The New Sudoku Players' Forum

by **ArkieTech** » Sat May 24, 2008 8:54 pm

Sorry about that I had the wrong puzzle. It wasn't the one-trick pony it was the brain-basher. Can anyone see an easier solution?

I found 2 xy-wings advancing the puzzle to where a third xy-wing broke it.

Code: Select all: 4...3...1..2.7.4...9.6.4.7...93.57..76.....93..41.78...4.7.3.6...6.8.2..9...5...7 *-----------------------------------------------------------* | 4 58 7 | 2589 3 89-2 | 6 258 1 | | 6 1358 2 | 58 7 *18 | 4 358 9 | | 158 9 1358 | 6 *12 4 | 35 7 258 | |-------------------+-------------------+-------------------| | 128 128 9 | 3 6 5 | 7 124 ^24 | | 7 6 15 | 28 4 *28 | 15 9 3 | | 235 235 4 | 1 9 7 | 8 ^25 6 | |-------------------+-------------------+-------------------| | 1258 4 158 | 7 12 3 | 9 6 58 | | 135 7 6 | 49 8 19 | 2 134-5^45 | | 9 128 138 | 24 5 6 | 13 148 7 | *-----------------------------------------------------------* xy-wings *-----------------------------------------------------------* | 4 58 7 | 29 3 89 | 6 258 1 | | 6 138 2 | 5 7 *18 | 4 3-8 9 | | 158 9 1358 | 6 *12 4 | 35 7 *28 | |-------------------+-------------------+-------------------| | 128 128 9 | 3 6 5 | 7 124 24 | | 7 6 15 | 8 4 2 | 15 9 3 | | 235 235 4 | 1 9 7 | 8 25 6 | |-------------------+-------------------+-------------------| | 1258 4 158 | 7 12 3 | 9 6 58 | | 135 7 6 | 49 8 19 | 2 134 45 | | 9 128 138 | 24 5 6 | 13 148 7 | *-----------------------------------------------------------* xy-wing

dan

by **wintder** » Sat May 24, 2008 11:01 pm

Original puzzle is the only way to try.

by **StrmCkr** » Sun May 25, 2008 6:16 am

i found a move... not easier by any means but its there..

Line box reduction + singles brings it to here. (just befor simple sudoku does an x wing from starting off.)

Code: Select all: *-----------------------------------------------------------* | 4 5@-8 7 | 29(8)@-5 3 89@-2 | 6 28@-5 1 | | 6 13 (8)@-5 2 | 5@-8 7 18@ | 4 38@-5 9 | | 1358 9 1358 | 6 12@ 4 | 35@ 7 258@ | |-------------------+-------------------+-------------------| | 128 128 9 | 3 6 5 | 7 124 24 | | 7 6 15 | 28 4 28@ | 15 9 3 | | 235 235 4 | 1 9 7 | 8 25 6 | |-------------------+-------------------+-------------------| | 1258 4 158 | 7 12@ 3 | 9 6 58 | | 135 7 6 | 49 8 19@ | 2 1345 45 | | 9 1238 138 | 24 5 6 | 13 1348 7 | *-----------------------------------------------------------*

there is a ALmost Unique rectangle of 58's connected to an als/als chain

{the chain is some what like this i can't write it out nicely... just plug 1 and the 2 into R3C4 and note the effects. }

(1289) R1259C6 = (1=2)R37C5 - (5=8)R1C2,R2C4 = (35)R3C7=(358) =(358)R2C8+(258)R3C8

=> R1C4, R2C2, R2C8,R1C9,R3C13 <>5 & R1C6<>2, R1C2,R2C4<>8,

brings it to here

Code: Select all: *-----------------------------------------------------------* | 4 5 7 | 29 3 89 | 6 28@ 1 | | 6 138 2 | 5 7 18 | 4 38@ 9 | | 138 9 138 | 6 12 4 | 35@ 7 258 | |-------------------+-------------------+-------------------| | 128 128 9 | 3 6 5 | 7 124 24 | | 7 6 15@ | 8 4 2 | 15@ 9 3 | | 235 23 4 | 1 9 7 | 8 25@ 6 | |-------------------+-------------------+-------------------| | 1258 4 158 | 7 12 3 | 9 6 58 | | 135 7 6 | 49 8 19 | 2 1345 45 | | 9 1238 138 | 24 5 6 | 13 1348 7 | *-----------------------------------------------------------*

next i use an aic
(2=8)R1C8-(8=3)R2C8-(3=5)R3C7-(1=5)R5C7-(2=5)R6C8- (5=1)R4C3 = > R6C1 <>5

the rest
solves as singles.

