41 terms this course uses, defined once each. Inside a lesson, the first mention of any of them is underlined — hover it for the short version, or follow it here.
No term matches that
Six faces, twenty-six visible pieces, and three kinds of piece that can never trade jobs.
The single piece in the middle of a face. On a 3×3 it never moves, so it names the face's colour for the whole solve.
This is why every recognition cue on this site is written against the centres: they are the one thing that cannot lie to you. On a 4×4 there is no single centre piece — you build each face's centre out of four, which is the first thing that puzzle asks of you.
A two-coloured piece between two corners. A 3×3 has twelve.
A three-coloured piece at a vertex. A cube of any size has exactly eight.
An edge can never become a corner and a corner can never become an edge, whatever you do to the cube. That is the unbreakable rule the first lesson is about, and it is why the method can solve one kind of piece at a time.
The place a piece belongs, named by the faces around it — the front-right slot, the UF slot.
One coloured square. A corner shows three of them, an edge two, a centre one.
One letter, one quarter turn — and the handful of modifiers that go with it.
A fixed sequence of turns you run as a unit to reach a known result.
Every algorithm on this site is machine-verified against a cube simulator before it ships, so a wrong one cannot reach you.
A short algorithm your hands run as one motion rather than as separate letters — the building block bigger algorithms are made of.
The trigger R U R' U'. This course calls it righty, and the whole beginner method is built from it.
A turn of a middle layer, between two outer faces — M, E and S on a 3×3.
A turn that takes two or more layers at once, written lowercase or with a w — r, Rw, 3Rw.
Wide turns are what a big cube adds to your notation, and the layer count matters: Rw takes two layers, 3Rw takes three. The 4×4 and 5×5 parity algorithms differ by exactly one such token.
Turning the whole cube rather than a layer — x, y and z. Nothing is solved by a rotation; it only changes what you are looking at.
Adjust the Upper Face: the free U turn that lines a case up before you start, or squares the top up afterwards.
It is free because it costs one move and never disturbs anything below the top layer — which is why case pictures on this site are not normalised to a single angle.
Moving your hands mid-algorithm to reach the next turn. Fewer regrips, faster solve.
Take a piece away, do something, put it back — the shape behind every big-cube centre insert.
The setup moves undo themselves, so everything the grab disturbed comes home except the piece you meant to move.
The steps every method shares, and the names this course uses for them.
Four edges of one colour placed around their centre, with their side colours matching. The first step of the solve.
The cross plus the four corners under it — one whole face and the row beneath it.
The four edges between the first and last layers.
The final face and the row around it — the four edges and four corners left when the first two layers are done.
First Two Layers: solving each corner and its middle-layer edge together as a pair, instead of one layer then the other.
It roughly halves your first-two-layers move count and is the single biggest change between the beginner method and CFOP.
Orient the Last Layer: make the whole top face one colour, ignoring where the pieces sit.
Two-look OLL does it with ten algorithms; full OLL does it in one look with 57.
Permute the Last Layer: slide the top pieces into their places, once they are all facing the right way.
Two-look PLL does it with six algorithms; full PLL does it in one look with 21.
Orient the Corners of the Last Layer — the corner half of two-look OLL, seven cases.
Turn a piece so its sticker faces the right way, without caring where it sits.
Move a piece to where it belongs, without caring which way round it is.
The fifteen seconds before the timer starts, spent planning — enough for the whole cross and often the first pair.
Watching for the next piece while your hands are still finishing this one. Pauses, not turn speed, are where solves lose time.
A sequence of turns that mixes the cube up, or sets up one specific case to drill.
What to look at, and the shapes worth having a name for.
Two corners on one face showing the same colour, with a different colour between them — a pair of headlights facing you.
Headlights are the fastest thing to spot on a last layer, which is why most PLL recognition starts by counting them.
Three stickers of one colour in a row across a face — a solved side of the last layer.
The one-yellow-corner algorithm, R U R' U R U2 R'. The most useful single algorithm in the last layer.
The corner cycle R U' L' U R' U' L: it holds one top corner still and rotates the other three.
The two corners that need to trade places share an edge of the top face.
The two corners that need to trade places sit across the top face from each other.
A 4×4 and a 5×5 are a 3×3 wearing extra pieces. These are the extra pieces.
The big-cube method this course teaches: build the centres, pair the edges, then solve the result as a 3×3.
Everything you already know is the last step. What a big cube adds is the two steps before it, plus parity.
A faster ordering of reduction: two centres, the cross, the last four centres, then the edges — so the cross is already done when you reach the 3×3 stage.
Short for double edge: the two or three pieces that behave as one edge on a big cube, once they are joined.
A 4×4 dedge is two wings; a 5×5's is two wings plus a midge, and it is often called an edge group there. Either way it is what a big cube has instead of a single edge piece — and joining all twelve of them is what reduction is.
One of the two outer pieces of a big-cube edge pair. A 4×4 and a 5×5 both have 24.
The middle piece of a 5×5 edge group — a true edge piece, which is why a 5×5's last layer obeys 3×3 law.
Pairing several edges at once by holding a slice out and letting the free layer feed pieces in.
A state a 3×3 can never reach, but a big cube can — a flipped edge pair, or two edge pairs swapped. It needs its own algorithm.
Parity is not a mistake you made. It arrives in about half of all solves, because a big cube's identical-looking pieces let you finish reduction in a position an odd number of swaps away from solved.