chat prefs...
|
12:29 am
bigshow
darn it, I think my brain has shrunk
12:29 am
bigshow
i'm stuck!
2:40 am
sprite
my browser had an interesting loading problem... showed this expert as to be empty of any given numbers :)
2:40 am
h
hmm - might be easier that way... :)
2:41 am
sprite
or not! LOL... haven't made my own sudokus yet. But refreshing solved that part of the puzzle...
2:41 am
h
wellp the first 5 cells from crasshatching were easy-enough... time to see what this expert is all about...
2:42 am
sprite
got 8 filled now and still not hard
2:53 am
sprite
halfway and kinda stuck
2:55 am
h
ditto, 18 filled... im afraid greens arent going to change anything, but time to try em...
3:05 am
h
greens didnt help - dont see any chains/wings that i can understand - going red/guess now...
3:05 am
sprite
greens did help here, am seeing the light at the end of the tunnel... now I only hope it isn't a train
3:06 am
h
cool, congrats... *keeps dropping red-crumbs*
3:08 am
sprite
got it. Thought I had a unique rectangle and hit red, but I could go back and find the mistake.
3:10 am
h
neato - my first guess hasnt solved, nor conflicted - i hate that...
3:11 am
sprite
you'll figure it out, as ever :) GL and cul8r
3:13 am
h
ooph - conflict, finally... i hope *goes back and starts the other side of the red-trail*
3:18 am
Dazza
going purple
3:29 am
h
done, but i needed three guesses - sigh...
3:57 am
h
i tried plugging this one into scanraid and i honestly have no idea how it solves the thing even tho it tells me every step... apparently unique rect does apply (i still dont get this one), but when it got to *an evil strategy*, it looked like voodoo to me... gl everyone...
4:19 am
augurus
Just to reiterate, Unique Rectangle emphasizes the rule that a proper Sudoku has only one solution, thus if those four cells all contain the same two candidates, then that Sudoku is invalid. Make more sense or confuse you more?
4:27 am
h
i always have trouble deciding what-can-be-eliminated in a UR, one? both? either? (apparently not) i only seem to understand it when the squares are xy,xy,xy,xyz... its all voodoo, i say... j/k
4:36 am
drwho
Difficulty score 41. I did a lot of guessing.
4:40 am
augurus
You're applying a difficulty score to something you guessed? Does this mean that your criteria for difficulty is the difficulty of the guess?
4:46 am
h
lol - the computer gives the score. the human (tries it indep of computer) needs to guess... :)
4:51 am
augurus
Instead of worrying about other people's credibility, I should actually rest my head on a pillowy star.
7:28 am
irv
To H: what is scanraid? how do i acquire one?
7:33 am
bigshow
irv, it's a website dude, scanraid.com
9:38 am
gman
I think there will be a few LESS ELITES tomorrow morning after this one!! A real head scratcher!!
9:39 am
OrangeSlice
It's got me doing singles chains right now.
9:45 am
UnikeTheHunter
Go. Well, I got one cell without greens.
9:45 am
OrangeSlice
Oh you'll get a handful before you need to green.
9:50 am
bunnyvicious
done
9:50 am
UnikeTheHunter
That's pretty good, Bunny.
9:50 am
UnikeTheHunter
Did you guess?
9:52 am
UnikeTheHunter
Well, I guess there's no shame in going green at this point.
9:53 am
OrangeSlice
I just made a move that seems logical, but that I've never seen documented anywhere.
9:53 am
UnikeTheHunter
Oh, there was some shame! I forgot to check for loners.
10:00 am
OrangeSlice
Ok, no, my logic was apparently illogical. My logic teacher says that too, in fact.
10:01 am
OrangeSlice
Back to madness.
10:10 am
justdefyu
hey hey
10:14 am
ShadeEagle
slice, what was the move you did?
10:19 am
OrangeSlice
Well, I have two aligned cells with 3/4 in them, and two parallel cells with 2/3/4/5 and 2/3/5/6 in them, I concluded that in order to avoid the 'deadly pattern' those cells must either be 2 and 5 or 2 and 6, which leads me to infer that one of them must be '2' and so I can remove it from another cell that the can both see.
10:19 am
OrangeSlice
However, it broke the puzzle.
10:19 am
OrangeSlice
er, the second thing should be 2/3/4/6.
10:19 am
ShadeEagle
meaning it made it incredibly easy to solve, right/
10:20 am
ShadeEagle
er, ?
