The New Sudoku Players' Forum

by **ArkieTech** » Sat Jul 04, 2015 1:24 am

Code: Select all: *-----------* |...|8..|..4| |.85|..9|..3| |3..|2..|..7| |---+---+---| |.4.|3..|.6.| |2.3|...|9.1| |.1.|...|.4.| |---+---+---| |4..|..2|..8| |6..|5..|21.| |8..|..7|...| *-----------*

Play/Print this puzzle online

by **Leren** » Sat Jul 04, 2015 1:32 am

Code: Select all: *--------------------------------------------------------------* |b179 269-7 269 | 8 1567 3 | 156 59 4 | |b17 8 5 | 1467 1467 9 | 16 2 3 | | 3 b69 4 | 2 156 156 | 1568 589 7 | |--------------------+--------------------+--------------------| | 59-7 4 89 | 3 12789 18 | 78 6 25 | | 2 a67 3 | 467 45678 4568 | 9 78 1 | | 59-7 1 689 | 679 26789 68 | 3 4 25 | |--------------------+--------------------+--------------------| | 4 59 19 | 169 1369 2 | 57 357 8 | | 6 3 7 | 5 48 48 | 2 1 9 | | 8 259 129 | 19 139 7 | 4 35 6 | *--------------------------------------------------------------*

ALS XZ Rule: X = 6, Z = 7: (7=6) r5c2 - (6=7) r1c1, r2c1, r3c2 => - 7 r1c2, r46c1; stte

Leren

by **pjb** » Sat Jul 04, 2015 2:41 am

Code: Select all: 179 2679 269 | 8 1567 3 | 156 59 4 17 8 5 | 1467 1467 9 | 16 2 3 3 69 4 | 2 156 156 | 1568 589 7 ------------------------+----------------------+--------------------- 579 4 a89 | 3 12789 18 |a78 6 25 2 67 3 | 467 45678 4568 | 9 78 1 579 1 689 | 679 26789 68 | 3 4 25 ------------------------+----------------------+--------------------- 4 b59 1-9 | 169 1369 2 |b57 357 8 6 3 7 | 5 48 48 | 2 1 9 8 259 12-9 | 19 139 7 | 4 35 6

(9=7)r4c37 - (7=9)r7c27 => -9 r7c3, r9c3; stte

Phil

by **SteveG48** » Sat Jul 04, 2015 3:32 am

Code: Select all: *--------------------------------------------------------------------* | 179 2679 269 | 8 1567 3 | 156 59 4 | | 17 8 5 | 1467 1467 9 | 16 2 3 | | 3 d69 4 | 2 156 156 | 1568 589 7 | *----------------------+----------------------+----------------------| | 59-7 4 a89 | 3 12789 18 |a78 6 25 | | 2 d67 3 | 467 45678 4568 | 9 8-7 1 | | 579 1 689 | 679 26789 68 | 3 4 25 | *----------------------+----------------------+----------------------| | 4 c59 b19 | 169 1369 2 | 57 357 8 | | 6 3 7 | 5 48 48 | 2 1 9 | | 8 c259 b129 | 19 139 7 | 4 35 6 | *--------------------------------------------------------------------*

(7=89)r4c37 - (9)r79c3 = r79c2 - (9=67)r25c2 => -7 r4c1,r5c8 ; stte

by **JC Van Hay** » Sat Jul 04, 2015 6:45 am

Code: Select all: +--------------------+-------------------+---------------+ | 179 69(27) 9(26) | 8 1567 3 | 156 59 4 | | 17 8 5 | 1467 1467 9 | 16 2 3 | | 3 69 4 | 2 156 156 | 1568 589 7 | +--------------------+-------------------+---------------+ | 579 4 89 | 3 12789 18 | 78 6 25 | | 2 -6(7) 3 | 467 45678 4568 | 9 78 1 | | 579 1 89(6) | 679 26789 68 | 3 4 25 | +--------------------+-------------------+---------------+ | 4 59 19 | 169 1369 2 | 57 357 8 | | 6 3 7 | 5 48 48 | 2 1 9 | | 8 259 129 | 19 139 7 | 4 35 6 | +--------------------+-------------------+---------------+ L3-Wing {7C2 2R1 6C3} - (6=7)r5c2; stte

by **Leren** » Sat Jul 04, 2015 7:09 am

Code: Select all: *---------------------------------------------------------------* | 179 d269-7 c269 | 8 1567 3 | 156 59 4 | | 17 8 5 | 1467 1467 9 | 16 2 3 | | 3 69 4 | 2 156 156 | 1568 589 7 | |---------------------+--------------------+--------------------| | 579 4 89 | 3 12789 18 | 78 6 25 | | 2 a67 3 | 467 45678 4568 | 9 78 1 | | 579 1 b689 | 679 26789 68 | 3 4 25 | |---------------------+--------------------+--------------------| | 4 59 19 | 169 1369 2 | 57 357 8 | | 6 3 7 | 5 48 48 | 2 1 9 | | 8 259 129 | 19 139 7 | 4 35 6 | *---------------------------------------------------------------*

H2 Wing: (7=6) r5c2 - r6c3 = (6-2) r1c3 = (2) r1c2 => - 7 r1c2; stte

Leren

by **bat999** » Sat Jul 04, 2015 2:37 pm

Code: Select all: .-------------------.-------------------.---------------. | a179 2679 j269 | 8 1567 3 | 156 5-9 4 | | 17 8 5 | 1467 1467 9 | 16 2 3 | | 3 69 4 | 2 156 156 | 1568 589 7 | :-------------------+-------------------+---------------: | b579 4 89 | 3 12789 18 | e78 6 25 | | 2 c67 3 | 467 45678 4568 | 9 d78 1 | | b579 1 689 | 679 26789 68 | 3 4 25 | :-------------------+-------------------+---------------: | 4 g59 19 | 169 1369 2 | f57 357 8 | | 6 3 7 | 5 48 48 | 2 1 9 | | 8 259 129 | 19 139 7 | 4 35 6 | '-------------------'-------------------'---------------'

This is my attempt at an AIC...
(9)r1c1=(59)r46c1-(7)r46c1=(7)r5c2-(7)r5c8=(8)r5c8-(8)r4c7=(7)r4c7-(7)r7c7=(5)r7c7-(5)r7c2=(9)r7c2-(9)r13c2=(9)r1c3 => r1c8<>9

Simplified...
(9)r1c1=(59)r46c1-(7)r46c1=(7)r5c2-(7=8)r5c8-(8=7)r4c7-(7=5)r7c7-(5=9)r7c2-(9)r13c2=(9)r1c3 => r1c8<>9
=> - 9 r1c8; stte

Is my AIC OK, or is there a better way with the Eureka notation?

Also, I can't see how this works when the chain is read from right to left, or does it not matter?

Is it good enough to say...
a is either 9 or not 9.
If a is 9 then r1c8 is not 9 (obviously).
And if a is not 9 then j is 9 (through the chain) and r1c8 is not 9 again.

by **SteveG48** » Sat Jul 04, 2015 3:00 pm

bat999 wrote:
Code: Select all
.-------------------.-------------------.---------------. | a179 2679 j269 | 8 1567 3 | 156 5-9 4 | | 17 8 5 | 1467 1467 9 | 16 2 3 | | 3 69 4 | 2 156 156 | 1568 589 7 | :-------------------+-------------------+---------------: | b579 4 89 | 3 12789 18 | e78 6 25 | | 2 c67 3 | 467 45678 4568 | 9 d78 1 | | b579 1 689 | 679 26789 68 | 3 4 25 | :-------------------+-------------------+---------------: | 4 g59 19 | 169 1369 2 | f57 357 8 | | 6 3 7 | 5 48 48 | 2 1 9 | | 8 259 129 | 19 139 7 | 4 35 6 | '-------------------'-------------------'---------------'

This is my attempt at an AIC...
(9)r1c1=(59)r46c1-(7)r46c1=(7)r5c2-(7)r5c8=(8)r5c8-(8)r4c7=(7)r4c7-(7)r7c7=(5)r7c7-(5)r7c2=(9)r7c2-(9)r13c2=(9)r1c3 => r1c8<>9

Simplified...
(9)r1c1=(59)r46c1-(7)r46c1=(7)r5c2-(7=8)r5c8-(8=7)r4c7-(7=5)r7c7-(5=9)r7c2-(9)r13c2=(9)r1c3 => r1c8<>9
=> - 9 r1c8; stte

Is my AIC OK, or is there a better way with the Eureka notation?

Also, I can't see how this works when the chain is read from right to left, or does it not matter?
Is it good enough to say...
a is either 9 or not 9.
If a is 9 then r1c8 is not 9 (obviously).
And if a is not 9 then j is 9 (through the chain) and r1c8 is not 9 again.

Pretty good, Bat. I would make some minor changes:

(9=17)r12c1-(7)r46c1=(7)r5c2-(7=8)r5c8-(8=7)r4c7-(7=5)r7c7-(5=9)r7c2-(9)r13c2=(9)r1c13 => r1c8<>9

In the last term, I understand your reasoning for writing (9)r1c3 rather than (9)r1c13 because of the first term. However, I think it's better to write c13 as being the actual logical conclusion from the preceding term.

As far as reading it, your description in just fine.

by **bat999** » Sat Jul 04, 2015 3:31 pm

SteveG48 wrote:... I would make some minor changes...

Thanks Steve

(9=17)r12c1-...
OK, I understand this.
not 9 makes these two the 17 pair.

But this...
=(9)r1c13...
It seems to say "makes either r1c1 or r1c3 a 9".
(The logical conclusion is that 9 is on row 1 of box 1, but not in column 2)
Then I need to think to myself..."but it can't be r1c1 because we already set it to not 9 at the start of the chain so it must be the other one"... => r1c3=9
8-)

by **daj95376** » Sat Jul 04, 2015 3:38 pm

Hello bat999,

Congratulations on learning to use Eureka notation. I'll do what I can to help. Unfortunately, some of my comments won't agree with what SteveG48 wrote.

Starting at your simplified chain, you can: merge the interaction in r46c1, ignore <8> in [box 6], drop repeated values in simple links, and drop parens around single values:

Code: Select all: (9)r1c1=(59)r46c1-(7)r46c1=(7)r5c2-(7=8)r5c8-(8=7)r4c7-(7=5)r7c7-(5=9)r7c2-(9)r13c2=(9)r1c3 => -9 r1c8; stte becomes 9r1c1 = (59-7)r46c1 = r5c2 - r5c8 = r4c7 - (7=5)r7c7 - (5=9)r7c2 - r13c2 = 9r1c13 => -9 r1c8; stte

Your original chain is really a network because you "remember" the initial assumption of -9r1c1 to derive =9r1c3. That's why you can't make sense of it when reading from r-to-l. To correct this, you need to include r1c1 in your last term. This is now a bidirectional AIC instead of a network.

To be technically correct, the term (95-7)r46c1 should be (95-57)r46c1; but convention has resulted in it being shortened and accepted.

Note: the term 9r1c13 is correct, but it may need explaining. What's happened is that <9> has been eliminated in r13c2 of [box 1]. What's left is <9> in r1c13. This is perfectly acceptable logic. It just takes a little mental gymnastics to follow it when reading r-to-l. As far as I'm concerned, an AIC may be technically bidirectional, but the Eureka notation only needs to make sense when reading from l-to-r.

Personally, I'd shorten your chain and use:

Code: Select all: 9r1c1 = (59-7)r46c1 = r5c2 - r5c8 = r4c7 - (7=5)r7c7 - (5=9)r7c2 => -9 r13c2; lclste

BTW: Am I the only one whose/who's bothered by the illogic of using stte in one instance and using lclste in another???

_

by **bat999** » Sat Jul 04, 2015 4:31 pm

daj95376 wrote:... I'll do what I can to help...

Code: Select all: 9r1c1 = (59-7)r46c1 = r5c2 - r5c8 = r4c7 - (7=5)r7c7 - (5=9)r7c2 - r13c2 = 9r1c13 => -9 r1c8; stte

"As far as I'm concerned, an AIC may be technically bidirectional, but the Eureka notation only needs to make sense when reading from l-to-r."

I'm not sure about the "mental gymnastics".
Not sure how to read this from right to left.
Ideally I would like to say "If r1c3 is not 9 then r1c1 is 9"

But ...r13c2 = 9r1c13 from r-to-l seems to try and say something like
"If r1c1 and r1c3 are not 9 then r1c1 is 9"
or maybe
"If r1c1 or r1c3 are not 9 then r1c1 is 9"
I haven't figured it out (yet).

