if anyone, or any software solver, can come up with a solution without using chains. However please don't post any alternative solutions until the "tag" has finished.

Apart from my steps at the start of this "tag", all the moves so far have come from forum members in Germany!

When I was working through the moves, I didn't initially understand step 20a, which Afmob has now edited to clarify it. An alternative way to do step 20a is

Step 20a. R2C4 in 18(5) cage <> 8 since only combo with 8 is {12348} when R4C2 = 4 clashes with R34C1 = [64] because R2C23+R3C2={123} for this combo.

Both ways are valid. We were just looking at different ways to show an elimination.


Sorry I haven't yet(?) been able to contribute any more steps beyond my initial message. I'll have another look at this puzzle tomorrow, together with any more progress that may have been made.
Last edited by Andrew on Wed Jan 16, 2008 11:06 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 »

A simple follow up to the eliminations in the last step.

29. R159C8 (step 6) = 12 = {138/147/237} (other combinations eliminated)
29a. 2 of {237} must be in R5C8 -> no 2 in R1C8

30. Naked pair {13} in R1C89, locked for R1 and N3 -> R12C5 = [23], R2C9 = 6, R1C67 = [65], clean-up: no 7 in R3C9, no 4 in R9C6, no 3 in R9C7

Wow! I didn't expect so much from step 30 when I started on step 29. :D

Maybe these eliminations will remove some combos from the 5-cell cages. I’ll leave it to others to work that out; it’s getting late!
Last edited by Andrew on Wed Jan 16, 2008 11:00 pm, edited 1 time in total.
Afmob
Expert
Expert
Posts: 103
Joined: Sat Sep 22, 2007 5:36 pm
Location: MV, Germany

Post by Afmob »

31. 2 locked in Innies C1234 for C4
31a. R46C4 <> 5 (step 11a)

32. Innies R9 = 279{18/45}; R9C2 <> 8 since 15(3) can't have 1 and 8 (step 9a)
32a. 15(3) = 5{19/46} <> 2

33. 10(3): R7C6 <> 4 because R67C7 <> 5

34. 24(5): R4C8 <> 7 because 3,6 only possible there

35. Outies R12 = 15(2+2): R4C2 @ 18(5) <> 8 because R34C2 would be [38]
-> Outies wouldn't be 15 since R3C8 <> 1,3
35a. 18(5) = 125{37/46}

The grid looks like this now:

Code: Select all

.-----------------------.-----------------------.-----------.-----------------------.-----------------------.
| 48          48        | 7           9         | 2         | 6           5         | 13          13        |
|           .-----------'-----------------------&#58;           &#58;-----------------------'-----------.           |
| 9         | 125         125         1457      | 3         | 14578       2478        2478      | 6         |
&#58;-----------&#58;           .-----------------------+-----------+-----------------------.           &#58;-----------&#58;
| 236       | 12356     | 12356       14578     | 14578     | 14578       24789     | 2479      | 489       |
|           |           |           .-----------'           '-----------.           |           |           |
| 478       | 1234567   | 12345689  | 1234        1456789     345789    | 12346789  | 1234569   | 459       |
&#58;-----------'-----------+-----------'-----------------------------------'-----------+-----------'-----------&#58;
| 126         156       | 345689      345678      456789      45789       789       | 124         134       |
&#58;-----------.-----------+-----------.-----------------------------------.-----------+-----------.-----------&#58;
| 234678    | 123456789 | 12345689  | 1234        1456789     345789    | 123467    | 123456789 | 379       |
|           |           |           '-----------.           .-----------'           |           |           |
| 234678    | 12346789  | 1234689     345678    | 1456789   | 12357       1346      | 134569    | 359       |
&#58;-----------&#58;           '-----------------------+-----------+-----------------------'           &#58;-----------&#58;
| 156       | 12346789    123468      1345678   | 4569      | 12345789    13469       134569    | 78        |
|           '-----------.-----------------------&#58;           &#58;-----------------------.-----------'           |
| 15          49        | 13468       13568     | 4789      | 358         146       | 78          2         |
'-----------------------'-----------------------'-----------'-----------------------'-----------------------'
mhparker
Grandmaster
Grandmaster
Posts: 345
Joined: Sat Jan 20, 2007 10:47 pm
Location: Germany

Post by mhparker »

Hi guys,

Here are some more moves to keep us going for a while:

36. 24(5) at R2C6 (steps 20c, 23e) = {12489/12678/13578/14568/23478}
36a. {1356} only in R2C6+R4C8
36b. -> {13578/14568} both blocked
36c. -> 24(5) = {12489/12678/23478} (no 5)

37. 5 in C8 locked in 23(5) within R678C8
37a. -> no 5 in R8C6

38. 5 in R2 locked in 18(5) within R2C234
38a. -> no 5 in R34C2

39. R8C159 = [198/567/648]
(Note: [657] blocked because it would force both of R9C58 to 8)
39a. -> no 5 in R8C5
39b. cleanup: no 8 in R9C5

40. Innies R9: R9C1258 = 25(4) = {1789/4579}
40a. AIC: (5)R7C9=(5)R4C9,(8)R3C9-(8=7)R8C9-(7=8)R9C8-(5)R9C1=(5)R9C46
40b. -> either R7C9 contains a 5, or R9C46 contains a 5
40c. -> no 5 in R7C456 (common peers)

