View previous topic :: View next topic |
Author |
Message |
| Puzzler
| Joined: 04 May 2009 | Posts: 36 | : | | Items |
|
Posted: Tue May 19, 2009 2:14 pm Post subject: More confusion |
|
|
In the second example, the author identifies {R9, C6} as an exception of {R2,C5} and thus painted Green, but this is also an exception of {R1,C4}, so couldn't it also be painted Blue? this makes a difference in which cells are identified as exceptions in later iterations.
|
|
Back to top |
|
|
| Lunatic
| Joined: 11 Mar 2007 | Posts: 166 | : | Location: Ghent - Belgium | Items |
|
Posted: Tue May 19, 2009 3:43 pm Post subject: |
|
|
This is the rule for the Exception Cell:
If all but one and only one cell of all the candidates in a house are peers of cells colored with the same color, then this Exception cell can be colored with the SAME COLOR.
From the view of cell {R9, C6}, ALL other candidate cells - in this case there's only one other candidate cell {R9, C5} - within it's house peers to cells ALL colored GREEN (cell {R9, C5} peers with GREEN cell {R2, C5}), therefore {R9, C6} can be colored GREEN.
There is no way that cell {R9, C5} peers to the BLUE cell {R1, C4}, so cell {R9, C6} certainly cannot be painted BLUE as well. _________________ Marc
~~~<><~~~<><~~~<><~~~<><~~~ |
|
Back to top |
|
|
| daj95376
| Joined: 05 Feb 2006 | Posts: 349 | : | | Items |
|
Posted: Tue May 19, 2009 4:49 pm Post subject: |
|
|
Puzzler,
It would appear that you are confused on the terminology and, more importantly, on the concepts. It doesn't help that Example #2 is overly complicated.
*) Cell [r1c4] is colored Blue.
*) Cell [r2c5] is colored Green because it is a conjugate cell in [b2].
*) Cell [r4c4] is colored Green because it is a conjugate cell in [c4].
*) There are three cells -- [r124c7] -- in [c7] for the candidate.
*) Cell [r2c7] sees Green cell [r2c5].
*) Cell [r4c7] sees Green cell [r4c4].
*) Cell [r1c7] is colored Green.
*) Candidate cells that see Blue cell [r1c4] and Green cell [r1c7] can have the candidate eliminated. Thus, [r1c13]<>6 follows. |
|
Back to top |
|
|
| Puzzler
| Joined: 04 May 2009 | Posts: 36 | : | | Items |
|
Posted: Tue May 19, 2009 4:59 pm Post subject: Ok. I think! |
|
|
OK, I think i get it.
When analyzing {R1,C4} I can't eliminate {R9, C5} just because its in the same row as a cell of ANY color (In this case {R2,C5}, Green)
When analyzing {R2,C5} I CAN eliminate {R9, C5} because its in the same row as a cell that is GREEN {R2,C5}
Next dumb question
In sample 6 the grid contains two chains - the first of which creates the collision that the author explains.
In the second chain of {R9,C6}--> {R9,C9} there appears to be a Trap at {R7,C7}.
Can I safely eliminate this cell, even though there is a prior collision?
|
|
Back to top |
|
|
| daj95376
| Joined: 05 Feb 2006 | Posts: 349 | : | | Items |
|
Posted: Tue May 19, 2009 5:20 pm Post subject: Re: Ok. I think! |
|
|
Puzzler wrote: | In sample 6 the grid contains two chains - the first of which creates the collision that the author explains.
In the second chain of {R9,C6}--> {R9,C9} there appears to be a Trap at {R7,C7}.
Can I safely eliminate this cell, even though there is a prior collision?
|
Quite often, an example may contain a number of single-digit elimination patterns. I can't make a determination based on your description.
I will say that there is a Skyscraper in [r2] and [r9] for this candidate that leads to [r7c7]<>1. It can be interpreted as X-Colors as well.
More importantly, the Skyscraper is easier than X-Colors to use, and it produces the same final results in Example #6. |
|
Back to top |
|
|
| Lunatic
| Joined: 11 Mar 2007 | Posts: 166 | : | Location: Ghent - Belgium | Items |
|
Posted: Tue May 19, 2009 5:21 pm Post subject: Re: Ok. I think! |
|
|
@Puzzler
I see that you've tried to implement the links for the pictures at Sudopedia, but that it's not working properly...
You must 'click' on the picture first, and then copy the link at the bottom of the picture from that page...
This is example 6:
_________________ Marc
~~~<><~~~<><~~~<><~~~<><~~~ |
|
Back to top |
|
|
| daj95376
| Joined: 05 Feb 2006 | Posts: 349 | : | | Items |
|
Posted: Tue May 19, 2009 5:30 pm Post subject: |
|
|
It would be nice if the cell labels matched the steps listed in the explanation. Then we'd know:
* Which conjugate cells went with Step (1)
* Which cell colors were derived through conjugates in step (2)
* That no cell colors were derived in step (3), and finally
* That step (4.3) was used for the conclusion
Other than the above, a perfectly understandable example. |
|
Back to top |
|
|
| Lunatic
| Joined: 11 Mar 2007 | Posts: 166 | : | Location: Ghent - Belgium | Items |
|
Posted: Tue May 19, 2009 5:32 pm Post subject: |
|
|
daj95376 wrote: | More importantly, the Skyscraper is easier than X-Colors to use, and it produces the same final results in Example #6. |
So is the Two-String Kite.
In example 1 cell {r2,c8} will also loose candidate 2 based upon the Two-String Kite principle.
The Kite is the 'bottom left box' and the two strings are 'column 1' and 'row 9'. Any cell that can see both string end cells, {r2,c1} and {r9,c8}, looses the candidate.
Example 1:
_________________ Marc
~~~<><~~~<><~~~<><~~~<><~~~ |
|
Back to top |
|
|
| Puzzler
| Joined: 04 May 2009 | Posts: 36 | : | | Items |
|
Posted: Tue May 19, 2009 6:07 pm Post subject: Ah, much better! |
|
|
Ah, much better to be able to see the example! Thanks Marc
In the sample above, the grid contains two chains - the first of which creates the collision that the author explains, and the cells in this first chain are the ones the author highlights.
There is a second conjugate chain in the bottom right corner, consisting of {R9,C6}--> {R9,C9}.
Using the X-Colors rules (as I understand them!) there appears to be a trap at {R7,C7}. ({R7,C7} is in house 9 which contains a Blue Cell {R9,C9}, and in the same column as {R2,C7} which is an exception of {R3, C4}, which is itself an exception of {R9,C6}, which is the conjugate of {R9,C9} (the blue cell!)
In this particular case, it is valid to remove the trapped cell, but can I make the assumption that it is always safe to remove a Trapped Cell from a second chain, even though there is a collision in another chain?
Thx
PS. Is there any way to post graphics in a message? It would be so much easier for you to see what I'm talking about if I could post a screenprint. (I don't have a website to upload the images to, otherwise I would) |
|
Back to top |
|
|
| Lunatic
| Joined: 11 Mar 2007 | Posts: 166 | : | Location: Ghent - Belgium | Items |
|
Posted: Tue May 19, 2009 9:18 pm Post subject: |
|
|
Puzzler wrote: | PS. Is there any way to post graphics in a message? It would be so much easier for you to see what I'm talking about if I could post a screenprint. (I don't have a website to upload the images to, otherwise I would) |
Not that I'm aware off. I reserved a folder on my website for my picture's I use in this forum. _________________ Marc
~~~<><~~~<><~~~<><~~~<><~~~ |
|
Back to top |
|
|
| strmckr
| Joined: 24 Apr 2009 | Posts: 52 | : | | Items |
|
Posted: Tue May 19, 2009 10:06 pm Post subject: |
|
|
muti coloring combins several move typs into a single application.
i find it covers these three things
finned / sashimi: x-wings
2-string kytes
empty rectangles
two string kyte.
uses
(2)R1C2 = (2)R8C1 - (2)R9C2= (2)R9C8 => R2C8 <>2
two rows with only 2 candidates intersecting in a box.
Code: | / N / | / N / | / / /
N . . | . . . | . . .
/ . . | . . . | . . .
--------+-------+-------
/ . . | . . . | . . .
N . . | . X . | . . .
/ . . | . . . | . . .
--------+-------+-------
/ . . | . . . | . . .
/ . . | . . . | . . .
/ . . | . . . | . . . |
KEY: '/' <=> cells void of candidate 'N'
'N' <=> cells with candidate 'N'; all are required
"-X" <=> potential eliminations of candidate 'N'
or
they can be like this example
Code: | / N # | / N / | / / /
N . . | . . . | . . .
# . . | . . . | . . .
--------+-------+-------
/ . . | . . . | . . .
N . . | . X . | . . .
/ . . | . . . | . . .
--------+-------+-------
/ . . | . . . | . . .
/ . . | . . . | . . .
/ . . | . . . | . . . |
where # marks cells that could hold a fin candidate.
the elimination is the same.
the move your picture that is represetning a kyte
is an empty rectangle as well.
{worded correctly for a kyte but the representation in color looks odd} {i would have hilighted the cells individually}
but it is also shows exactly where to find the er.
Code: | / N / | . . . | . / .
N N N | . . . | . X .
/ N / | . . . | . / .
--------+-------+-------
. . . | . . . | . / .
. . . | . . . | . / .
. . . | . . . | . / .
--------+-------+-------
. . . | . . . | . / .
. N . | . . . | . N .
. . . | . . . | . / . |
KEY: '/' <=> cells void of candidate 'N'
'N' <=> cells with candidate 'N'; not all are required in the box. {min is 2}
"-X" <potential> R8C6 <1> R7C4 <1> R8C6 <1> R7C4 <1> r8c6 <1> r7c4 <>1 |
|
Back to top |
|
|
| strmckr
| Joined: 24 Apr 2009 | Posts: 52 | : | | Items |
|
Posted: Wed May 20, 2009 1:50 am Post subject: |
|
|
http://www.setbb.com/sudoku/viewtopic.php?p=11153&mforum=sudoku#11153
in this picture there is two
2-string kytes:
Code: |
R2C7 = R2C6 - R3C4 = R8C4 => R8C7 <>1
other chain won't write... =>R7C4 <>1
|
edit: this is glitchign alot for me.
can any one else replicate the error?
i cannot post the second chain at all.
ive tried writign it in diffrent orders as well..
not a clue on it.
end edit;
the cells that see 2 cover green cells can be safely removed.
only when they can see a green cell that is from the opposit blue diretion.
thats
these 2 cells specifically.
R8C6, R7C4
there is also removed as well.{but not applicable in the examle}
R3C7,r2c8
Last edited by strmckr on Wed May 20, 2009 7:01 am; edited 1 time in total |
|
Back to top |
|
|
| daj95376
| Joined: 05 Feb 2006 | Posts: 349 | : | | Items |
|
Posted: Wed May 20, 2009 4:03 am Post subject: |
|
|
strmckr: I don't follow the specifics of your complaint ... but I have a guess.
1) Try checking the box for Disable HTML in this post
2) Try checking the box for Disable Smilies in this post
If one of these doesn't solve your problem, then maybe someone else will have a solution. |
|
Back to top |
|
|
| strmckr
| Joined: 24 Apr 2009 | Posts: 52 | : | | Items |
|
Posted: Wed May 20, 2009 6:28 am Post subject: |
|
|
when i type this in with normal stuff enabled it casues glitches to appear in the post..
[code]
R2C7 = R2C6 - R3C4 = R8C4 => R8C7 <>1
R3C4 = R3C8 - R2C7= R7C7 => R7C4 <>1
[/code]
and it changes the post as well.
when you go to edit it.
i'll add a reply below with the stuff enabled.
Last edited by strmckr on Wed May 20, 2009 7:00 am; edited 2 times in total |
|
Back to top |
|
|
| strmckr
| Joined: 24 Apr 2009 | Posts: 52 | : | | Items |
|
Posted: Wed May 20, 2009 6:30 am Post subject: |
|
|
Code: | R2C7 = R2C6 - R3C4 = R8C4 => R8C7 <1> R7C4 <>1 |
|
|
Back to top |
|
|
|