A hidden lens for good cubing
Intended for cubers who average 30 seconds or faster.
There is danger. A young adventurer, Peaku, is trying to rescue his sister. In the heart of a labyrinth filled with monsters, he's found a clue to her whereabouts, and now he needs to escape before the whole place collapses.
Peaku chances upon a speed potion. It's tightly sealed with a cork. With the labyrinth crawling with monsters, it's dangerous to slow down to open a bottle. A monster might find him and slow him down even more. But the sooner he uncorks the potion, the sooner he can use it to escape.
So. When is the right time to drink the potion?
Cubers are in a similar predicament.
They can solve Edge Orientation (EO) right away and get a speed boost, but that stubborn step may cost them. They can delay and finish solving EO near the end, but then the solve is almost over already. Solving EO somewhere in the middle mixes the risks and benefits.
But, what is EO?
With edge orientation, we can define every edge on the cube as good or bad. An edge is good if it can be completely solved with justR, L, U, and Dmoves. Click on the two examples below to see what that looks like.
In the bad edge example, placing the edge in the correct location leaves it "flipped". The stickers on the edge do not line up with the centers. It's impossible to solve this kind of flipped edge withR, L, U, and D. Check for yourself!
This ability to put every edge into one of two categories (good or bad) wouldn't be interesting, except that good edges are usually faster to solve.
We can use the ends of our fingers to do mostU and Dmoves, and our wrists for L and Rmoves. It just so happens that we can use these four types of moves to solve any good edge. This is awesome since it means you can rapidly alternate between using your fingers and your wrists to solve good edges. Certain combinations of these moves are especially fast, likeR U R' U'.
We'll later explore how to deliberately useFand ymoves to turn bad edges into good edges, making the entire solve faster.
In the videos earlier, we had to attempt to solve the edge to check if it was good or not. It'd be a lot more useful if we didn't have turn the cube to check. Then there's no wasted moves. But to do this, we'd need to find some relationship between the colors of cube.
We'll focus on CFOP in this lesson, but these concepts will apply to any method.
The cube's centers don't move much in CFOP. What if we recognized EO by comparing the edges to the centers in some way? Which sticker on the edge would you look at? Consider this problem before you keep reading.
Going back to the good edge video, the red-green edge "connects" to the red center by doing U'. In other words, the red stickers line up on both pieces. Then, we didR'to solve the edge.
If we were forced to use F or B to solve it, that'd automatically make it a bad edge. So the edge has to be able to connect to the right or left center using R U L D moves for it to be a good edge.
Thankfully, since we only checking how pieces move with R U L D, there's only so many places that each sticker on one edge can be. The sticker is limited to certain "orbits". Let's look at how the sticker moves around. We'll color one sticker white and the other magenta. As you watch, remember that none of these moves change an edge's orientation.
In CFOP, this is useful for OLL, but especially F2L, for reasons we'll look at later.
Using this new orbit knowledge, we can make some statements about F2L edges and recognizing EO:
With the magenta orbit, we're basically making the opposite statement compared to the white orbit. This makes sense, since if one sticker connects to the left or right center, the other must not.
So which orbit do you use to recognize orientation? It's up to you. From the image above, you can see that the magenta orbit is more visible for top edges. For edges on the left or right sides, the white orbit is more visible.
If this is a bit confusing, don't worry. We'll go through some examples next. By the end of the lesson, you'll see several ways your new EO knowledge can help speed up your solves.
Find the red-green edge on the cube below. Is it good or bad?
Let's use the magenta orbit. The sticker on this edge that is within the magenta orbit is green. Green matches the front center color. That's enough to tell you it's a good edge, as long as you know that you're looking at an F2L edge.
We can confirm it's good by solving it with R U L D moves. Click the play button below.
Below is one more example. Again, we're looking at the red-green edge. But the edge is in the middle layer, so try using the white orbit this time. Good or bad?
The sticker on this edge that's within the white orbit is red. When we use this orbit, the sticker has to match the left or right center color. The red sticker doesn't match the right center that's next to it, but it does match the left. Another good edge. Let's verify:
These approaches show how one sticker can be used to determine EO. But you still need to know the color of both stickers. Otherwise, you might be looking at an OLL edge or a cross piece.
If a both stickers aren't visible, you can either try to remember what it was from past moves (lookahead, an advanced skill), deduce what the sticker is, or barring those options, tilt the cube to check directly. You should avoid rotating or doing moves to get information, since that is wasted effort and would slow you down.
If you know OLL, you already know how to orient last layer edges. You solve the top cross, which flips all the edges so that all of the same color is on top.
But it's useful to know that an OLL edge in the middle layer can still considered be oriented or not. Sometimes you'll try to influence OLL during F2L in various ways.
A correctly oriented OLL edge will always have the top center color in the magenta orbit. Knowing that, is the white-green edge good or bad? Try to remember the orbit diagram and how it was made rather than referring back to it.
