The New Sudoku Players' Forum

[ArkieTech]

Code: Select all

 *-----------*
 |...|.2.|...|
 |.1.|.4.|..3|
 |7..|9..|.62|
 |---+---+---|
 |..7|..4|...|
 |..9|538|4..|
 |...|7..|5..|
 |---+---+---|
 |58.|..1|..9|
 |9..|.7.|.1.|
 |...|.8.|...|
 *-----------*


Play/Print this puzzle online

[eleven]

Code: Select all

+----------------------+----------------------+----------------------+
| 36     9      356    | 1      2      67     |d78     4578   4578   |
| 26     1      256    | 8      4      67     | 79     579    3      |
| 7      4      8      | 9      5      3      | 1      6      2      |
+----------------------+----------------------+----------------------+
| 1238   5      7      | 6      19     4      | 2389   2389  a18     |
| 126    26     9      | 5      3      8      | 4      27    a167    |
| 13468  36     1346   | 7      19     2      | 5      389   a168    |
+----------------------+----------------------+----------------------+
| 5      8      234    | 234    6      1      | 237    2347   9      |
| 9      236    2346   | 234    7      5      |c2368   1     b48     |
| 12346  7      12346  | 234    8      9      | 236    2345   45     |
+----------------------+----------------------+----------------------+

(7=8)r456c9-r8c9=r8c7-(8=7)r1c7 => r1c9<>7

[SteveG48]

Code: Select all

 *---------------------------------------------------------------------*
 | 36     9      356    | 1      2      67     |ce78    c4578  b458-7  |
 | 26     1      256    | 8      4      67     | c79    c579    3      |
 | 7      4      8      | 9      5      3      |  1      6      2      |
 *----------------------+----------------------+-----------------------|
 | 1238   5      7      | 6      19     4      |  2389   2389   18     |
 | 126    26     9      | 5      3      8      |  4      27    a167    |
 | 13468  36     1346   | 7      19     2      |  5      389    168    |
 *----------------------+----------------------+-----------------------|
 | 5      8      234    | 234    6      1      | d237   d2347   9      |
 | 9      236    2346   | 234    7      5      | d2368   1      48     |
 | 12346  7      12346  | 234    8      9      | d236   d2345   45     |
 *---------------------------------------------------------------------*


7r5c9 = r1c9 - (7=8945)b3p1245 - (45=23678)b9p12478 - (8=7)r1c7 => -7 r1c9 ; stte

[pjb]

Code: Select all

 36      9       356    | 1      2      67     | 78    a4578  b4578   
 26      1       256    | 8      4      67     | 79     579    3     
 7       4       8      | 9      5      3      | 1      6      2     
------------------------+----------------------+---------------------
*1238    5       7      | 6      19     4      | 2389  *2389   1-8     
 126     26      9      | 5      3      8      | 4      27     167   
*13468   36      1346   | 7      19     2      | 5     *389    16-8   
------------------------+----------------------+---------------------
 5       8       234    | 234    6      1      | 237    2347   9     
 9       236     2346   | 234    7      5      | 2368   1     c48     
 12346   7       12346  | 234    8      9      | 236    2345   45     


Kraken x-wing of 8s at r46c18, fin at r1c8: (8-4)r1c8 = r1c9 - (4=8)r8c9 => -8 r46c9; stte

Phil

[Leren]

Code: Select all

*--------------------------------------------------------------*
| 36    9     356    | 1     2     67     | 78   c4578C 4578B  |
| 26    1     256    | 8     4     67     | 79    579   3      |
| 7     4     8      | 9     5     3      | 1     6     2      |
|--------------------+--------------------+--------------------|
| 1238  5     7      | 6     19    4      | 2389 d2389D 1-8    |
| 126   26    9      | 5     3     8      | 4     27    167    |
| 13468 36    1346   | 7     19    2      | 5    d389D  16-8   |
|--------------------+--------------------+--------------------|
| 5     8     234    | 234   6     1      | 237  b2347  9      |
| 9     236   2346   | 234   7     5      | 2368  1    a48A    |
| 12346 7     12346  | 234   8     9      | 236  b2345  45     |
*--------------------------------------------------------------*

M Wing Type 7A : (8=4) r8c9 - r79c8 = (4-8) r1c8 = r46c8 => - 8 r46c9; stte, or

M Wing Type 7B : (8=4) r8c9 - r1c9  = (4-8) r1c8 = r46c8 => - 8 r46c9; stte

Leren

[bat999]

