I am having great difficulty expressing this clearly, so i apologise if this doesn't make much sense.
There is a puzzle consisting of two nested cylinders, with a maze carved on the inner surface of the outer and a knob on the outer surface of the inner cylinder, so that one must find one's way through the unseen maze with the knob to move the inner cylinder from one end to the other.
In order to twist the faces of a Rubik's cube, all one need do is turn them in a single plane. If the "axles" of the central faces were mazes, it would then be necessary to pull the face out gradually, through a maze embossed on the interior surface of the axle, turn it at the outer end, then find one's way through another maze to put the faces back into place. For each ninety degree turn, a different route would be needed to do this. Since there are six axles, there would be a total of twenty-four mazes. Then there is the problem of actually doing the cube.
-- nineteenthly, Sep 13 2005

cryptex http://www.cryptex.org/
Might be a part of this idea. [zeno, Sep 14 2005]

the difficult one for people with brains po_92s_20su_20doku_20mayhem
[po, Sep 14 2005]

Not bad for an explanation, but I think an illustration would help. Sounds interesting (even though I can't quite visualise it).
-- Adze, Sep 13 2005


I agree that there should be an illustration but i can't draw, so i'd have to describe it to someone who could, and apparently i can't describe it either.
-- nineteenthly, Sep 13 2005


How do you complete a maze while turning one face unless the face can be pulled in and out along the axis that is 90 degrees to the direction of rotation?
-- wagster, Sep 13 2005


I understand what you're describing, and you are a diabolical bastard, [nineteenthly]. For being responsible for cumulative decades of lost time, frustration and anguish, I regret that I can only vote for this once.
-- Detly, Sep 13 2005


[wagster], you do pull the face in and out in order to do the maze.
Maybe there's a way to include mazes on each of the faces which also must be done to operate the puzzle.
-- nineteenthly, Sep 13 2005


If someone would please explain to me how I can link a sketch I would be gratefull, because I can visualise it and I think I have this covered. Therefore a major bun. I love rubic's cube's and this would keep me off the streets for months.

Actually it would be quite difficult to draw as a whole. You'd have to draw the two cylinders separately.
-- Susan, Sep 13 2005


[Susan], that'd be great. If you email me it i'll try to put it somewhere i can link to it.
-- nineteenthly, Sep 13 2005


Well in that case I understood it correctly the first time. I still don't understand how you'd build it though, maybe [Susan] does.

<sp> Rubics cube's, no wait, Rubik's cubes', no... oh who cares...
-- wagster, Sep 13 2005


If you wanted to build it, you could start with a piece consisting of six cylindrical protuberances, each with a knob sticking out sideways. To this could be added six hollow closed cylinders each containing an engraved maze, which could be teased gradually down onto the protuberances. The difficult bit would be attaching the corner and edge pieces, because they could easily fall off when the face was separated. However, if there were electromagnets of the right polarity attached to the edges of the centre piece with steel pieces on the surrounding pieces, and lifting the centre face switched them on, it wouldn't fall apart.
There would of course have to be something like a battery to power all this, but it could be recharged by photovoltaic cells on the faces or perhaps by a dynamo-like arrangement connected to the central faces.
-- nineteenthly, Sep 13 2005


Oh ok [nineteenthly] that is not the way I read it (the knobs all over) but I have emailed you my picture of how I see it.

The way I read it: You have two cylindrical Rubic's cubes (they already exist just can't find them on google) to simplify I used just three layers.

You have a knob (dot) that you need to get through an invisible maze (from bottom to top). The only pointer to where the maze leads, is to turn the layers and then pull out a section you have turned. The colour of the (maze) section you need to follow is (e.g) purple. you can turn 5 (6 sections) times to get the colour desired for solving the maze. If the colour is not correct you re-align the inner section. If you hit the right section the knob (dot) will lower and you can proceed to the next level.

Please tell me if this were the puzzle you were thinking of or no(t).

