dailysudoku.com :: View topic - Astraware Diabolical 14/12/10

dailysudoku.com
Discussion of Daily Sudoku puzzles

FAQ

Memberlist

Usergroups

Profile

Astraware Diabolical 14/12/10

dailysudoku.com Forum Index -> Other puzzles

View previous topic :: View next topic

Author

Message

peterj

Joined: 26 Mar 2010
Posts: 974
Location: London, UK

Posted: Tue Dec 14, 2010 8:53 am Post subject: Astraware Diabolical 14/12/10

Code:

*-----------*
|.2.|35.|...|
|...|...|8.5|
|8..|71.|.2.|
|---+---+---|
|1..|...|47.|
|..7|.6.|5..|
|.42|...|..8|
|---+---+---|
|.3.|.21|..6|
|2.9|...|...|
|...|.47|.5.|
*-----------*

The ratings on the Astraware site are patchy as they don't include turbot moves. However, this one looks reasonably tough! Rating 24,900 (SE 7.3)

[Edit. 1 hr later] Unless I have missed something obvious, this puzzle looks pretty unreasonable. For me - an Extreme. Though perhaps someone can find an elegant solution.

(c) Astraware

peterj

Joined: 26 Mar 2010
Posts: 974
Location: London, UK

Posted: Tue Dec 14, 2010 2:35 pm Post subject:

This was definitely "extreme" for me! Bunch of my first attempts made absolutely no dent in it at all. After a long time I found this path...

Code:

*-----------------------------------------------------------*
| 49 2 146 | 3 5 8 | 167-9 1469 1479 |
| 347 167 1346 | 246 9 246 | 8 1346 5 |
| 8 569 3456 | 7 1 46 | 36-9 2 (49) |
|-------------------+-------------------+-------------------|
| 1 5689 3568 | 258 38 2359 | 4 7 29 |
| 39 89 7 | 1248 6 2349 | 5 139 129 |
| 359 4 2 | 15 7 359 | 1369 1369 8 |
|-------------------+-------------------+-------------------|
| 457 3 45 | 58 2 1 | (79) 89 6 |
| 2 57 9 | 568 38 356 | (17) 148 (147) |
| 6 18 18 | 9 4 7 | 2 5 3 |
*-----------------------------------------------------------*

#1 (9=4)r3c9 - ANP(4=17)r8c79 - (7=9)r7c7 ; r13c7<>9

Then using much the same move but with an HP (in fact they can be combined using memory - I realised later) .. HPs often seem to do real damage to hard puzzles?

Code:

*---------------------------------------------------------------*
| 4(9) 2 146 | 3 5 8 | 167 1469 1479 |
| 347 167 1346 | 246 9 246 | 8 1346 5 |
| 8 56(9) 3456 | 7 1 46 | 36 2 (49) |
|-------------------+-------------------+-----------------------|
| 1 5689 3568 | 258 38 2359 | 4 7 29 |
| (39) 89 7 | 1248 6 2349 | 5 1(3)9 129 |
| 359 4 2 | 15 7 359 | 1(36)-9 1(36)9 8 |
|-------------------+-------------------+-----------------------|
| 457 3 45 | 58 2 1 | (79) 89 6 |
| 2 57 9 | 568 38 356 | (17) 148 (147) |
| 6 18 18 | 9 4 7 | 2 5 3 |
*---------------------------------------------------------------*

#2 HP(36)=(3)r5c8 - (3=9)r5c1 - r1c1=r3c2 - (9=4)r3c9 - ANP(4=17)r8c79 - (7=9)r7c7 ; r6c7<>9

That cracked it for a final ...

Code:

#3 Type 4 UR(17) r18c79 ; r18c9<>1

Starting to look worryingly like the Eureka forum!

daj95376

Joined: 23 Aug 2008
Posts: 3854

Posted: Tue Dec 14, 2010 6:00 pm Post subject:

Neither my solver nor HoDoKu liked it, either.

Code:

(1=8)r9c2 - r5c2 = r5c4 - (8=5)r7c4 - r8c46 = (5-7)r8c2 = (7)r2c2 => r2c2<>1

(6=4)r3c6 - (4=9)r3c9 - (9=2)r4c9 - r4c4 = (2)r2c4 => r2c4<>6

r8c1 2-String Kite <> 5 r6c6

(3)r4c5 = r8c5 - (3=5)r8c6 - (5=7)r8c2 - r2c2 = (7-3)r2c1 = (3)r56c1 => r4c3<>3

(9=4)r3c9 - r8c9 = (4-8)r8c8 = (8-9)r7c8 = (9)r7c7 => r13c7<>9

Yeah, I know the stupid Kite looks out of place but, trust me, it's needed.

storm_norm

Joined: 18 Oct 2007
Posts: 1741

Posted: Wed Dec 15, 2010 10:46 am Post subject:

Code:

