# Thread: People's Rubic's Cube strategies

1. ## People's Rubik's Cube strategies

For people who can solve Rubik's Cube within a few minutes, how common is it for them to have figured it out on their own as opposed to getting tips or reading instructions?

2. Originally Posted by BarIII
For people who can solve Rubic's Cube within a few minutes, how common is it for them to have figured it out on their own as opposed to getting tips or reading instructions?
Pre internet, almost all of them. Now, almost none of them

3. Then the pre-internet guys are pretty impressive. I'd like to read some well written accounts of how people figured it out on their own. I figured out a move when I was a kid that helped me get one side. As an adult I figured out another couple of moves that got me two squares away from solving it after like 30 minutes.

4. Originally Posted by Sistamatic
Pre internet, almost all of them. Now, almost none of them
I disagree. I think that more people can solve it now with the internet than they could before, but starting with the people who can solve it reliably in a few minutes, I'm sure the ratio is pretty close in terms of how many used instructions and I'm sure the ratio is pretty high.

I think solving the rubix cube once is doable for someone with decent spatial abilities and with a lot of perseverence ... if you have the idea of solving the bottom two layers first, that's very doable without instructions. Afterwards if you find some ways to play with the top layer while being able to return the bottom two to the solved state, then you'll eventually get lucky (I'm not sure how many permutations are left with the bottom two layers fixed but I guess it's not that many and in any case, it would not be hard to do better than random).

On the other hand, independently developing a method to solve the cube in a couple of minutes say, would be quite a bit more work. I guess you'd need to do like above ... play with ways to permute the top layer while preserving the bottom two, taking notes on how the top layer changes and then figuring out a way of combing those things. I met a Russian girl here in Chile who could solve it in about four or five minutes using her own method. She bascially solved the bottom two layers and then had two recipes to solve the rest. That for me was impressive. She told me it took her about 15 hours to figure it out. But she's like a postdoctoral CS researcher ... not a "normal person", haha. I also saw her solve a 4x4 cube (one of th guys at work has a big collection) during a three hour talk for the first time without instructions, but as she explained, a lot of the methods for the 3x3 can be applied to a 4x4 once you do some "pre-processing" of the cube, which is to group the centres and the pairs of non-corner edges. This sort of "reduction" is very typical in CS.

In my case, I think I solved the cube in college maybe once or twice without instructions but could not do it reliably. Then I got instructions on the "beginner's method" and could solve it in a few minutes. Then I forgot about it for like a decade. Some homeless guy outside the university recently bought one for me again so I've been practising a bit again, can do it in about two minutes (following instructions for a standard beginner's method).

One of the students in a previous class I taught was a speed cuber who could do it in like 11 seconds. He used to bring a cube everywhere with him and just play with it in one hand.

/

edit: just to put things in context, it took Rubik himself a month to solve the first cube.

5. Good. I'm smart again.

6. I just googled the algorithm, and even then it took me a while to develop the muscle memory to solve a rubik's cube... as for solving it by myself... ain't nobody got time (or mental ability) for that

7. Oh! Well, I quickly peel off each colored square, then I "fix" the problem

8. I solve a 3x3 in just under a minute on average (I'm a casual cuber taking a slow, lazy road toward speed cubing, you might say). I learnt through a mix of self-taught and reference-material-using honing/refining. Basically I looked at overviews of the approaches used, and then worked to "reverse engineer" the moves ("algorithms") used to accomplish the various incremental effects. For instance, I heard many methods start by solving the "top", often "white" layer, and so I set about figuring out how to do that. Then came middle layer (edge pieces, etc). And so on. For me the last layer was (obviously enough, I figure) the most complex to sort out (since the moves must not render the prior two layers unsolved, etc). My resulting method is basically a tweaked version of what many tutorials give as a "beginner" method (layer-by-layer), with a dash of opportunistic F2L (I'm slowly transitioning into that approach -- evolving the latter from the former very lazily).

Edit: the "lazy" aspect of my progress from here could be explained differently/more completely, actually. I refuse to just follow algorithms given. I look at the effect of the moves (for instance, permute this, swap that, cycle here, etc) and then try to reverse engineer/rediscover the moves myself -- so that I understand them. This is similar to how I go about (not) using libraries in software development if/when I can avoid it -- I'll study how they work, their API, their effects, inputs and outputs, etc -- and then set out to implement my own, so that I can properly understand not just what but why/how. This is how I go about cubing. So it turns out, for me anyhow, that the algos used for many speed cubing methods involve essentially dozens of long strands of otherwise rote movements -- that are neither the sort of thing I care to remember, nor have the available time or energy to sit down and reverse-engineer. (By contrast, the method I use involves 5 or so sequences, applied under only about as many patterns/cases). So my "opportunistic F2L" above refers to scenarios I've worked out for myself -- the simpler-to-spot and/or resolve cases (3 of them, so far) -- where going for a F2L corner+edge move I've figured out is bound to shave a few seconds off of the "flat" layer-by-layer sequences I know otherwise.

9. Originally Posted by Lurker
Oh! Well, I quickly peel off each colored square, then I "fix" the problem
That's so dainty of you!

Here's how I do it:

Spoiler: Don't try this at home

10. This could be in the "feel old" thread but I'll put it here.

I had a Rubic's cube BEFORE the internet. No way I could figure out how to solve it. I was still in grade school. But back then, schools in my area had this program where a couple of times a year, students could by books from a special catalog brought in by the school. They were interesting books designed to appeal to young readers that you wouldn't see normally in a bookstore. Anyways, one was on how to solve a Rubic's cube so I bought it.

The method it taught was to solve the corners first, then the middle pieces. So it gave you some moves to get either a corner or middle piece moved without disrupting other squares already in place. I doubt this was an efficient method. Took a lot of moves. But at least I got the cube which had been haunting me since I first mixed up the colors back in order.

#### Posting Permissions

• You may not post new threads
• You may not post replies
• You may not post attachments
• You may not edit your posts
•