The New Sudoku Players' Forum

by **ArkieTech** » Fri Jan 25, 2013 7:50 am

Code: Select all: *-----------* |1..|...|3..| |...|817|...| |.2.|...|.6.| |---+---+---| |..8|.4.|..5| |.9.|...|.3.| |6..|.3.|9..| |---+---+---| |.1.|...|.4.| |...|652|...| |..7|...|..9| *-----------*

Play/Print this puzzle online

by **Leren** » Fri Jan 25, 2013 8:05 am

Code: Select all: *--------------------------------------------------------------------------------* | 1 78 459 | 2 6 45 | 3 789 478 | | 49 6 3 | 8 1 7 |c245 259 24 | | 78 2 45 | 45 9 3 | 78 6 1 | |--------------------------+--------------------------+--------------------------| | 37 37 8 | 9 4 6 | 12 12 5 | |a45 9 1 | 57 2 58 |b478 3 6 | | 6 45 2 | 157 3 158 | 9 78 478 | |--------------------------+--------------------------+--------------------------| | 2-5 1 6 | 3 7 9 |d258 4 28 | | 489 48 49 | 6 5 2 | 17 17 3 | | 235 35 7 | 14 8 14 | 6 25 9 | *--------------------------------------------------------------------------------*

m-wing: (5=4) r5c1 - r5c7 = (4-5) r2c7 = r7c7 => r7c1 <> 5; stte

Leren

by **ArkieTech** » Fri Jan 25, 2013 12:16 pm

Code: Select all: *--------------------------------------------------* | 1 78 459 | 2 6 45 | 3 789 478 | | 49 6 3 | 8 1 7 |e24-5 259 24 | | 78 2 45 | 45 9 3 | 78 6 1 | |----------------+----------------+----------------| | 37 37 8 | 9 4 6 | 12 12 5 | |c45 9 1 | 57 2 58 |d478 3 6 | | 6 45 2 | 157 3 158 | 9 78 478 | |----------------+----------------+----------------| |b25 1 6 | 3 7 9 |a258 4 28 | | 489 48 49 | 6 5 2 | 17 17 3 | | 235 35 7 | 14 8 14 | 6 25 9 | *--------------------------------------------------* s-wing 5r7c7=r7c1-(5=4)r5c1-r5c7=r2c7 => -5r2c7; stte

by **tlanglet** » Fri Jan 25, 2013 2:24 pm

Code: Select all: *--------------------------------------------------* | 1 78 459 | 2 6 45 | 3 789 478 | | 49 6 3 | 8 1 7 |*245 259 *24 | | 78 2 45 | 45 9 3 | 78 6 1 | |----------------+----------------+----------------| | 37 37 8 | 9 4 6 | 12 12 5 | |*45 9 1 | 57 2 *58 |*478 3 6 | | 6 *45 2 | 157 3 *158 | 9 78 *478 | |----------------+----------------+----------------| | 25 1 6 | 3 7 9 |*258 4 *28 | | 489 48 49 | 6 5 2 | 17 17 3 | | 235 35 7 | 14 8 14 | 6 25 9 | *--------------------------------------------------*

Almost BUG-Lite (24)b3, (28)b9, (48)b6, (45)b4, (58)b5: internal SIS 5r2c7, 5r7c7, 7r5c7, 7r6c9, 1r6c6 => r5c6=8
(5-4)r2c7=(4-8)r5c7=8r5c6
||
5r7c7-(5=8)r9486c8-r6c6=8r5c6
||
(7-8)r5c7=8r5c6
||
7r6c9-(7=8)r6c8-r6c6=8r5c6
||
(1-8)r6c6=8r5c8 => r5c6=8

As sometimes happens, the long fun solutions prove to be useless...... :roll:

Next I noticed another Almost BUG-Lite (58)b5, (45)b4, (48)b7: internal SIS 1r6c6, 9r8c1
My first solution resulted in the same useless result as the prior AB-L(). So I used another logic path:

(4=9)r2c1-ABL(458)=1r6c6-(1=4)r9c6=r1c6-4r3c4 => r3c3<>4

which did more damage than the first but still required another step to complete the puzzle.

