DIY - Rubik's Cube - Blind Man's Cube - Metal Rubiks cube

Step 3 Obtain 54 Round or Raised Square Metal Tabs of 6 different raised shape designs

You are kinda on your own for this one, I pried mine off of an old chair in my garage and filed the backs off with a dremel. If using an authentic sized rubik's cube, these will need to be smaller than 5/8th's of an inch square, and preferably no deeper than 1/16th of an inch (or the cube will become noticably cumbersome in size). If you can only find 5 shapes... don't fret... remember that a single side missing shapes is still an identifyer (provided all other sides have identifyers). Other options (not as cool looking in my opinion) are sandpaper, wood, cardboard, plastic, etc. Scrapbook or craft stores might be a good place to shop for something like this.

WARNING 1: The Designs must FEEL different from each other. Each shape needs to have a distinguishing characteristic to differentiate it by touch rather than sight. THIS IS IMPERATIVE.

WARNING 2: It is best to ensure that each of the 6 shapes is quadratically symetric* in design (i.e. turning the shape 90 degrees yeilds same design as before turned). This ensures that after mixing and solving... the cube looks complete (note that one of my shapes does not follow this rule).

Many who are new to Rubik's cubes are unaware of the fact that if you were to draw arrows on each square of the cube when you buy it, solving it to perfection is not as easy as some pieces will not be soundly placed (arrows in all same directions). See picture for understanding... you see the colors are all in place... but the cube is not truly solved. Unless you know how to solve a cube like this... you will want to ensure the designs on the tabs are symetric as I mentioned.

• - I am not sure if this is even a word, it just made sense in my juvenile vocabulary.

Comments

[SaiyanKirby]

Another thing people don't know is that unless you take the cube apart and put it back together wrong, there will always be an even number of incorrectly solved centers, unless they are all correct. That being said, if you can solve a Rubik's Cube with a beginner's Layer method, then you can at least make sure 5 of the centers are correct while solving, thus the last will with automatically have to fall into the right direction.

[P4nz3r]

that's not true: centers never move, they can only spin in place the opposite centers are always opposite all other pieces can move, but corners stay as corners and non-corners stay as non-corners

[kelseymh]

The terminology of "incorrectly solved centers" is just relativity: either the centers are fixed (and always solved) while the sides and corners move around them, or you "solve" a framework of sides and corners in situ while the corners move through that framework. Arguing about which perspective is "right" is equivalent to arguing about the "reality" of centrifugal vs. centripetal forces.

[gtig]

The centers are not "always solved" in the true sense of the word. Arguing about which perspective is "right" does merit some gravity. According to Rubiks there is "only one solved state" which is denoted as "all faces of 9 squares having the same color". However this is not one solution, there are hundreds (thousands?, I am lazy today and don't want to research the math, anyone else game) of possible "true" states that merit the cube being "solved" by Rubik's rationale. That is because even in a solved state... the 6 centers still have 4 possible states each of rotation without you knowing if it is in the "original pre-scrambled state". Only with a supercube can you truly know if you solved it 100%... as the centers have an arrow denoting the "truly" solved or unsolved state of the center peices. That is of course if you want to get down to that level of solving... if Rubik's solution befits your lifestyle... so be it.

[jray5]

No, there is only 1 solved state. The centers are fixed with the edge peices rotating around the cube, in the x, y, and z vertices. So if we are looking at the blue face, and the orange face is to the right, then there is an edge piece in between them that is both blue and orange. This piece needs to return to that same spot. That particular piece has two faces, which cannot move Independently of each other, and must be in that particular spot to be correct. There is only 1 solution to the Rubik cube, and only 1 state that fits the bill.

[kelseymh]

Good points. I was using "solved" in the same sense as Rubik: the visual (or tactile, given this I'ble's topic :-) appearance of the exterior surfaces, rather than the internal degrees of freedom.

Something I don't know is whether it is possible to manipulate a cube such that a given center (or combination of centers) can be rotated relative to the edges and corners, while ending up with the same solid-color faces. This is your second point; it may in fact be impossible given the engineering of the joints.

As for the number of solved states, we can do the math here. The corners provide a reference frame -- because each one has three unique colors,
their positions relative to one another are fixed, so therefore there is only 1 solved state for them. With the corners fixed, each edge in turn can have only one position and orientation, and therefore there is also a single solved state.
That leaves only the four internal degrees of freedom for the six centers; the total number of such states would be 4^{6 = 2}12 = 4,096.

Again, I don't know whether those states are reachable. If they are, then your discussion above about the meaning of "solved" is on point. If not, then the solution state is unique for any cube which is not disassembled and reassembled.

[ScavengerHack]

The centers can only be rotated in pairs. So instead of 4^6 = 4,096 states. We have (4^6)/2 = 2048 states. If there are arrows on the faces, then only one will be the solved state. Rotating the centers in pairs is not difficult however. So getting the "quadrilateral symmetry" for the tiles is not needed.

[gtig]

There are permutations that allow for cycling the center in a supercube, they are asked about a lot on boards, and I don't know them by heart. thanks for hitting the math... so there are 4,095 possible solutions that are not true... very interesting. That being said... I hate supercubes. lol.

[gruntking741]

Grandpa: I solved it!!!! Kid:Solved what? Grandpa: Supercube Kid: In like what, a week? Grandpa: 72 years!!! Kid: You have no life (Grandpa has a heart-atack and promtly dies)

[Sagar Gondaliya]

ditto

[spock155]

double ditto

[trogdoroth]

I think your thinking of "radial symmetry" but it's not really a big deal.

[HannahLegutki]

Im not here to r=write a big paragraph about centers or anything, i just have one question: where do u get the tiles?

[MegaMaker]

This is explained in the instructable.

[thecooltodd]

ROTATED CENTERS

Imagine a picture cube where all the pieces are permutated (in the right place) and all the cubies are correctly oriented (rotated) except the middles. It is IMPOSSIBLE to rotate a single face 90 degrees. If you rotate one center 90 degrees you ALWAYS rotate another center 90 degrees.

It is possible to have 5 centers correctly oriented and be missing the 6th. Here's how. Assume you have a perfectly solved cube and two faces, let's say Face#1 and Face#2. The following two steps can rotate a single face 180 degrees.
1. Rotate Face#1 clockwise 90 degrees and Face #2 clockwise 90 degrees simultaneously
2. Rotate Face#1 clockwise 90 degrees and Face #2 counter clockwise 90 degrees simultaneously
Notice that Face#1 has been rotated clockwise twice whereas Face#2 was rotated clockwise and then counter clockwise (leaving it in it's original position). On a side note, by "face" I mean "center"

I learned this while solving picture cubes, which are not very different from regular cubes. Just slightly different.

[tommylovesjamie]

you were spot on with your explanation, but the term quadratically symmetric was a little off. I think you would have been better off saying vertically and horizontally. but I think we all understood what you meant. kudos for the work on this. I plan to make one for my sis-in-law. she is a little on the vision impaired side.

[rlpowell]

"quadrilateral symmetry" is the correct term; it means reflected across both the horizontal and vertical axes. See http://en.wikipedia.org/wiki/Symmetry

-Robin

[irxo]

love the brass. not only more challenging.. but decisively steampunk. well done.