Ow, ow, its just like the cryptex that [zeno] is working on. back to you [zeno]
-- Susan, Sep 13 2005


Susan, this isn't quite it. I'll cut and paste my email to you here:
I understand your design, but it isn't quite what i meant. However, i think yours is actually more practical and a good idea.
What i had in mind was a cube with a maze on each column that connects the middle square of each face to the centre of the cube, so in order to rotate each face of what is otherwise a conventional Rubik's cube, it has to be pulled out, moved through an elaborate maze on the actual column, then moved back again to one of four possible exits, again on the column, to the centre. The problem with this idea is how to keep the side and edge pieces from falling off while this is being done, which could probably be solved by a system of magnets.
As i said though, i think your design is both very worthwhile and more practical than mine. I just wish i could draw it.
-- nineteenthly, Sep 13 2005


Anything I can't quite visualize gets my plus. Somebody post a sketch soon...please.
-- 2 fries shy of a happy meal, Sep 14 2005


Yeah, where's [Susan]'s drawing?
-- wagster, Sep 14 2005


I went to a Rubiks Cube party last week, actually. Show up wearing an item of clothing of each of the six colours, get assigned a colour on arrival (and trade with others until you get the one you really wanted), and then swap clothes until your attire is entirely of your assigned colour.

I started off looking like Dr. Who on acid, and ended up pretty much the same (but without the Dr. Who part). I had to take the beret off when I realised it sort of suited me.
-- Detly, Sep 14 2005


[Susan]'s drawing was pretty good but she doesn't know how to put it up for you to see. And [nineteenthly] said it wasn't quite what he/she had in mind anyway.

[nineteenthly], is this like a mix of the rubic cube and the cryptex? See link. Might be a help in understanding this idea.
-- zeno, Sep 14 2005


Didn't they have one of these in the movie "Hellraiser"? And when you completed the puzzle a bunch of demons showed up to poke you with sharp things? And some people liked it? Sure, I'll give a croissant!

Also, wasnt there something like this in Davinci Code? A puzzle with an inner maze carved into a cylinder? Full of acid, if I recall. Not quite sharp things and demons, but still ++!
-- bungston, Sep 14 2005


Upon concideration I have decided to bun this idea. I don't care for games like the mentioned cube or anything that resembles it. I'm no good at it and I don't like playing games like that. (maybe that's why). But the idea certainly has a lot of merit for those who do like games like this. Also it is an intelligent idea. So big plus for you.
-- zeno, Sep 14 2005


[Bungston], please check link. No acid, just vinegar to dissolve parchment.
-- zeno, Sep 14 2005


I'm getting into a bit of a mess about this. The cryptex looks like [Susan]'s idea, a bit, and certainly has merit, but it isn't what i mean.
Initially, it looks like a Rubik's cube. However, the faces must be pulled out before they can be turned. In order to pull them out, one must turn the column ending in the central face through a maze which is carved on the column, which has an inner and outer part. To return the face to its original place on the cube, that is, to put it back so that it again resembles a conventional Rubik's cube, one must negotiate one of four mazes going in the opposite direction to the first one. The major problem in building this would be keeping the corner and edge pieces connected to the central one while turning the face, hence my reference to magnets. I appreciate that this is very unclear without a diagram.
(Later) I finally seem to have managed to link to a diagram of what i mean. Does it make sense at all?
-- nineteenthly, Sep 14 2005


Yes it makes perfect sense and now I understand exactly what you're on about. The bun abides.
-- wagster, Sep 14 2005


Erm.... I still dont get it... :(
-- chocolateraindrops, Sep 14 2005


I'm sorry, [chocolateraindrops], i had a feeling that it might confuse some people. The maze at top right is rolled up inside the cylinder underneath it, and is an example maze. Each one would actually be different. I imagine that eventually one would learn all twenty-four mazes, but even then the orientation of the cube would be satisfyingly confusing.
-- nineteenthly, Sep 14 2005


What [wagster] said.
-- zeno, Sep 14 2005


I have to say the hellraiser bit back there does kinda scare me...I was going to say that the greatest thing about this idea is just trying to visualize it. Once you have a preliminary image in your mind you can actually play this puzzle through. Fascinating! It seems as though I had noticed an aspect of reality similar to this....
-- obesemind, Sep 15 2005


two cylinders inside of each other?... you could draw the maze on two flat surfaces facing each other so that one was the whole maze and one was a couple of precision placed squares the same width as the path in the maze...the squares are kind of following each other through the course of the maze so that when turned properly they would all progress through to the end. Roll this up creating two cylinders, one inside the other. The goal is to eventually pull the outer cylinder off of the inner. For a longer maze the inside of the outer cylinder would face in towards the outer face of the inner cylinder. The inner cylinder would be solid the outer would have one end closed. There is a handle on the opposite end of each so as to resemble a rolling pin sort of. Twist and pull until it comes apart. This is what I primarily envisioned when I read the idea.
-- obesemind, Sep 15 2005