by **Carcul** » Sun May 25, 2008 11:10 am

Referring to the first grid of ArkieTech,

[r8c4]-4-[r8c9]=4=[r4c9]=2=[r3c9]-2-[r3c5]=2=[r7c5]-2-[r9c4]-
-4-[r8c4], => r8c4<>4.

by **ArkieTech** » Sun May 25, 2008 11:18 am

Thanks all. I got my puzzles mixed up an you saved me.

dan.

by **ronk** » Sun May 25, 2008 1:49 pm

StrmCkr wrote:next i use an aic
(2=8)R1C8-(8=3)R2C8-(3=5)R3C7-(1=5)R5C7-(2=5)R6C8- (5=1)R4C3 = > R6C1 <>5

Huh? r1c8 doesn't even see r6c1.

by **StrmCkr** » Mon May 26, 2008 1:11 am

yes it does indirectly. from an alternating chain

you only need to look at both options of
r1c8, or R2C8

as the outcome of both nodes are identical.

its like this.

when:

R1c8 = 8 => r2c8 (3) => R3c7 ( 5) => R5C7(1) => R5C3 (5)
R1C8 = 2 => R6C8 (5) > R5c7 (1) => R5c3 (5)

therefore R5C3 <> 1 , R6c1 & R5C7<> 5

Code: Select all: *-----------------------------------------------------------* | 4 5 7 | 29 3 89 | 6 28@ 1 | | 6 138 2 | 5 7 18 | 4 38@ 9 | | 138 9 138 | 6 12 4 | 35@ 7 258 | |-------------------+-------------------+-------------------| | 128 128 9 | 3 6 5 | 7 124 24 | | 7 6 15@ | 8 4 2 | 15@ 9 3 | | 235 23 4 | 1 9 7 | 8 25@ 6 | |-------------------+-------------------+-------------------| | 1258 4 158 | 7 12 3 | 9 6 58 | | 135 7 6 | 49 8 19 | 2 1345 45 | | 9 1238 138 | 24 5 6 | 13 1348 7 | *-----------------------------------------------------------*

its based of the sets marked wiht the @ symbol on this grid

by **ronk** » Mon May 26, 2008 3:38 am

StrmCkr wrote:yes it does indirectly. from an alternating chain {AAic i belive it actually is : edit for the error in identity}

you only need to look at ...

I understand the deduction. What I don't understand is why someone with 183 posts still can't properly write a simple xy-chain in NL notation or AIC notation. There is nothing "indirect" about the chain ends in either notation. One should be able to look at the ends of an AIC and see the derived strong inference, e.g.,

(5)r5c3 = (5)r6c8 ==> r6c1<>5,

or (5)r3c7 = (5)r6c8 ==> r5c7<>5

Because of the bivalues of the xy-chain, the AIC looks a bit different, of course. For the latter (the shortest), it looks like ...

(5=3)r3c7 - ... - (2=5)r6c8 ==> r5c7<>5

You should be able to fill in the ... between the ends.

BTW: An xy-chain is not an AAIC, its' an AIC. BTW2: The "official" language for the Players' Forums is NL notation, not AIC notation. For instance, Mike Barker and I use NL notation here, and AIC notation on the Eureka forum.

by **StrmCkr** » Mon May 26, 2008 6:31 am

casue i still can't figure out how anyone writes them out.

as much as i hate to admit it i have a learning disabiltiy. I am ADHD. i can absorbe techniques and apply the methods quickly, but i miss the finer points on presentation and litteration of such things ( I never pay much atention to things not of intrest to me). its on of my pitfalls and a hindernce most of the time. i need help to over come it.

links to and descriptions starting with the basics
in very simple clear writting techniques would greatly apreciated.

some of the chains i read from right to left or left to right didnt always work out the same. till i strated looking out side the noted links.

i can read(through trial and error); them how they work but still muddled on how or why they are written that way.

and so,

i cannot figure out what the diffrence between strong and weak links are, how there represented. is it a strong link when AB=AB, and a weak link when: Ab-AC (both connected, or only singles)

or
why i don't show all values that are in each cell some times and not in others, i have a huge list of stuff im still trying to figure out.

i basically need to learn presentation stuff from scratch.

coupled with the fact i don't have proper grammer, punctiation etc to formulate a coherant argument. this also impeads my ablity to understand your points some times as well. i have to re read it many times.