R, which puts the wrong color on top.EO is useful for CFOP, but it is an integral part of the ZZ method. Learning ZZ provides great intuition for how EO really works. Check here2 for an excellent introduction to EO and ZZ by crystalcuber. His guide is the source of the idea of orbits and related concepts.
So far, we've only shown how to use R, L, U, and D moves to solve good edges. But if an edge is bad, what do you do?
Doing y rotations is the most typical way to deal with bad edges. If the F2L edge is in the top layer, rotating flips F2L edge orientation. Good F2L edges turn into bad edges, and bad become good.
Below, we show the same cube with different rotations applied. Only the good F2L edges are highlighted.
Notice how the red-green edge is highlighted in both images, and how the orange-green edge between the red and blue centers is highlighted in neither. If you apply the EO recognition rules from earlier, you'll see that rotating only flips EO of the top and bottom layer edges! The middle F2L edges are unaffected.
Due to this difficulty with altering their EO, bad F2L edges in the middle layer are some of the worst cases you can get. Often, you'll want to solve something else and destroy this case in the process.
If you do have to solve them, you either use F moves, or you can take the edge out of the middle layer before you rotate and then solve it. Which of these you choose can depend on the case, the pairs after it, and your own preferences.
Influencing OLL edges during F2L can be done in a structured way using algorithm sets like ZBLS1.
A less structured way is to sometimes use triggers like sledgehammer (R' F R F') to insert a pair and orient some edges. This technique is often called partial edge control. Compare the EO of OLL after inserting normally and after inserting with sledgehammer.
Using sledgehammer flips the orientation of the front and right edges. If those edges weren't oriented to start, they would be after doing sledge.
OLLs with more edges oriented tend to be slightly better. There are a couple other inserts worth considering for this reason:
You can also mirror these inserts to use lefty moves. You should start all of them with your thumbs on the front face of the cube, and do all of them without regripping. However, you don't have to use these.
The main downside of partial edge control is that it can add a pause. You can't always use these inserts because you won't always get a connected pair with certain edges not oriented. And then you might need to pause and determine what insert to use.
You can get better at determining what insert you will use in advance, eliminating the pause. But the amount of practice this takes is significant. You might decide it's not worth it for you.
Peaku runs through the labyrinth's hedges and comes toe-to-toe with an orc. They fight, and Peaku defeats it unharmed. He decides it's as good a time as any to drink that potion.
Peaku opens his bag to shattered glass and the last drips of the potion. He curses his luck.
A shard of the glass glimmers in the torchlight. Peaku picks it out and holds it in front of his eye. The world appears to open up. The glass has some property that makes the hedges look almost invisible! He heads for the exit without any more guesses or wrong turns.
Recognizing the orientation of the edge gives information on how to solve that edge. You can use this info to reduce the number of moves or rotations in your solves.
If your solution to a pair with a good edge doesn't use only RULD moves, you typically would have made a mistake. A good edge will never require you to:
Once you've identified a bad solution, you can experiment to try to find a better one, or look up an alg or tutorial.
On rare occasions, you might rotate or use F moves anyway. For example, with this case, you could rotate to avoid a longer, clunkier RUL solution.
Using EO this way may give you many things to fix. Fixing them will slow you down at first, but as you get familiar with the solutions, you will eventually get faster as well.
Pairs with good edges are typically faster to solve than bad-edge pairs, so when there's multiple pairs to solve, it's nice to solve good-edge pairs first.
Doing x, y, and z rotations in a solve is essentially time spent doing nothing. In a perfect world, only one y rotation would be needed in a solve. You can solve cross and all good-edge pairs, rotate to turn all the bad edges into good edges, and finally solve the remaining pairs. And indeed, EO is a powerful tool for you to see how to reduce unnecessary rotations and F moves.
But reality is more complicated. What if a very fast pair would require a rotation before all good-edge pairs are solved? What if you have multiple bad edges in the middle layer? The choice is yours.
Here's a hard example. Which pair is better to solve here: orange-blue, orange-green, or are they about the same? Why?
Once you've recognized an edge as good or bad, it's possible to keep that knowledge in your brain. At times, this can be easier than remembering both colors on the edge.
Say in the last example you decide to solve the orange-blue pair first. You see orange-green is a bad edge. Since you'll use RULD moves to solve orange-blue, you know, whatever happens to orange-green, it'll still be a bad edge. So you might choose to rotate immediately after you solve orange-blue even if you don't know exactly what case it is.
This example is a bit too simple, since you'll be able to keep track of orange-green in other ways just as easily. A more realistic example might be remembering the orientation of the red-green edge as you solve these two other pairs.
Additionally, keeping track of whether an edge is good or bad might make it easier to recognize the case. After all, you no longer need to look very closely at it.
A video by J Perm goes into much more detail on these ideas.3 He relates EO more strongly to lookahead concepts. Using EO with lookahead is pretty hard to do because lookahead itself is hard. You would have to remember details about an edge while you solve something else. If you are still working on pair solutions, it may make sense to first focus on how EO can help you improve efficiency and pair choice.