The New Sudoku Players' Forum

by **ncantoral** » Sun Jun 22, 2008 6:12 am

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

Code: Select all: .---------------------.---------------------.---------------------. | 15 12 125 | 678 78 3 | 9 67 4 | | 349 6 349 | 457 2 45 | 8 137 137 | | 8 7 34 | 46 9 1 | 236 5 26 | :---------------------+---------------------+---------------------: | 134 1348 7 | 48 5 2 | 16 136 9 | | 59 59 6 | 1 3 7 | 4 2 8 | | 2 48 13 | 9 6 48 | 5 137 137 | :---------------------+---------------------+---------------------: | 147 124 124 | 3 178 68 | 1267 9 5 | | 13579 123 12359 | 57 4 569 | 1267 8 26 | | 6 59 8 | 2 17 59 | 137 4 13 | '---------------------'---------------------'---------------------'

my sudoku knowlegde has been used up after staring for so long.
I get these off of a site called sudokupuzz and the difficulty range is vast. I have come accross numerous puzzles rated 7.2 - 8.4 SE

by **daj95376** » Sun Jun 22, 2008 3:42 pm

While you're waiting for a solution with insight ...

Code: Select all: r7c5 =8= r7c6 =6= r7c7 -6- r4c7 -1- r789c7 =1= r9c9 -1- r9c5 -7- r7c5 => [r7c5]<>7 r7c7 =6= r7c6 =8= r7c5 =1= r9c5 =7= r9c7 -7- r7c7 => [r7c7]<>7 ____________________________________________________________________________________ Sashimi X-Wing => [r6c2]<>4

by **submacrolize** » Sun Jun 22, 2008 10:38 pm

You can view the solution here.

Click on the gray squares that you want to see. Right click to view the whole solution

by **eleven** » Mon Jun 23, 2008 5:58 pm

Found some UR eliminations, but it kept to be nasty:

xyz-wing to remove 5 from r3c4, a sashimi x-wing to remove 3 from r4c7, a kite removes 4 from r6c3. All of them already done by ncantoral.

Code: Select all: *--------------------------------------------------------* | 15 12 125 | 678 78 3 | 9 67 4 | | 349 6 349 | 457 2 45 | 8 #137 #137 | | 8 7 34 | 46 9 1 | 236 5 26 | |---------------------+----------------+-----------------| | 134 1348 7 | 48 5 2 | 136 *136 9 | | 59 59 6 | 1 3 7 | 4 2 8 | | 2 48 13 | 9 6 48 | 5 #137 #137 | |---------------------+----------------+-----------------| | 147 124 124 | 3 178 68 | 1267 9 5 | | 13579 123 12359 | 57 4 569 | 1267 8 26 | | 6 59 8 | 2 17 59 | 137 4 13 | *--------------------------------------------------------*

Now these UR eliminations:
UR 26 in r38c79: r8c7<>6
UR 17 in r26c89: r6c9<>1 (note r6c9=1 -> r2c8=1)
UR 37 in same cells: r6c9<>3 (r6c9=3 -> r2c8=3)

Bringing me here after skysraper for 3:

Code: Select all: *-------------------------------------------------------* | 15 12 125 | 678 78 3 | 9 67 4 | | 349 6 349 | 457 2 45 | 8 #17 13 | | 8 7 34 | 46 9 1 | 236 5 26 | |---------------------+----------------+----------------| | 134 1348 7 | 48 5 2 | 16 136 9 | | 59 59 6 | 1 3 7 | 4 2 8 | | 2 48 13 | 9 6 48 | 5 13 7 | |---------------------+----------------+----------------| | 147 124 124 | 3 178 68 | 1267 9 5 | | 13579 123 12359 | 57 4 569 | 127 8 26 | | 6 59 8 | 2 #17 59 | 137 4 13 | *-------------------------------------------------------*

Now also the 1 in r6c8 can be eliminated by the UR 17 in the same cells. (If you cant see it, reinsert 1 in r6c9 and 7 in r2c9 and r6c8.)
This gives an xyz-wing 124 to eliminate 2 from r8c2.

Code: Select all: *----------------------------------------------------* | 15 12 25 | 678 78 3 | 9 67 4 | | 349 6 349 | 457 2 45 | 8 17 13 | | 8 7 34 | 46 9 1 | 236 5 26 | |-------------------+----------------+---------------| | 34 348 7 | 48 5 2 | 16 16 9 | |#59 #59 6 | 1 3 7 | 4 2 8 | | 2 48 1 | 9 6 48 | 5 3 7 | |-------------------+----------------+---------------| | 147 124 24 | 3 178 68 | 1267 9 5 | |#1379 13 2359 | 57 4 #569 | 127 8 26 | | 6 #59 8 | 2 17 #59 | 137 4 13 | *----------------------------------------------------*

59 MUG pattern eliminates 5 from r8c1.

Code: Select all: *-------------------------------------------------------* | 15 12 25 | 678 78 3 | 9 67 4 | | 349 6 349 | 457 2 45 | 8 17 13 | | 8 7 34 | 46 9 1 | 236 5 26 | |---------------------+----------------+----------------| | 34 348 7 | 48 5 2 | 16 16 9 | | 59 59 6 | 1 3 7 | 4 2 8 | | 2 48 1 | 9 6 48 | 5 3 7 | |---------------------+----------------+----------------| | 147 124 24 | 3 178 68 | 1267 9 5 | | 1379 13 2359 | 57 4 569 | 127 8 26 | | 6 59 8 | 2 17 59 | 137 4 13 | *-------------------------------------------------------*

But now i didnt find something nice.

Lets try the local notation:
r8c1-7-r7c123-1-r7c5=1=r9c5-1-r9c79=1=r78c7-1-r4c7-6-r7c7=6=r7c6-6-r89c6-5-r8c4-7-r8c1

In my own:
r8c1=7 -> r7c123=124 -> r7c5<>1 -> r9c5=1 -> r78c7=1 -> r4c7=6 -> r7c6=6 -> r89c6=59 -> r8c4=7

A kite then solves it.

by **Bigtone53** » Mon Jun 23, 2008 6:14 pm

.. but will submacrolize grow up one day and realise the difference between this erudite description of how, and the brute force result that no-one is interested in?

by **wintder** » Mon Jun 23, 2008 6:21 pm

Bigtone53 wrote:.. but will submacrolize grow up one day and realise the difference between this erudite description of how, and the brute force result that no-one is interested in?

Doubt it. If I could I would ban him/her/it from posting.

by **hobiwan** » Mon Jun 23, 2008 9:19 pm

We've had chains and URs, so let's try ALS:

Code: Select all: .---------------------.----------------------.---------------------. | 15 12 125 | 6-78 78 3 | 9 A67 4 | | 349 6 349 | 457 2 45 | 8 137 137 | | 8 7 34 | 46 9 1 | 236 5 A26 | :---------------------+----------------------+---------------------: | 134 1348 7 | 48 5 2 | 16 136 9 | | 59 59 6 | 1 3 7 | 4 2 8 | | 2 48 13 | 9 6 48 | 5 137 137 | :---------------------+----------------------+---------------------: | 147 124 124 | 3 178 68 | 1267 9 5 | | 13579 123 12359 | B57 4 B569 | 1267 8 C26 | | 6 59 8 | 2 17 B59 | 137 4 13 | '---------------------'----------------------'---------------------' Almost Locked Set XY-Wing: A=r1c8,r3c9 - {267}, B=r8c46,r9c6 - {5679}, C=r8c9 - {26}, Y,Z=2,6, X=7 => r1c4<>7 Naked Triple: 6,8,4 in r134c4 => r2c4<>4 .----------------------.---------------------.----------------------. | 15 12 125 | 68 B78 3 | 9 67 4 | | 349 6 349 | 57 2 45 | 8 137 137 | | 8 7 34 | 46 9 1 | 236 5 26 | :----------------------+---------------------+----------------------: | 134 1348 7 | 48 5 2 | 16 136 9 | | 59 59 6 | 1 3 7 | 4 2 8 | | 2 48 13 | 9 6 48 | 5 137 137 | :----------------------+---------------------+----------------------: | A147 A124 A124 | 3 A178 68 | 126-7 9 5 | | 13579 123 12359 | 57 4 569 | 1267 8 26 | | 6 59 8 | 2 B17 59 | B137 4 B13 | '----------------------'---------------------'----------------------' Almost Locked Set XY-Wing: A=r7c1235 - {12478}, B=r9c79 - {137}, C=r19c5 - {178}, Y,Z=1,8, X=7 => r7c7<>7 .---------------------.----------------------.----------------------. | 15 12 125 | 68 78 3 | 9 67 4 | | 349 6 349 | 57 2 45 | 8 137 137 | | 8 7 34 | 46 9 1 | 236 5 26 | :---------------------+----------------------+----------------------: | 134 1348 7 | 48 5 2 | A16 136 9 | | 59 59 6 | 1 3 7 | 4 2 8 | | 2 48 13 | 9 6 48 | 5 137 137 | :---------------------+----------------------+----------------------: | 147 124 124 | 3 178 68 | A126 9 5 | | 135-79 123 12359 | B57 4 B569 | A1267 8 C26 | | 6 59 8 | 2 17 B59 | 137 4 13 | '---------------------'----------------------'----------------------' Almost Locked Set XY-Wing: A=r478c7 - {1267}, B=r8c46,r9c6 - {5679}, C=r8c9 - {26}, Y,Z=2,6, X=7 => r8c1<>7 Hidden Single: r7c1=7 Skyscraper: 4 in r2c1,r3c4 (connected by r4c14) => r2c6,r3c3<>4 Singles

by **Glyn** » Mon Jun 23, 2008 9:34 pm

wintder wrote:
Bigtone53 wrote:.. but will submacrolize grow up one day and realise the difference between this erudite description of how, and the brute force result that no-one is interested in?

Doubt it. If I could I would ban him/her/it from posting.

I think you may have hit the nail on the head with 'it', sub, macro,ize sounds like a relative of g.r.emlin

by **Bigtone53** » Mon Jun 23, 2008 9:56 pm

Glyn wrote:I think you may have hit the nail on the head with 'it', sub, macro,ize sounds like a relative of g.r.emlin

That would explain many things. It is an annoyance rather than a problem but as Daj95376 pointed out, there is a supporter out there. If we still have a moderator what does he/ she / (it?) think :?:

by **Draco** » Tue Jun 24, 2008 8:23 pm

ncantoral wrote:...
my sudoku knowlegde has been used up after staring for so long.
I get these off of a site called sudokupuzz and the difficulty range is vast. I have come accross numerous puzzles rated 7.2 - 8.4 SE

FWIW, I've also noticed that SudokuPuzz frequently delivers puzzles of fairly random difficulty, with little regard to the level requested. And sometimes it delivers puzzles at the level requested. Hard to know what you're getting unless you score it with SE or another program.

If you are looking for more reliable puzzle sources, a couple of sites I've found that tend to deliver puzzles in line with the requested difficulty level are Life.com and Player.co.uk. No endorsement of these sites is implied beyond noting I've seen many good puzzles come from them.

Cheers...

- drac

by **ncantoral** » Tue Jun 24, 2008 8:53 pm

FWIW, I've also noticed that SudokuPuzz frequently delivers puzzles of fairly random difficulty, with little regard to the level requested. And sometimes it delivers puzzles at the level requested. Hard to know what you're getting unless you score it with SE or another program

yes, its a big variation in difficulty. I am not trying to promote the site, in fact I think its a bit lackluster for the site not to better regulate its difficulty variation. box of chocolates comes to mind

. if the average user finds that site and thinks that all fiendish puzzles are easy/hard, whatever the case may be, then its greatly misleading.

all that said, I am in the camp that believes its totally subjective and up to the player to make his/her sudoku experience what it is regardless of difficulty.

and thanx for the reference to the other sites. the vanhegan site is very nice for sampling puzzles on many difficulty levels. but even that site has had some easier than normal extremes and harder than normal fiendish.