{written things that make sence to me make no sence at all to most others.}

makes it very hard for me to write and get an idea or move shown and understood.

ps. what is NL language ( i know its the langue of choice here) just don't know what it is etc, or what it stands for.

also
how it it jsut an AIC when the 2 cannot be placed in R6C8 - as it leads to 1 in RC53 instead of a 5. which is diffrent then the other placements.

i thought AIC chains had to have all points of insertion leading to the same deduction,

by **ronk** » Mon May 26, 2008 12:32 pm

StrmCkr wrote:casue i still can't figure out how anyone writes them out.

as much as i hate to admit it i have a learning disabiltiy. I am ADHD.

I'm sorry to read that. As I've no experience in dealing with ADHD and have never been the patient type, I don't think I can be of help. Perhaps in a few months, when I should have more time.

In the mean time, I can occasionally post the correct notation for some of your chains -- as I've done on Eureka. I also suggest that you don't try to post complex chains until you've conquered the notation for simple ones.

NL stands for nice loop. It's a notation system that explicitly uses inferences (links) between cells or groups of cells, also called nodes. Inferences between candidates of [edit: the same node are usually] implicit. The Forcing chains: Terminology and Definition thread is a good reference.

In contrast, in AIC notation all inferences between candidates are explicitly stated.

by **DonM** » Mon May 26, 2008 5:27 pm

StrmCkr: Though I know it must be difficult to have to explain why it has been difficult for you to pickup notation, it's helpful that you have so that the rest of us can be as supportive as possible. I know something about ADHD & learning disorders and for a start, I would suggest that you stick with AIC notation & not try to learn nice loop notation for the time being. The good news is that you have made real progress with AIC notation and if you keep working at it, I have no doubt that you will master it even with the more difficult chains.

The other good news is that you appear to have well above average sudoku solving skills which is another reason to continue to work on your AIC notation skills so that your solving can be better appreciated by others. My guess is that your talent to see fairly complex Boolean relationships in your head in more difficult puzzles has made it especially difficult because they can be even more difficult to express clearly in AIC or nice loop notation. So, Ronk gives some good advice above:...don't try to post complex chains until you've conquered the notation for simple ones. If you do that just for a little while, the skill to notate more complex chains should come fairly quickly cause you're a pretty smart guy

.

by **ttt** » Tue May 27, 2008 5:10 pm

Hi StrmCkr and All,

I have no doubt that you are talent than me, It seems that you are strong for ALSs…
For me, all of you : StrmCkr, ronk, DonM, Mike, Myth, Carcul, Jeff, tarek, Steve K, DBP, gsf… and re’born, if you are not here then Sudoku is nothing… I like Sudoku because of you show me the elegant of logic to find TRUTH.
Thanks to All,
ttt

by **StrmCkr** » Thu May 29, 2008 8:49 am

The other good news is that you appear to have well above average sudoku solving skills which is another reason to continue to work on your AIC notation skills so that your solving can be better appreciated by others. My guess is that your talent to see fairly complex Boolean relationships in your head in more difficult puzzles has made it especially difficult because they can be even more difficult to express clearly in AIC or nice loop notation. So, Ronk gives some good advice above:...don't try to post complex chains until you've conquered the notation for simple ones. If you do that just for a little while, the skill to notate more complex chains should come fairly quickly cause you're a pretty smart guy .

thanks for the compliments,

yes the complexaties in chains i notice quickly are hard to explain beaceus of there size..

tend to make it very hard to find a way to show it in entireity... as i did in ER 1 & 2.. and mikes version of it on here..

and often make it even harder for me to word

yes i'll take the advice and try learing writting out the smaller stuff first. then move up gradually and see how it goes.

lets see if i can explain how i see the map of a puzzle.

i see it in paritally as the numbers them selves or

for example
.. i see shapes and patterns formed by paired numbers or singles rather then the numbers them selves

Code: Select all: x --- x | | \ x---------x | | x - x

do u see the x's or
an sideways and mirrored L?
i see the L - from this shape all the areas's in side the white small squares(formed by the shapes outlines) can't be other x's ( parts of patterns).

i can do the same fora few named patterns
swordfish, turbots, empty rectangles, kites(skyscrappers), and a few other patterns that i have found to be the same their named.

i find shapes like this all over the puzzle, msot are known other i haven't seen else where.

.

The New Sudoku Players' Forum

Another one trick pony

Another one trick pony