Following my last entry and upon reading more of the comments I have realized that my rolling pin maze is what you intend for each of the axis of each section of the rubix cube. The best way to understand this theory is to first form a vivid picture of a rubix in your mind. Then manipulate it there as though it were a real one. By recalling that one must twist a whole side of a rubix cube to solve the puzzle this can be more easily visualized. Instead of simply twisting, one must pull the section away from the cube by twisting slightly back and forth as well as small to and fro motions through an inner maze such as the "rolling pin". Once the maze is accomplished the section is fully away from the cube and can be twisted. Once twisted, one must traverse through another maze, or the same one backwards, to set the section back down to the cube. Essentialy there would be four mazes on each "side" of the axis corresponding to the four sides of the square section. There are six sides to a cube, and each side has four mazes to traverse forwards and backwards. Six times four is twenty four times two equals forty eight puzzles one must master and memorize in order to solve the already perplexing puzzle of the rubix cube. The link to a picture near the top of this page actually does a fair job of illustrating the rubix part. As far as the mazes go, my comment above about the cylinders must be shrunk to a very thin "rolling pin" where the handles are the center of the cube and the section being twisted. The maze would have to be carved into flat pieces of metal using an etching process of some sort, they have software that runs machines that do this sort of thing, it would probably even create a maze and position the squares so that they could actually proceeded through the maze. Then another machine would carefully roll the two pieces together, locking them together into the axis. Does this describe well your picture?
-- obesemind, Sep 15 2005


So close.

One more and you would have had a hat trick.
-- 2 fries shy of a happy meal, Sep 15 2005


//I went to a Rubiks Cube party last week, actually. Show up wearing an item of clothing of each of the six colours, get assigned a colour on arrival (and trade with others until you get the one you really wanted), and then swap clothes until your attire is entirely of your assigned colour.//

How come I never get invited to parties like that?
-- wagster, Sep 15 2005


[obesemind], yes that does describe my idea. I like your other one as well. There could be a whole family of good puzzles based on misunderstandings of this idea. I hadn't thought of doing it in software: that could work really well, though it wouldn't be a substitute for an actual physical one of these. I don't think there are forty-eight possible routes though, unless you count the same route going backwards. I suppose there might as well be though.
-- nineteenthly, Sep 15 2005


[wags] - possibly because your friends aren't nerds...
-- Detly, Sep 15 2005


I suspect I have sufficient nerd friends to be able to host a Rubic party. Those who aren't attracted by the "Rubic" part I can always reel in with "swap clothes".
-- wagster, Sep 15 2005


//I still don't understand how you'd build it though, maybe [Susan] does//.
No, unfortunately. Seeing the picture I find that it is totally different to how I read it. And also I think this can not be baked.

Making it neccessary to go through the maze before you can turn, you need a way of preventing the turn if you do not go through the maze. I cannot see how that can be done. Well, actually, I Can, but it would make a very big cube.

The bun still stands though if only for giving me something to chew on.
-- Susan, Sep 15 2005


The cube needn't be _that_ big. The mechanism can be compressed at right angles to the direction of rotation, so the knob is a flange slightly narrower than the width of the perpendicular paths, and the lateral paths can be quite narrow. There is no need to rotate the actual knob itself. This implies that the column needn't be very long. To my mind, the biggest problem is preventing faces from disintegrating when they're rotated.
-- nineteenthly, Sep 15 2005


I hadn't thought of that. That is a real problem. I shall give it some thought.
-- wagster, Sep 15 2005


My solution was to make the middle square from a magnet and put pieces of steel along the adjacent edges of the edge pieces, but that might make it difficult to turn, which is why i started going on about electromagnets. The maze and flange arrangement could act as a switch, allowing a current to flow from the point when they come into contact on the inner side of the maze, which would turn on the magnet and attract the edges. The corner pieces could interlock as they do on a conventional Rubik's cube.
-- nineteenthly, Sep 15 2005


I went to look at your sketch link [nineteenthly] and, well it might not confuse everyone, but it sure confuses me.
For some reason the destination seems to be my flickr photo account page.

That's just so...weird.
-- 2 fries shy of a happy meal, Jan 11 2011


It goes to Flickr for me too, but not your account. I'll do another diagram and put it somewhere else.
-- nineteenthly, Jan 11 2011



random, halfbakery