"Log Cabin", "10", and "Buzz" [SPOILERS]
Kory Heath
This is certainly confirming the conclusion I came to in a previous thread: a pattern that functions as a puzzle when it's completely revealed will be too hard to figure out while playing. Of course, I think you designed these before we had that discussion, and anyway, we're all still in the experimentation phase of pattern-design.
One nice thing about Paradigm is that, unlike Zendo, it's not particularly painful to play a pattern that's too hard. Even if you're totally lost, you can just take an intuitive stab in the dark, and the worst that can happen is that you miss. The game always drives to a close, and if everyone's decided the pattern is too hard, it moves pretty quickly.
Here are the things we figured out about each of the patterns. [SPOILERS FOLLOW]
Log Cabin
Fascinating setup, with some tantalizing patterns, but we never quite figured it out. The most concrete piece of data we found was that yellow tiles can only be touching (orthogonally adjacent to) blue tiles or other yellow tiles, and that each blue tile must be touching at least one yellow tile. For almost the entire pattern, a similar pattern held for brown tiles: they can only touch purple tiles or other brown tiles (although purple tiles don't have to touch brown tiles). However, late in the game we found one place where a brown tile was next to a blue tile. At that point, we knew we were lost.
Anyway, I always suspected that these "rules" were just emergent features of some more basic coloring rule, but we couldn't figure it out. I suspect it has something to do with the "fractal" or "recursive" nature of the pattern: a small square tile has a tile above and below, then that group has a tile on either side, then that group has a tile above and below, etc.
10
This was the pattern that we figured the most out about. Each row of the pattern contains blocks of colors in sequence (either R-B-G or R-G-B), so we just figured out the various sequences. However, at the end it still felt like we were missing the big picture of why each sequence was the way it was. We assume that there's some mathematical rule here. I'm sure the name "10" is a clue, but we don't know what it means. BTW, the strong horizontal symmetry of this pattern leads to what I call the "unhappy-seat" effect. I make an incorrect guess, and my opponent just guesses the corresponding tile on the opposite side of the board and then passes it back to me, putting me back in the unhappy-seat. It doesn't break the game, but it's something to be aware of as a pattern designer.
Buzz
We figured out very little about this one. We tried lots of ways of forming paths with the arrows, but didn't come up with anything that worked. About halfway through the game I began to suspect that the arrows were red-herrings that didn't mean anything, but I still don't know if I'm correct. There are long diagonal lines of green that seem to mean something, and the placement of the red tiles seems quite regular. In fact, if we hypothesize that yellow tiles are formed when red and green combine, then this pattern can be viewed as a bunch of diagonal green lines combined with a red "knight's move" tiling, on a blue background. If that is indeed the pattern, it might be better to add an extra line of tiles at the bottom of the pattern, which shows one more red-green collision, and centers the red group in the pattern. This might make it a bit easier to figure out.
If my hypothesis is correct, this is a neat pattern that may not quite work. If you eliminated the arrows and made all the tiles square, you'd have a quite playable pattern - a nice visual pattern with just enough asymmetry to keep things interesting. The question is whether even this relatively simple pattern is too complex to be mixed with the "gimmick" of the red-herring arrows (which I like in principle). It may be that the gimmick requires for the underlying pattern to be even simpler - something very basic, like my Cornerstone pattern.
-- Kory
carlto...@gmail.com
the patterns may be recognizable once you get going but it also
functions as somewhat of a puzzle. If you figure out the basis of the
pattern design you will be able to fill the pattern in easily. As in
most things like this (Zendo Included) it's always easier if you know
the answer. Also Perhaps we can have a SPOILERS thread where we can
discuss the ins and outs of individual patterns and some one new to a
particular patterns can avoid those posts. ( Iamy need to use these
groups a bit more to see how this might work.)
Ryan McGuire
I intended it to be a field of red*. The yellow "seed" tiles were
placed "randomly". Then blue tiles were anything just one away from
the yellow (that isn't already yellow) and purple is anything TWO away
from the yellow (that isn't already yellow or blue).
(Note: You could treat it as a yellow field with red seeds and purple
and blue 1 and 2 spaces from the red.)
> However, late in the game we found one place where a brown tile was next to a blue tile.
Oof. Master error. I better fix that. Even if the pattern is
determined to be too hard, I want it to at least be "correct".
The name comes from the Log Cabin quilting pattern.
* On my monitor at home the "brown" looked a lot redder and had higher
contrast with the purple. At work, I now notice that they are almost
the same. I'll make the brown tiles more red while I'm fixing the
error.
----
10:
I thought the fact that there were 27 triplets of tiles and three
colors would give it away even before any tiles were revealed. I also
thought that it would be obvious that having the bottom row offset by
half a tile would make it semi-obvious that it was a single zigzagging
sequence of 27 triplets of tiles. No, huh?
I'll tell you what, I'll go add some white space to make the
zig-zagginess more obvious. Maybe that will help.
The name comes from the an math "pun": In every numbering system, there
are 10 different digits.
----
Buzz:
Your one suspicion is correct: the tile shape is a red herring. This
wasn't digustingly obvious once the whole pattern was revealed? Huh.
You can treat it as diagonal green lines and red knight's move "lines"
(crossing at a yellow tile). Another way to look at it is as a
sequence of numbers (seven per row) where the multiples of three are
green, multiples of five are red, and multiple of 3 and 5 are yellow.
I see what you're saying about making the puzzle one line (or more)
taller to give more examples of yellow tiles.
The name comes from the game "Bizz Buzz". Players take turns counting
and numbers that have certain properties are replaced with words.
Perhaps prime numbers are "Bizz", multiples of 4 are "Buzz, and perfect
squares are "Fizz". Ryan: 1. Kory: Bizz. Carlton: Bizz. Ryan: Fizz
Buzz. Kory: Bizz. Carlton: 6. New things can added as the game goes
on. Triangular numbers (3, 6, 10, 15, 21) are "Wang", etc. Numbers
with at least two digits the same are "Murb". I figured multiples of 3
and 5 would be hard enough.
----
Red Herring:
The name doesn't refer to any misdirection -- just the herring-bine
pattern.
----
The promised uploads shouldbe available later tonight (Monday, July 24,
Eastern).
Ryan McGuire
The new version of 10 has more whitespace.
Buzz is one line taller. (Maybe it should be even taller.)
http://www.ryanker.com/Looney/LogCabin.svg
http://www.ryanker.com/Looney/10.svg
http://www.ryanker.com/Looney/Buzz.svg
Ryan McGuire
http://www.ryanker.com/Looney/Bleeding.svg
and
http://www.ryanker.com/Looney/Dude.svg