by **Steve K** » Wed Jun 25, 2008 2:48 am

Here is a stab at it: start, some singles,
Locked 6 r13c4 => r8c4<>6
Locked 8 r2c79 => r2c46<>8
AAIC: (5=4)r2c6-[(4)kite: r6c6=r6c23-r45c1=*r2c1]=*(4-7)r7c1=(7)r8c1-(7=5)r8c4 => r89c6, r123c4<>5 singles

by **Bigtone53** » Thu Jun 26, 2008 7:30 pm

Glyn wrote:
I think you may have hit the nail on the head with 'it', sub, macro,ize sounds like a relative of g.r.emlin

That would explain many things. It is an annoyance rather than a problem but as Daj95376 pointed out, there is a supporter out there. If we still have a moderator what does he/ she / (it?) think

On reflection, now that I am not full of beer-fuelled rage and know that sub--macro--lise is just a program, I am happy to treat it as just background noise. Annoying but there. However, iffy things seem to happen on your machine if you write quoting the name it actually uses.

by **daj95376** » Thu Jun 26, 2008 9:49 pm

Upon review of eleven's UR effort, it's too bad he didn't tackle cell [r6c8] for URs.

Code: Select all: [r6c8]=1|7 -> UR 17 in [r26c89] => ([r6c8]=3, [r6c3]=1, [r6c9]=7, XYZ-Wing) +-----------------------------------------------------------------------+ | 15 12 125 | 678 78 3 | 9 67 4 | | 349 6 349 | 457 2 45 | 8 137 137 | | 8 7 34 | 46 9 1 | 236 5 26 | |-----------------------+-----------------------+-----------------------| | 134 1348 7 | 48 5 2 | 16 136 9 | | 59 59 6 | 1 3 7 | 4 2 8 | | 2 48 13 | 9 6 48 | 5 137 137 | |-----------------------+-----------------------+-----------------------| | 147 124 124 | 3 178 68 | 1267 9 5 | | 13579 123 12359 | 57 4 569 | 1267 8 26 | | 6 59 8 | 2 17 59 | 137 4 13 | +-----------------------------------------------------------------------+

daj95376 motto: If you can't be original, hindsight is always second best.

by **Glyn** » Fri Jun 27, 2008 12:58 am

I'm slightly confused by one of the UR's which appears not to be there.
First the ones that I think are ok.
From eleven
UR 26 in r38c79 avoidable with candidates labelled a,b or c => r8c7<>6

Code: Select all: .------------------.------------------.----------------------. | 15 12 125 | 678 78 3 | 9 67 4 | | 349 6 349 | 457 2 45 | 8 137 137 | | 8 7 34 | 46 9 1 | 23a6 5 2a6 | :------------------+------------------+----------------------: | 134 1348 7 | 48 5 2 | 16 136 9 | | 59 59 6 | 1 3 7 | 4 2 8 | | 2 48 13 | 9 6 48 | 5 137 137 | :------------------+------------------+----------------------: | 147 124 124 | 3 178 68 | 1267 9 5 | | 13579 123 12359| 57 4 569 | 1b2(-6)7c 8 26a | | 6 59 8 | 2 17 59 | 137 4 13 | '------------------'------------------'----------------------'

Followed by UR 17 in r26c89 avoidable with candidates a or b. =>r6c9<>1

Code: Select all: .------------------.------------------.---------------------. | 15 12 125 | 678 78 3 | 9 67 4 | | 349 6 349 | 457a 2 45 | 8 137 137 | | 8 7 34 | 46 9 1 | 236 5 26 | :------------------+------------------+---------------------: | 134 1348 7 | 48 5 2 | 16 136 9 | | 59 59 6 | 1 3 7 | 4 2 8 | | 2 48 1b3 | 9 6 48 | 5 137 (-1)37a | :------------------+------------------+---------------------: | 147 124 124 | 3 178 68 | 1267 9 5 | | 13579 123 12359| 57 4 569 | 127 8 26 | | 6 59 8 | 2 17 59 | 137 4 13 | '------------------'------------------'---------------------'

The UR 37 in r26c89 can't happen as a 1 is mandatory in r2c89. So no eliminations there.
I'm not sure daj95376's post clarified things for me

The New Sudoku Players' Forum

another gem of a "fiendish" named puzzle

another gem of a "fiendish" named puzzle

Re: another gem of a "fiendish" named puzzle