+-----------------+-----------------+------------------+
| 49 2 146 | 3 5 8 | 1679 1469 1479 |
| 347 167 1346 | 24-6 9 246 | 8 1346 5 |
| 8 569 3456 | 7 1 (46) | 369 2 (49) |
+-----------------+-----------------+------------------+
| 1 5689 3568 | (258) 38 2359 | 4 7 (29) |
| 39 89 7 | 1248 6 2349 | 5 139 129 |
| 359 4 2 | 15 7 359 | 1369 1369 8 |
+-----------------+-----------------+------------------+
| 457 3 45 | (58) 2 1 | 79 89 6 |
| 2 57 9 | (568) 38 35-6 | 17 148 147 |
| 6 18 18 | 9 4 7 | 2 5 3 |
+-----------------+-----------------+------------------+

ALS[(6)r8c4 = (2)r4c4]r478c4 - (2=9)r4c9 - (9=4)r3c9 - (4=6)r3c6; r8c6 and r2c4 <> 6

Code:

+--------------------+-------------------+--------------------+
| 4(9) 2 146 | 3 5 8 | 1679 1469 1479 |
| 347 167 1346 | 24 9 246 | 8 1346 5 |
| 8 56(9) 3456 | 7 1 46 | 369 2 (49) |
+--------------------+-------------------+--------------------+
| 1 5689 568-3 | 258 8(3) 2359 | 4 7 29 |
| (39) 89 7 | 1248 6 249-3 | 5 139 129 |
| 359 4 2 | 15 7 359 | 1369 1369 8 |
+--------------------+-------------------+--------------------+
| 457 3 45 | 58 2 1 | 79 89 6 |
| 2 57 9 | 6 (38) 35 | 17 1(48) 17(4) |
| 6 18 18 | 9 4 7 | 2 5 3 |
+--------------------+-------------------+--------------------+

(3=9)r5c1 - (9)r1c1 = (9)r3c2 - (9=4)r3c9 - (4)r8c9 = (4-8)r8c8 = (8-3)r8c5 = (3)r4c5; r4c3 and r5c6 <> 3

Code:

+-----------------+-----------------------+--------------------+
| 49 2 146 | 3 5 8 | 1679 1469 1479 |
| 47 167 1346 | 24 9 246 | 8 1346 5 |
| 8 569 3456 | 7 1 46 | 369 2 (49) |
+-----------------+-----------------------+--------------------+
| 1 5689 568 | 58(2) 38 359(2) | 4 7 (29) |
| 39 89 7 | 8-1(24) 6 9(24) | 5 139 129 |
| 359 4 2 | (15) 7 59 | 1369 1369 8 |
+-----------------+-----------------------+--------------------+
| 457 3 45 | (58) 2 1 | 79 89 6 |
| 2 57 9 | 6 3(8) 35 | 17 1(48) 17(4) |
| 6 18 18 | 9 4 7 | 2 5 3 |
+-----------------+-----------------------+--------------------+

ahs(24)r5c46 = (2)r4c46 - (2=9)r4c9 - (9=4)r3c9 - (4)r8c9 = (4-8)r8c8 = (8)r8c5 - (8=5)r7c4 - (5=1)r6c4; r5c4 <> 1

Code:

+-------------------+-----------------+---------------------+
| (49) 2 146 | 3 5 8 | 167-9 146-9 147-9 |
| (47) 167 1346 | 24 9 246 | 8 1346 5 |
| 8 56-9 3456 | 7 1 46 | 369 2 (49) |
+-------------------+-----------------+---------------------+
| 1 5689 568 | 258 38 2359 | 4 7 29 |
| 39 89 7 | 248 6 249 | 5 139 129 |
| 359 4 2 | 1 7 59 | 369 369 8 |
+-------------------+-----------------+---------------------+
| 45(7) 3 45 | 58 2 1 | 79 89 6 |
| 2 (57) 9 | 6 (38) (35) | 17 1(48) 17(4) |
| 6 18 18 | 9 4 7 | 2 5 3 |
+-------------------+-----------------+---------------------+

(9=4)r3c9 - (4)r8c9 = (4-8)r8c8 = (8-3)r8c5 = (3-5)r8c6 = (5-7)r8c2 = (7)r7c1 - (7=4)r2c1 - (4=9)r1c1;
r3c2 <> 9
r1c789 <> 9

Code:

+---------------+--------------+-------------------+
| 9 2 146 | 3 5 8 | 167 146 147 |
| 47 167 1346 | 24 9 246 | 8 1346 5 |
| 8 56 3456 | 7 1 46 | 36-9 2 (49) |
+---------------+--------------+-------------------+
| 1 689 68 | 258 38 35 | 4 7 29 |
| 3 89 7 | 248 6 24 | 5 19 129 |
| 5 4 2 | 1 7 9 | 36 36 8 |
+---------------+--------------+-------------------+
| 47 3 45 | 58 2 1 | (79) 89 6 |
| 2 57 9 | 6 38 35 | (17) 148 (147) |
| 6 18 18 | 9 4 7 | 2 5 3 |
+---------------+--------------+-------------------+

xy-wing with the ever elusive als middle term
(9=7)r7c7 - als[(7=4)r8c79] - (4=9)r3c9; r3c7 <> 9
which coincidentally is peterj's first move.

daj95376

Joined: 23 Aug 2008
Posts: 3854

Posted: Wed Dec 15, 2010 11:15 pm Post subject:

This puzzle has a much simpler solution than the one I originally presented. However, it relies on performing a chain before performing 2x critical UR eliminations.

