Demonstrating Full PLL

There's more than one way to solve a Rubik's cube, but typically, with the Beginner's Method, you follow these steps:

(first layer)
1. solve the edges (cross)
2. solve the corners

(middle layer)
3. solve the edges

(last layer)
4. orient the edges (cross)
5. orient the corners (face)
6. permute the corners
7. permute the edges

You can solve the first layer intuitively (without memorization) pretty easily, but there are also some simple tricks you can learn (e.g., daisy method, corner shimmy) if you don't want to use your brain. The middle layer requires learning two algorithms that are very similar (just mirrored), and will need to be applied multiple times. The last layer can be solved in four steps, requiring the memorization of 5 algorithms of varying difficulty, that may need to be applied more than once in any given solve. So, you can solve a Rubik's cube - every single time, no matter the scramble - only needing to memorize as little as 7 or 8 algorithms (2 of which are just mirrored versions of others).

It's neither the flashiest nor the most efficient way to solve the puzzle, but it's one of the easiest to learn. More difficult methods typically require you to memorize more algorithms to solve individual cases in a more tailored fashion, with less repetition. Beginner's CFOP, which I learned last fall, reduces your solve from seven to six steps, only one of which requires any repetition at all (and not even every time). After solving the cross, you can solve the first two layers (F2L) simultaneously, by pairing edges with corners. This can be done intuitively, once you learn the patterns. The last layer still requires four steps, but by learning 14 algorithms (plus one that is only a minor variation of another one), you can solve it every time by executing only 4 of those algorithms.

The advantage of Advanced CFOP is that it reduces the last layer from 4 down to 2 steps (orientation of the last layer, or OLL, followed by permutation of the last layer, or PLL), so that for every given scramble, assuming you're solving the first two layers intuitively*, the cube can be solved with the execution of only two algorithms! You can see why it's so much faster. The downside is that in order to do this, you have to commit 78 different algorithms to memory (57 for OLL, 21 for PLL) - although this includes the 14 you already learned. Yeah, it's a lot. Is it worth it? If you're a casual cuber, then almost certainly not. But if you want to be in the big leagues, it's mandatory. Besides, if you enjoy cubing, and practice every day, it gives you something new to learn. Just think of the feeling of mastery it will impart once you're done.

I've got a long way yet to completing Advanced CFOP. But I just reached a notable milestone - I have committed each of the 21 algorithms required for full PLL to memory. I bought a bulk order of 20 speedcubes to celebrate, so I could demonstrate my progress!



*I've been doing advanced intuitive F2L for so long I apparently forgot what the beginner's version was like. In F2L, you're looking for a first layer corner and a middle layer edge with matching colors, so you can pair them up and then slot them where they belong (repeating the process until all four corner-edge pairs are solved).

In the beginner's version, you first put the matching corner and edge in the top layer, and then pair them up according to one of just three variations. The advanced version streamlines the process (again, by learning more cases) by learning how to most efficiently match each pair even when they're not both in the top layer. I think there's about thirty different cases to learn (including all the mirrors). It was rough going at first, but I laid them all out in my head (with the help of diagrams I made on my computer), and practiced, practiced, practiced until I was able to remember how to solve each case without looking it up.

This is still the slowest part of any solve for me (which is typical among speedcubers, from what I can tell), because it takes time to hunt for the pieces all over the cube. That's why I'm practicing look-ahead, where instead of looking at what you're doing, you rely on muscle memory and focus on figuring out what you're going to do next, to reduce time wasted on the recognition phase.