Code: Select all

.-------------------.-------------.---------------------.
| 36     9    356   | 1    2   67 | c78     4578   4578 |
| 26     1    256   | 8    4   67 |  79     579    3    |
| 7      4    8     | 9    5   3  |  1      6      2    |
:-------------------+-------------+---------------------:
| 1238   5    7     | 6    19  4  |  2389   2389  f18   |
| 126    26   9     | 5    3   8  |  4      2-7   f167  |
| 13468  36   1346  | 7    19  2  |  5      389   f168  |
:-------------------+-------------+---------------------:
| 5      8    234   | 234  6   1  | b237   a2347   9    |
| 9      236  2346  | 234  7   5  | d2368   1     e48   |
| 12346  7    12346 | 234  8   9  |  236    2345   45   |
'-------------------'-------------'---------------------'

(7)r7c8 = r7c7 - (7=8)r1c7 - r8c7 = r8c9 - (8=7)r456c9 => -7 r5c8; stte

[bat999]

...(7=8)r456c9-r8c9=r8c7-(8=7)r1c7 => r1c9<>7

Hi
Not meant as a criticism, only a commment.

imho this is another of those fake dubious pincer attacks.

Trivial stream
(7)r5c9 - (7)r1c9
Second stream
(7=8)r456c9 - r8c9 = r8c7 - (8=7)r1c7 - (7)r1c9
=> r1c9<>7

But stop and look...
(7=8)r456c9 - r8c9 = r8c7 - (8=7)r1c7 - (7)r1c9
There are now no sevens in column 9.
Alarm bells are ringing.
Somebody screams "omg, if we remove that 7 from r5c9 we won't be able to complete the puzzle!"
=> r5c9=7

imho in cases like this it's appropriate (for me) to use discontinuity/contradiction instead of a pincer.
It tells us that r5c9=7, not just that r1c9<>7.

Code: Select all

.-------------------.-------------.---------------------.
| 36     9    356   | 1    2   67 | c78    4578  d458-7 |
| 26     1    256   | 8    4   67 |  79    579    3     |
| 7      4    8     | 9    5   3  |  1     6      2     |
:-------------------+-------------+---------------------:
| 1238   5    7     | 6    19  4  |  2389  2389  a18    |
| 126    26   9     | 5    3   8  |  4     2-7   a7-16  |
| 13468  36   1346  | 7    19  2  |  5     389   a168   |
:-------------------+-------------+---------------------:
| 5      8    234   | 234  6   1  |  237   2347   9     |
| 9      236  2346  | 234  7   5  | b2368  1     a48    |
| 12346  7    12346 | 234  8   9  |  236   2345   45    |
'-------------------'-------------'---------------------'

(7=8)r456c9 - r8c9 = r8c7 - (8=7)r1c7 - (7)r1c9 = (7)r5c9 => -16 r5c9, -7 r1c9,r5c8; stte

[eleven]

Bat, have you read David's nice comment on this topic ?

However it is the players who explain their eliminations using Nice Loop logic and terms while using AIC notation conventions who do the community a great disservice. In particular a 'continuous loop' in AIC terminology seems to be a 'type 2 discontinuous loop' in NL terminology. The confusion this mix and match approach causes is further compounded by the style issues regarding the best way to notate different situations which affects both systems.

Please don't mix "Nice loop thinking" and "AIC thinking". And, even worse, don't mix "Nice loop thinking" and AIC writing.

My way is to use logical thinking. And then write it in any notation, hopefully understood by the readers.

I never read the NL definitions, i did not know, what type it was, but i never had a problem to understand it (at least a correct one), when looking at the grid.

FYI
Standard AIC's with alternating strong and weak links, starting and ending with a strong one, are OR connections. All you see on the left side of "=" is in OR relation to all on the right side of '='. That's probably the property, on which, what you call "pincer attack", is based.
In this case (7=8)r456c9 - r8c9 = r8c7 - (8=7)r1c7 gives you the OR relation 7r456c9=7r1c7. Since the 7 must be here or there, it can't be in all cells, which see both, i.e. r123c9 and r456c8. So you can eliminate it there. This logical deduction is not notated in AIC (it is trivial here, less trivial in other cases). You don't need a contradiction or a closed loop.

[bat999]

...Please don't mix "Nice loop thinking" and "AIC thinking". And, even worse, don't mix "Nice loop thinking" and AIC writing...

Aha
This has been discussed before.
People see Eureka notation and assume it's an AIC (probably because many of the posts on this forum use AICs in their solutions).