[Note: Alternative explanation for those unfamiliar with Eureka notation:
R7C9<>5 -> R4C9=5 -> R3C9=8 -> R8C9=7 -> R9C8=8 -> R9C1<>5 (combinations h25(4), step 40) -> R9C46 = {5..}
-> if R7C9 is not a 5, R9C46 must contain a 5
-> 5 locked in R7C9+R9C46
-> no 5 in R7C456 (CPE)]


41. 5 now unavailable to 10(3) at R6C7 = {127/136} (no 4)

42. 5 in R7 locked in N9 -> not elsewhere in N9

43. either R8C4 = 5, or...
43a. ...R8C1 = 5 (only other place in R8) -> R8C59 = [67] (step 39)
43b. -> no 6,7 in R8C4

44. Outies C12: R28C34 = 20(4)
44a. R8C34 cannot contain both of {68} due to R8C159 (step 39)
44b. -> [1568/2468/5168] blocked
44c. -> R28C34 = [1748/1784/2738/2765/2783/5438/5465/5483/5735]
(Note: [2585] impossible, because R28C4 cannot both contain a 5)
44d. -> no 1,5 in R2C4; no 1,2 in R8C3; no 1 in R8C4

45. 5 in R2 locked in N1 -> not elsewhere in N1

46. 18(5) at R2C2 = {12357/12456}
46a. -> must have exactly 1 of {47}, which must go in R2C4
46b. -> no 4,7 in R4C2

47. 7 in C2 locked in 29(5) at R6C2 = {14789/23789/25679/34679/35678}
(Note: Thus, if 29(5) contains a 4, it must also contain a 9 (locked within R678C2))
47a. R678C2+R8C3 cannot contain both of {49} due to R9C2
47b. -> no 4 in R678C2+R8C3

48. 4 in C2 locked in C1 outies (R159C2) = [459/864]
48a. -> no 1 in R5C2
48b. cleanup: no 6 in R5C1

49. CPE: R4C6 sees all 5s in C5
49a. -> no 5 in R4C6

Code: Select all

.-----------------------.-----------------------.-----------.-----------------------.-----------------------.
| 48          48        | 7           9         | 2         | 6           5         | 13          13        |
|           .-----------'-----------------------&#58;           &#58;-----------------------'-----------.           |
| 9         | 125         125         47        | 3         | 1478        2478        2478      | 6         |
&#58;-----------&#58;           .-----------------------+-----------+-----------------------.           &#58;-----------&#58;
| 236       | 1236      | 1236        14578     | 14578     | 14578       24789     | 2479      | 489       |
|           |           |           .-----------'           '-----------.           |           |           |
| 478       | 1236      | 12345689  | 1234        1456789     34789     | 12346789  | 123469    | 459       |
&#58;-----------'-----------+-----------'-----------------------------------'-----------+-----------'-----------&#58;
| 12          56        | 345689      345678      456789      45789       789       | 124         134       |
&#58;-----------.-----------+-----------.-----------------------------------.-----------+-----------.-----------&#58;
| 234678    | 12356789  | 12345689  | 1234        1456789     345789    | 12367     | 123456789 | 379       |
|           |           |           '-----------.           .-----------'           |           |           |
| 234678    | 1236789   | 1234689     34678     | 146789    | 1237        136       | 134569    | 359       |
&#58;-----------&#58;           '-----------------------+-----------+-----------------------'           &#58;-----------&#58;
| 156       | 1236789     368         3458      | 469       | 1234789     13469       13469     | 78        |
|           '-----------.-----------------------&#58;           &#58;-----------------------.-----------'           |
| 15          49        | 13468       13568     | 479       | 358         146       | 78          2         |
'-----------------------'-----------------------'-----------'-----------------------'-----------------------'
Puzzle still not cracked, but at least we're still making sound progress...
Last edited by mhparker on Thu Jan 17, 2008 11:40 am, edited 2 times in total.
Cheers,
Mike
Nasenbaer
Master
Master
Posts: 167
Joined: Thu Jul 20, 2006 7:19 am
Location: Fellbach (Deutschland)

Post by Nasenbaer »

Nice work. But it took me a long time to figure out how you eliminated the 3 combos in step 47. Maybe you could add some explanations.

I tried to work on N5 by concentrating on 2 and 3 and proofed that they can't be both in the same cage in N5.

50. {2389} removed from 22(4) @r3c5
-- r4c46 = [23], r34c5 = [89] -> r89c5 = [67] -> r567 = {145} -> problem, can't place anything in r6c4
--> {2389} removed from 22(4) @r3c5

51. {2347} removed from 16(4) @ r6c4
-- r6c46 = [23], r67c5 = {47} -> problem, can't place anything in r89c5
--> {2347} removed from 16(4) @ r6c4

52. {2356} removed from 16(4) @ r6c4
-- r6c46 = [23], r67c5 = [56] -> r89c5 = {49} -> only possible combination left for 22(4) @ r3c5 is {1489} -> r4c4 = 4 -> r4c5 = 1 -> r3c5 = 8 -> r4c6 = 9 -> r4c9 = 5 -> problem two 8s for r3 in r3c5 and r3c9.
EDIT: Now it should be better to understand, thanks to Andrew

52a. -> 1 locked in 16(4) @ r6c4 -> no 1 in r4c5

Sorry, only one elimination.
Last edited by Nasenbaer on Thu Jan 31, 2008 5:48 pm, edited 1 time in total.
Post Reply