10:20 am
OrangeSlice
No, it didn't do anything at all, in fact. But I found another identical pattern somewhere else and did the same thing, and _that_ made it incredibly easy to solve and subsequently broke the puzzle.
10:20 am
ShadeEagle
ah, okay
10:21 am
ShadeEagle
yeah, I've done something like that on a couple of other puzzles (I forget which), although I have no idea if there's a name for it
10:21 am
OrangeSlice
I'm looking for a death blossom now.
10:22 am
ShadeEagle
a what?
10:22 am
OrangeSlice
A horrible, horrible maneuver that should never have to be used.
10:23 am
OrangeSlice
Also, the lack of (unlocked) bivalue cells in this puzzle is aggravating me.
10:24 am
ShadeEagle
yeah, I'm still early in this puzzle, but I've only found a few
10:24 am
UnikeTheHunter
All the 7's in the lower half seem significant, but I can't see how.
10:25 am
OrangeSlice
I've exhausted the possibilities of an xy- or xyz-wing, wxyz doesn't look too likely based on (again) the lack of unlocked bivalues. Singles chains led to nothing, though I didn't try every number, only those that seemed likely to give me something.
10:26 am
OrangeSlice
Almost locked sets and aligned pair exclusion are next.
10:27 am
UnikeTheHunter
Well, I'll take a break and come back to this later when my brain is less fuzzy.
10:27 am
OrangeSlice
I _will_ find the solution. I'm too stubborn to force anything.
10:27 am
bunnyvicious
Yes, I guessed :)
Sorry, the kiddos are runnin' circles around me.
10:27 am
ShadeEagle
this one has me worried; even when I'm making correct moves I don't think they're right, but I just broke open a little bit ... and I'm seeing a cascade beginning
10:31 am
Minski
Done! Had to use greenies and made 2 educated guesses but after that everything fell into place.
10:34 am
ShadeEagle
at this rate, I'm going to have greens in every cell before it's all said and done
10:37 am
OrangeSlice
The majority of moves I know depends on unlocked two-value cells, and this grid just doesn't have that many.
10:38 am
ShadeEagle
yeah, as do I - I have a few random tricks I can do with cascades (I don't know what their actual names are) and some three-valued cells, but this is just kind of a mess. I'm resorting to that trick you mentioned earlier to see it that unlocks anything (although I do have to ask: were the values you used there values in the puzzle?)
10:39 am
OrangeSlice
ah, yes.
10:40 am
ShadeEagle
okay - then I either haven't reached there yet or I'm going about it differently, I don't know
10:41 am
OrangeSlice
Starting with R6C3 and R6C9, head south two rows for the other two cells, just don't use it on the 4/5 pair nearby, that's what broke it for me.
10:41 am
OrangeSlice
In fact I could remove '3' from a certain cell based on what broke my puzzle, but I'm not doing it because that's illicit knowledge. lol
10:42 am
ShadeEagle
okay .... you're still a little ahead of me then. I can see that I'll eventually get to that spot, but I'm still missing some knowledge that'll let me eliminate a number from each of the other ones (not the R6 ones; I have those)
10:43 am
OrangeSlice
I have the entire puzzle greened (manually, of course)
10:44 am
ShadeEagle
was that a 2/3/5/6 in one of the cells? I just got something else
10:47 am
OrangeSlice
no, the cells are 2/3/4/5 and 2/3/4/6
10:47 am
ShadeEagle
okay, that makes way more sense
10:47 am
OrangeSlice
I mistyped the first time around.
10:48 am
ShadeEagle
yeah, I was looking at it (missed something that could be column-eliminated - whoops!) and realized that either my puzzle had bit it or there was a typo :)
10:50 am
ShadeEagle
okay, now I have to find the other time this occurs :)
10:56 am
ShadeEagle
...which I can't seem to do, I don't think.
10:59 am
ShadeEagle
does this incarnation need those matching bivalue cells? I only have a couple of instances of them right now, which may be the issue, and neither of them match up with anything corresponding
11:00 am
ShadeEagle
(...I think. I do have one potential match, but that resulted in a dead end, I think.)
11:04 am
OrangeSlice
Whoops, sorry.
11:04 am
OrangeSlice
It builds on unique rectangles.