Code:

(1=5)r6c4 - (5=8)r7c4 - (8=9)r7c8 - (9=7)r7c7 - (7=1)r8c7 => r6c7<>1

+--------------------------------------------------------------+
| 49 2 146 | 3 5 8 | 1679 1469 1479 |
| 347 167 1346 | 246 9 246 | 8 1346 5 |
| 8 569 3456 | 7 1 46 | 369 2 49 |
|--------------------+--------------------+--------------------|
| 1 5689 3568 | 258 38 2359 | 4 7 29 |
| 39 89 7 | 1248 6 2349 | 5 139 129 |
| 359 4 2 | 15 7 359 | 369 1369 8 |
|--------------------+--------------------+--------------------|
| 457 3 45 | 58 2 1 | 79 89 6 |
| 2 57 9 | 568 38 356 | 17 148 147 |
| 6 18 18 | 9 4 7 | 2 5 3 |
+--------------------------------------------------------------+
# 89 eliminations remain

r18c79 <17> UR via s-link <> 1 r1c9 critical
r18c79 <17> UR via s-link <> 1 r8c9 critical

r25c46 <24> UR Type 1.2223 r2c6<>24

daj95376

Joined: 23 Aug 2008
Posts: 3854

Posted: Thu Dec 16, 2010 2:47 am Post subject:

I recently made numerous modifications to my solver and I'm in the middle of running test puzzles. This is one of the test puzzles because my SIN() routine was acting up for awhile. Now, I'm just having fun with the results.

Code:

after basics
+--------------------------------------------------------------+
| 49 2 146 | 3 5 8 | 1679 1469 1479 |
| 347 167 1346 | 246 9 246 | 8 1346 5 |
| 8 569 3456 | 7 1 46 | 369 2 49 |
|--------------------+--------------------+--------------------|
| 1 5689 3568 | 258 38 2359 | 4 7 29 |
| 39 89 7 | 1248 6 2349 | 5 13(9) 129 |
| 359 4 2 | 15 7 359 | 1369 1369 8 |
|--------------------+--------------------+--------------------|
| 457 3 45 | 58 2 1 | 79 89 6 |
| 2 57 9 | 568 38 356 | 17 148 147 |
| 6 18 18 | 9 4 7 | 2 5 3 |
+--------------------------------------------------------------+
# 90 eliminations remain

Here's a SIN from my solver. It's a sequence of assignments that lead to a contradiction.

9r7c8 8r8c8 4r8c9 9r3c9 9r1c1 3r5c1 1r5c8 [b6]+2

The trick is to turn it into something that's more "logical". How about a discontinuous chain with a "lasso" and a fin candidate?

If the fin candidate is true: (9)r5c8-(9)r7c8

If the fin candidate is false:

(9-8)r7c8=(8-4)r8c8=r8c9-(4=9)r3c9-r3c2=r1c1-(9=3)r5c1-(93=1)r5c8-(1=29)r45c9-(9=4)r3c9-r8c9=(4-8)r8c8=(8-9)r7c8

(the "lasso" occurs in the first/last four cells of the discontinuous loop.)

Code:

after the dust clears
+--------------------------------------------------------------+
| 49 2 16 | 3 5 8 | 17 1469 1479 |
| 34 7 136 | 246 9 246 | 8 146 5 |
| 8 69 5 | 7 1 46 | 3 2 49 |
|--------------------+--------------------+--------------------|
| 1 689 36 | 28 38 5 | 4 7 29 |
| 39 89 7 | 248 6 2349 | 5 139 129 |
| 5 4 2 | 1 7 39 | 6 39 8 |
|--------------------+--------------------+--------------------|
| 7 3 4 | 5 2 1 | 9 8 6 |
| 2 5 9 | 68 38 36 | 17 14 147 |
| 6 1 8 | 9 4 7 | 2 5 3 |
+--------------------------------------------------------------+
# 47 eliminations remain

r34 X-Wing <> 9 r1c9,r5c29

r18c79 <17> UR Type 2.2233 <> 4 r3c9

peterj

Joined: 26 Mar 2010
Posts: 974
Location: London, UK

Posted: Thu Dec 16, 2010 8:00 am Post subject:

daj95376 wrote:

This puzzle has a much simpler solution than the one I originally presented.

Nice! I did spot the xy-chain too but dismissed the elimination as "not helpful". But as you spotted it creates the strong link in the UR(17) which seems critical in many solutions.

Good luck with SIN() !

Display posts from previous:

	dailysudoku.com Forum Index -> Other puzzles	All times are GMT
Page 1 of 1

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