(2=5)r9c8-(5=8)r941c2-r3c1=r3c7-r7c7=8r7c9 => r7c9<>2

Ted

by **eleven** » Fri Jan 25, 2013 4:06 pm

tlanglet wrote:Almost BUG-Lite (24)b3, (28)b9, (48)b6, (45)b4, (58)b5

This is no valid BUG-Lite. You can check this, because an "outside" number like 4r2c1 or 5r9c2 would leave a unique solution of the BL pattern. In a valid BL pattern all outside numbers seeing the same digit would make the pattern invalid.
[Added:]Maybe its easier to see it this way: If you have a solution with one of the 2 possible BL-pattern solutions, it is unique, if it contains a 4r2c1 or a 5r9c2.

by **Leren** » Fri Jan 25, 2013 8:37 pm

Code: Select all: *--------------------------------------------------------------------------------* | 1 78 459 | 2 6 45 | 3 789 478 | | 49 6 3 | 8 1 7 | 245C 259 24 | | 78 2 45 | 45 9 3 |b78 6 1 | |--------------------------+--------------------------+--------------------------| | 37 37 8 | 9 4 6 |b12 12 5 | |a45A 9 1 | 57 2 58 |b478B 3 6 | | 6 45 2 | 157 3 158 | 9 78 478 | |--------------------------+--------------------------+--------------------------| | 2-5 1 6 | 3 7 9 |b258D 4 28 | | 489 48 49 | 6 5 2 |b17 17 3 | | 235 35 7 | 14 8 14 | 6 25 9 | *--------------------------------------------------------------------------------*

(5=4) r5c1 - (4=5) r34578c7 => r7c1 <> 5; stte

or

(5=4) r5c1 - r5c7 = (4-5) r7c2 = r7c7 => r7c1 <> 5; stte

Leren

by **Luke** » Sat Jan 26, 2013 3:14 am

tlanglet wrote:
Code: Select all
*--------------------------------------------------* | 1 78 459 | 2 6 45 | 3 789 478 | | 49 6 3 | 8 1 7 |*245 259 *24 | | 78 2 45 | 45 9 3 | 78 6 1 | |----------------+----------------+----------------| | 37 37 8 | 9 4 6 | 12 12 5 | |*45 9 1 | 57 2 *58 |*478 3 6 | | 6 *45 2 | 157 3 *158 | 9 78 *478 | |----------------+----------------+----------------| | 25 1 6 | 3 7 9 |*258 4 *28 | | 489 48 49 | 6 5 2 | 17 17 3 | | 235 35 7 | 14 8 14 | 6 25 9 | *--------------------------------------------------*

Almost BUG-Lite (24)b3, (28)b9, (48)b6, (45)b4, (58)b5.......
Ted

Wow, buddy, that's one ambitious attempt at a BUG-Lite :!:

Each value in a BL must appear twice in all of a row/box/column. In this pattern, neither the 4 nor 5 in box 4 meets that criteria in the columns.

by **eleven** » Sat Jan 26, 2013 12:44 pm

Luke451 wrote:Each value in a BL must appear twice in all of a row/box/column.

Oh yes, so simple can it be said for deadly patterns with 2 candidates per cell.

by **tlanglet** » Sat Jan 26, 2013 1:28 pm

Thanks all for the various replies. I know the conditions for DPs, but I seemed to have been 'blinded" by my enthusiasm to solve all the inference conditions that I forget to truly analyze the pattern.

Ted

by **Marty R.** » Sat Jan 26, 2013 8:33 pm

I left another M-Wing on the table. :oops:

Used a short AIC and a four-cell XY-Chain.

The New Sudoku Players' Forum

AU Tough January 25, 2013

AU Tough January 25, 2013

Re: AU Tough January 25, 2013

Re: AU Tough January 25, 2013

Re: AU Tough January 25, 2013

Re: AU Tough January 25, 2013

Re: AU Tough January 25, 2013

Re: AU Tough January 25, 2013

Re: AU Tough January 25, 2013

Re: AU Tough January 25, 2013

Re: AU Tough January 25, 2013