Message #3059

From: David Reens <>
Subject: Re: [MC4D] Re: Introduction
Date: Wed, 28 Jan 2015 10:27:08 -0700

Hi Ray,

I know, its like what have I been doing for 20 years? haha

Yeah so in practice for the middle 3x3x3x1 layer I first just put the 2c
pieces in place (by this I mean the hypercubes that touch the outside of
the hypercube at 2 cubies, is that what the Nc notation I see everywhere
means? I should really go check out the wiki). So I’m moving sticks of
three that include one 2c piece and two 3c pieces, but just only focusing
on the 2c’s.

Once all the 2c’s are in place, I start doing just what you say. Move a
stick of three down to the bottom face, rearrange it to either transfer 3c
pieces to the bottom that don’t belong on that stick, or collect 3c pieces
from the bottom that do belong on it, then put it back. I think at this
stage its already required to do the inverse of anything you do to achieve
said rearrangement (SZ method as you call it), but since the bottom face is
largely unsolved, this is straightforward.

I looked at the Roice method briefly, it seems like a bit less headache
than what I did! I can remember people solving the 3^3 in a somewhat
similar way, at least with regard to starting with the upper cross. I’m not
familiar with the acronym LBL, but I’ll check out the wiki more thoroughly
soon. I think my method would have been a lot better if I had gotten around
to being comfortable with the macros, because I literally used the same
7-step edge-stick switcher in this middle reel over 50 times.

By the way, Melinda encouraged me to mention that I am the 200th 3^4
solver! Hopefully we can see many more in the coming years.

Speaking of many more, I and a friend have been thinking about engineering
some sort of tactile mechanism for interacting with the 3^4. Imagine
something like a 3^3 handheld with an internal gyroscope to detect overall
3D rotation and simpler sensors to detect twists. You could link 3D
rotation to the bottom face, and have the actual turns map to rotations of
the adjacent faces of the 3^4. Maybe you could put touch sensors on the
centers of the 3^3 that would activate the equivalent of a 4D overall
rotation with no twist (i.e. a ctrl+click). I think if something like this
existed, it would open up the 3^4 to the realm of speed cubing, which might
significantly increase visibility. Has anyone been thinking about something
like this??


On Wed, Jan 28, 2015 at 8:22 AM, Ray Zhao
[4D_Cubing] <> wrote:

> Hi David :3
> You’re a 26 year old with a first experience with cubing at 6. What is
> time?
> Even though you didn’t get to make 3^4 algs, you still translated LBL it
> seems onto the 3^4, and that’s already something. I think at least half the
> people on this group (including me) started with just following through
> Roice’s method (solution available on the webpage) for the whole thing.
> The method in general seems clear, but I’m not sure how the second
> layer/middle reel is solved. Since doing normal 3^3 algs on it moves 3x1
> sticks, are the sticks first manually joined on the last layer then
> inserted into the right place?
> For the last layer, the twist of having to do cube move+twist+cube move
> inverse is something I still credit Matthew for first telling me about.
> I’ve written the process down on the superliminal wiki under the SZ method,
> which is just a translation of the CFOP method into 4D.
> Of course, if you’re bored with not finding new algs, then try 3^5. x3
> Ray