11:04 am
ShadeEagle
I think I may be familiar with that term in practicality, but not in theory, if that makes sense
11:05 am
OrangeSlice
You can't have a square sharing the same two values because they would be interchangeable and the sudoku would not have a unique solution.
11:06 am
ShadeEagle
4 boxes can't share the same 2 values?
11:06 am
OrangeSlice
Right.
11:06 am
ShadeEagle
okay - then I did see it correctly
11:06 am
ShadeEagle
unfortunately, this leads to another more depressing conclusion ... I screwed something else up, maybe.
11:07 am
OrangeSlice
In this case, if we look at the locked pair of 3/4, we can see a rectangle formed by two other cells that also share 3/4, but that have other values.
11:07 am
ShadeEagle
yeah, I see it - I actually did see it earlier, but I think I have another problem with the board that's going to cause me to have to restart
11:07 am
OrangeSlice
Both of those cells can't reduce to 3/4 or else we would have a rectangle of 3/4 which is broken in and of itself.
11:08 am
ShadeEagle
...oh. wait, just kidding. I made the same assumption twice.
11:08 am
OrangeSlice
Thus, one of those cells has to not be 3 or 4.
11:08 am
ShadeEagle
and then that causes a bivalue (?) match with another cell, which allows the break in this case
11:08 am
OrangeSlice
Now, there's no documented unique rectangle strategy for four-value cells.
11:10 am
ShadeEagle
it should be able to be done in theory, but I'll have to figure out how later. even though I think I have this broken, I may have to go back in a minute and restart
11:11 am
OrangeSlice
But I built on the logic involved with a type-4 rectangle (I think) and made an assumption that since those two cells can't reduce to 3/4, they must be some combination of either 3 or 4 AND either 2 or 5 and either 2 or 6.
11:11 am
OrangeSlice
And actually, breaking it down like this is making me see why it's not a logical move. heh
11:11 am
ShadeEagle
....yeah, I am. nuts.
11:13 am
ShadeEagle
unfortunately, every time I attempted to solve the puzzle from that point I hanged myself. I'm starting over.
11:13 am
OrangeSlice
I applied the same logic using the 4/5 pair in the botttom-center box along with the 3/4/5/7 and 2/3/4/5 cells in the lower-left, which is where my puzzle broke and I decided that my idea wasn't going to work after all.
11:14 am
ShadeEagle
so in other words that idea doesn't work, or it does? you've totally confused me now
11:14 am
OrangeSlice
It does not.
11:15 am
ShadeEagle
in general, or just in that particular instance?
11:16 am
OrangeSlice
Logically, there's nothing that says I can eliminate the candidates that I did.
11:16 am
OrangeSlice
Therefore it's not a valid technique.
11:30 am
ShadeEagle
wow, solved it by accident
11:55 am
OrangeSlice
Solved it with a jellyfish.
12:21 pm
Krys10
a jellyfish on what number? I'm willing to try it..
12:22 pm
Krys10
(hi peep!) pop on IM
7:06 pm
Andy
good lord...
7:06 pm
Andy
here goes nuttin'
7:09 pm
bigshow
man, March 28, 2006, hardest puzzle *ever*
7:18 pm
helenkeller
definitely not an easy one
7:46 pm
Andy
done. plugged this into Sudoku Susser (Mac app) to get some help. Unfortunately, it suggests lots of comprehensive chains which are, essentially, guessing. Wish I could have gone without the guessing.
9:04 pm
spreadeagle
Okay, I've read the posts, and nobody seems to have a solution without guessing. Here goes nothing......
9:13 pm
spreadeagle
Hmmmm, lots of nothing. 11 filled in, going manual green.
9:18 pm
moedog
expert? i predict a large amount of sucking in my immediate future!!
9:20 pm
spreadeagle
I'm with you moe. Greens helped some, 18 filled in. But nothing left to try.
9:23 pm
moedog
looking back through the days comments is discouraging. don't do it!
9:24 pm
spreadeagle
I agree. Jellyfish? This late on a Saturday night? No way...
9:25 pm
moedog
making a few guesses...
9:26 pm
moedog
gonna tackle this one later tonight. getting a few beers might not make the puzzle any easier, but I won't care as much! :D
9:27 pm
spreadeagle
It hasn't helped me. See ya...
9:48 pm
spreadeagle
Done. Had to guess. I hate guessing. Maybe tomorrow I will find time to to figure out a real solution.
11:22 pm
coppershep
expet...ouch...can I beat the clock