But here it's merely a non-AIC chain being used to arrive at a conclusion.
So I will make it clear.
(7=8)r456c9 - r8c9 = r8c7 - (8=7)r1c7 - (7)r1c9 = (7)r5c9 -loop
=> -16 r5c9, -7 r1c9,r5c8; stte

[eleven]

Aha, so you know better, what an AIC chain and a "non-AIC chain" and an AIC loop is than all the others ?
Please explain them (i don't mind as long as i can understand it).
Or do you now have your own notation, which looks like AIC and is a valid AIC, but for you it is "non-AIC" ? You like to make simple things complicated.

Hm, maybe you misunderstood, that i said "This logical deduction is not notated in AIC". It should say, you don't need to write it (you can add unnecessary alternating links as much as you want, it always will be a valid AIC).

How would write a skyscraper ?
1r1c1=r1c7-r4c7=r4c2 => r123c2,r456c1<>1

[bat999]

...what an AIC chain and a "non-AIC chain" and an AIC loop is than all the others ?Please explain them...

No, I'm not gonna touch that third rail.

But you said:
"Standard AIC's with alternating strong and weak links, starting and ending with a strong one, are OR connections".
I agree.
But that property is not being used here (to eliminate the 7 in r1c9).
So don't read it as an AIC (even if you think it might qualify as one), just a chain being used to arrive at a conclusion (to keep the 7 in r5c9).

[eleven]

But that property is not being used here (to eliminate the 7 in r1c9).
So don't read it as an AIC (even if you think it might qualify as one), just a chain being used to arrive at a conclusion (to keep the 7 in r5c9).

My last try: This property IS used through your whole chain, and the result is: 7r5c9 OR 7r5c9 =>7r5c9. This the logical reason, that you can "keep the 7 in r5c9".
E.g. a chain 1r1c1-(1=2)r1c2-(2=1)r1c1 does NOT allow you to keep the 1 in r1c1.

[bat999]

My last try...

Yes, I understand where you're coming from now.

You have read it as an AIC and considered the logic.
(7=8)r456c9 ..... = (7)r5c9
7r5c9 OR 7r5c9 =>7r5c9

I have read it as a chain and considered the contradiction.
(7=8)r456c9 ..... = (7)r5c9
Not 7 in r5c9 forces 7 in r5c9 =>7r5c9.

[blue]

...Please don't mix "Nice loop thinking" and "AIC thinking". And, even worse, don't mix "Nice loop thinking" and AIC writing...

And above all, please (please please) don't master "Nice loop writing" (using "Nice loop notation"), and use it here, in an attempt to avoid criticism.

[DonM]

...Please don't mix "Nice loop thinking" and "AIC thinking". And, even worse, don't mix "Nice loop thinking" and AIC writing...

Aha
This has been discussed before.
People see Eureka notation and assume it's an AIC (probably because many of the posts on this forum use AICs in their solutions).

But here it's merely a non-AIC chain being used to arrive at a conclusion.
So I will make it clear.
(7=8)r456c9 - r8c9 = r8c7 - (8=7)r1c7 - (7)r1c9 = (7)r5c9 -loop
=> -16 r5c9, -7 r1c9,r5c8; stte

Eleven has been giving great guidance and advice, but, in the end, it looks like you should go back to the post that defined and started the use of AICs. Please note the quote from that article, which pretty much says that there are both discontinuous AICs and continuous AIC loops. When using these defined terms to find eliminations, there is no scenario under which one can use Eureka notation, use alternating inferences (of course) and say that non-AIC chains are involved.

I understand that, semantically, you may be saying that this is a loop and not a chain (or maybe not, I can't really tell what your point is), but an AIC is an Alternating Inference Chain regardless of the context or semantics. For whatever reason, a term 'AIL' (Alternating Inference Loop) never came into use...

http://forum.enjoysudoku.com/alternating-inference-chains-t3865.html

"Deductions
Quite simply, at least one or the other (possibly both) of the two endpoint candidates (or candidate premises) of an AIC is true. Any deductions that you can make based on that are valid. This tends to produce the best results if the endpoints either share a group, or if the endpoints involve the same candidate. When your chain endpoints satisfy one of those conditions, it is time to check for any deductions.

If the two endpoints candidates are weakly linked, then you have an AIC loop. In this case, you could cut the loop at any weak link and end up with a valid AIC. Thus, for every weak link in the loop, either one or the other of the candidates joined by that weak inference are true, and you can make all appropriate deductions based on that.

That is pretty much all you need.: