wu :: forums
        (http://www.ocf.berkeley.edu/~wwu/cgi-bin/yabb/YaBB.cgi)
      

        riddles >> general problem-solving / chatting / whatever >> Rubik's
        
(Message started by: TruthlessHero on Jun 8th, 2007, 8:04am)
Title: Rubik's
Post by TruthlessHero on Jun 8th, 2007, 8:04am
When I was younger, I always thought having a Cube made me look smart, even though, I didn't know what I was doing. I've unearthed my old one, and now, I can consistently get one side, regardless of color, but I ALWAYS mess it up going for side two. Any suggestions?
Title: Re: Rubick's
Post by Sameer on Jun 8th, 2007, 9:21am
easiest algorithm:

When you make first side: you will notice that all adjacent sides will have a small "T" ... top row and middle square done...

NExt step is to make the second row of adjacent sides.. with some trial and error you will be able to figure that out pretty fast...

So now at this point you will have 1 face done. and top two rows of adjacent faces done... now to the bottom face..
this is where it gets tricky... try to get a "+" in the bottom square.. once that is done go for the corners...
I figured this out when I was in high school... but I had plenty of time that part of my life ;) ... if you get frustrated.. search google..  ;D
Title: Re: Rubik's
Post by TruthlessHero on Jun 8th, 2007, 9:32am
Well, right now I have green done, the other sides of the plane look like so:

Code:
O Y R

W W Y

R R O

Y O W


As you can see, I don't have 3 in a row anywhere, necessary for said T's.
Title: Re: Rubik's
Post by Sameer on Jun 8th, 2007, 11:09am
hmmm I don't quite understand... Let's do this

TOp face - face1 - color green
Bottom face - face 6 - color blue
Adjacent faces (face 2 through 5) - colors orange, yellow, red, white

If you have face 1 green done.. then face 2 through 5 should have this... (listing for RED)

RRR
XRX
XXX
where X can be any color (notice the T of R's) you will see the same for orange, yellow and white. If you don't have this way then you didn't make the green correctly
Title: Re: Rubik's
Post by TruthlessHero on Jun 8th, 2007, 11:25am
Hmm, I guess I didn't make it right, but I don't think I've ever seen one of these Ts...I guess I'll keep making sides until I do.

Ok, now I have white on top, and T's for each color on the sides. The bottom doesn't have a + though...

Code:
Y     R     R
Y     Y     Y
R     B     Y


Do I need to get to the + config? Or can I work from here?
Title: Re: Rubik's
Post by Sameer on Jun 8th, 2007, 1:19pm
after you get 1 side and the T ... you need to get the adjacent square to T

So for adjacent side say Red we had

RRR
XRX
XXX

We need to get

RRR
RRR
XXX

Same for other three adjacent sides.. so once you do this you will have

Face 1 done.. adjacent faces top two rows done...

Then next step is to have the + on the bottom face.
Title: Re: Rubik's
Post by TruthlessHero on Jun 8th, 2007, 1:26pm
Right, I have all the T's, but I can't seem to get the + on the bottom. Is that necessary?
Title: Re: Rubik's
Post by Sameer on Jun 8th, 2007, 1:46pm

on 06/08/07 at 13:26:24, TruthlessHero wrote:
Right, I have all the T's, but I can't seem to get the + on the bottom. Is that necessary?


After T and before the + you need to get the two rows on adjacent sides.. essentially get the small "T" to become "=" .. see my e.g.
RRR     RRR
XRX-->RRR
XXX     XXX
Title: Re: Rubik's
Post by rmsgrey on Jun 9th, 2007, 8:12am
I find it more helpful to think about the 20 moving pieces of the Cube (12 edges and 8 corners - the centre panels in each face are fixed).

If you think about a corner piece, you're looking at a formation of three squares of three different colours, that defines the target position and orientation for that piece. Similarly, the edges are a pair of squares that define that piece's destination.

So you're not just trying to get a flat plane of a given colour, you're trying to get 8 solid pieces into the right places.

If you follow Sameer's approach from there, the next steps are to get the four equatorial edges in place, then the bottom four edges - by which point the remaining four corners will be in one of only 324 possible configurations and, hopefully, possible to sort out without losing what you had.

When I solve the cube, I tend to get the edges into place, and then tackle the corners, though my method of attack varies depending on my mood - the main idea is to pin down certain pieces and restrict what move-sequences I employ to ones that I know will affect them in controllable (and reversible) ways
Title: Re: Rubik's
Post by TruthlessHero on Jun 9th, 2007, 10:04am
Doesn't an edge include two corners? I'm confused with what you mean by doing the edges then the corners.
Title: Re: Rubik's
Post by Grimbal on Jun 9th, 2007, 10:47am
An edge is a cube that has 2 visible stickers.
A corner is a cube that has 3 visible stickers.
Title: Re: Rubik's
Post by rmsgrey on Jun 10th, 2007, 7:23am

on 06/09/07 at 10:04:34, TruthlessHero wrote:
Doesn't an edge include two corners? I'm confused with what you mean by doing the edges then the corners.

There are 2 types of movable piece on a Cube - 8 corner pieces, any of which can end up on any corner of the Cube and in any of three orientations, and 12 other pieces, that are each always in the middle of some edge of the Cube, and can each end up on any edge and in either of two orientations. The second type of piece is often referred to as an "edge" piece for convenience.

The 6 tiles in the centres of the faces can rotate in place, but never actually move (if you disassemble a Cube, you'll find an otherwise rigid 6-armed central mechanism that has the 6 face-centre tiles as the ends of the arms) so they can be discounted when it comes to a solution...

If you poke around online, you can find various articles about the Cube that might help and/or be of interest...
Title: Re: Rubik's
Post by rmsgrey on Jun 13th, 2007, 12:04pm
You-Tube has a number of "helpful" videos. This one (http://www.youtube.com/watch?v=2jzKflLF3VQ&NR=1) isn't perfect, but it's the best I've come across so far.
Title: Re: Rubik's
Post by JiNbOtAk on Jun 13th, 2007, 6:57pm
For the more adventurous ones, you could try this (http://www.geocities.com/meorhamzah/rubik.mpeg).  :P
Title: Re: Rubik's
Post by FiBsTeR on Jun 22nd, 2007, 9:41am
Unfortunately, "getting a side" is not very helpful in solving a Rubik's cube, because a "side" is 2-dimensional, and the process to solving the cube must be done in 3 dimensions. For example, you have to get your mind to see the cube not as 6 sides but as 3 layers.

A very easy site to learn the cube in "blocks" is http://lar5.com/cube/. This is my favorite method.

Watching instructional videos on YouTube is also useful if you are a visual learner.

http://youtube.com/watch?v=NdUwn1vgVc4 <-- me doing the Rubik's cube!
Title: Re: Rubik's
Post by SMQ on Jun 22nd, 2007, 10:27am
While I agree with your comments about learning to think of three-dimensional pieces and layers rather than just the colored stickers, as someone who solves cubes infrequently (and has no aspirations to set any speed records), I've found I do better with a method with as few memorized sequences as possible.

My current method (and the one I'll probably stick with) is based on this mathematical solution (http://alexfung.info/favorite/cube/cube.htm) but further reduced to only need four memorized sequences beyond basic single-piece positioning moves: one each to shuffle three corners, twist two corners, shuffle three edges, and flip two edges.  It's then pretty easy to put those together into a solution.  It's not very fast, but when somebody hands me a cube to solve in a couple years I'll still remember how to do it. ;)

--SMQ
Title: Re: Rubik's
Post by FiBsTeR on Jun 22nd, 2007, 1:43pm
1) If you're looking to simply solve the cube without memorizing any move sequences, there is a way to do so: http://www.ryanheise.com/cube/ . In the link I gave in my previous post ( http://lar5.com/cube/), you actually need only 3 memorized algorithms (one for swapping 2 corners, one for orienting 3 corners, and one for swapping 3 edges).

2) IMHO, memorizing move sequences isn't necessarily demanding (the most widely used speedcubing method, the Fridrich method, actively uses 56 algorithms). But learning algorithms isn't just sitting at a table with a Rubik's cube doing a sequence over and over until it's ingrained in your head. It's simply a matter of practice, solving the cube with maybe a page of notes in front of you. If you're using a method with only a handful of algorithms, I doubt you would need anything more than a dedicated hour or so. In all cases, though, after 3-4 weeks, I can guarantee you you won't forget an algorithm. Ever. If you're skeptical, when I first learned to solve the Rubik's cube blindfolded, I thought there was no way I could remember all 20 pieces. But, by having a good system, it's really easy, and even a week after I could recall the cube from memory. Of course, it's not an overnight thing (for example, whenever I went to a restaurant I tried to memorize different parts of the menu and recall them when I woke up the next morning).

/rant

Anyway, anyone who's willing to learn the Rubik's cube simply has to have the dedication to practice, no matter if you're using a method with 0 or 1211 algorithms.


Powered by YaBB 1 Gold - SP 1.4!
Forum software copyright © 2000-2004 Yet another Bulletin Board