So I will consider it to be a uni-directional AIC (or network, as you said), with my previous thoughts...
a is either 9 or not 9.
If a is 9 then r1c8 is not 9 (obviously).
And if a is not 9 then j is 9 (through the chain) and r1c8 is not 9 again.
Say nothing and let the readers figure it out for themselves. 8-)

by **David P Bird** » Sat Jul 04, 2015 4:48 pm

Bat,
When first mastering AICs read the chain from left to right
If you started with a strong link assume the first argument is false and track the consequences, when normally you'll find the last argument must be true.
Now read the chain backwards assuming the last argument is false when you should find the first one must be true.
For a chain that starts and ends with a strong link this proves that the two end arguments can't both be false.
However it doesn't prove that one is true and the other is false – they could both be true.

Less useful are chains that start and end with weak links. Now you have to read them assuming the terminal arguments are true.
This proves that at least one of them must be false (but similarly they could both be false).

With simple strong links using bi-value cells and bi-local digits you shouldn't need to check your chains after a while, but when you start using more complicated links it's useful to check that the inferences work in both directions using this method.

Authors who try to shorten their notations as much as possible shift the burden over to their readers who will need to take more time to understand the logic, and will find it harder to focus alternately on the chain and the grid. This will affect beginners more than most, but it's funny how quickly contributors forget the difficulties they had when they first started. Making a notation as terse as possible also increases the chances a logical error will be missed when it's being written. Having spent time trying to find a deduction, shortcutting the notation makes little sense to me, it's not as if we're saving paper here!

Like others here you have shown yourself to be strong willed and it's up to you how much of this you take on board. To a large extent it depends on a) what your ambitions are in improving your solving skills, & b) how much you would like to encourage others to join in the fun.

DPB

by **bat999** » Sat Jul 04, 2015 5:15 pm

David P Bird wrote:...Authors who try to shorten their notations as much as possible shift the burden over to their readers who will need to take more time to understand the logic, and will find it harder to focus alternately on the chain and the grid. This will affect beginners more than most, but it's funny how quickly contributors forget the difficulties they had when they first started. Making a notation as terse as possible also increases the chances a logical error will be missed when it's being written. Having spent time trying to find a deduction, shortcutting the notation makes little sense to me, it's not as if we're saving paper here!

Hi
Yes, I'll think about these tips..."merge the interaction in r46c1, ignore <8> in [box 6], drop repeated values in simple links, and drop parens around single values".
Will introduce them when I'm more confident.
Baby steps.
:lol:

by **DonM** » Sat Jul 04, 2015 10:18 pm

daj95376 wrote:BTW: Am I the only one whose/who's bothered by the illogic of using stte in one instance and using lclste in another???

I have never quite understood how or why the only objective in these threads has become the aim to achieve a single-stepper along with the apparent exclusion of any attempts at increasing one's skillset by solving more difficult puzzles.

Still, if the challenge is a single-stepper, then one is kidding one's self if they think an lclste solution is equivalent to a stte solution (assuming that the acronym's are being used correctly).

by **Ngisa** » Sat Jul 04, 2015 10:23 pm

Code: Select all: +--------------+-----------------+-------------+ | 17-9 2679 d269 | 8 1567 3 | 156 a59 4 | | 17 8 5 | 1467 1467 9 | 16 2 3 | | 3 d69 4 | 2 156 156 | 1568 589 7 | +--------------+-----------------+-------------+ | 579 4 89 | 3 12789 18 | 78 6 25 | | 2 67 3 | 467 45678 4568 | 9 78 1 | | 579 1 689 | 679 26789 68 | 3 4 25 | +--------------+-----------------+-------------+ | 4 59 19 | 169 1369 2 | 57 357 8 | | 6 3 7 | 5 48 48 | 2 1 9 | | 8 259 c129 | c19 c139 7 | 4 b35 6 | +--------------+-----------------+-------------+

(9=5)r1c8-(5=3)r9c8-(3=192)r9c345-(2=69)r1c3,r3c2 => -9r1c1; lclste

The New Sudoku Players' Forum

July 4, 2015

July 4, 2015

Re: July 4, 2015

Re: July 4, 2015

Re: July 4, 2015

Re: July 4, 2015

Re: July 4, 2015

Re: July 4, 2015

Re: July 4, 2015

Re: July 4, 2015

Re: July 4, 2015

Re: July 4, 2015

Re: July 4, 2015

Re: July 4, 2015

Re: July 4, 2015

Re: July 4, 2015