Introduction to the CFOP Method (Fridrich)
Learn the CFOP speedcubing method -- the most popular technique used by world-class speedcubers to solve the Rubik's Cube in under 10 seconds.
CFOP stands for Cross, F2L, OLL, PLL -- the four stages that make up the Fridrich method. Developed and popularized by Jessica Fridrich, CFOP is the dominant speedcubing method worldwide. Most sub-20 and sub-10 solvers rely on it. If you can already solve the cube with the beginner layer-by-layer method, transitioning to CFOP is the single biggest improvement you can make.
Cross
The solve begins by building a cross on the bottom face (typically white). Unlike the beginner method where you form the cross on top and flip it, advanced solvers work directly on the bottom, allowing them to look ahead for F2L pairs during the cross step. A good cross takes 8 moves or fewer and should be planned entirely during the 15-second inspection.
F2L -- First Two Layers
F2L is the most important step to master. Instead of solving the first layer corners and middle layer edges separately (as in the beginner method), you pair each corner with its matching edge and insert them as a single unit into the correct slot. There are 41 standard F2L cases, but most speedcubers learn them intuitively rather than memorizing algorithms. Efficient F2L is what separates a 60-second solver from a 20-second solver.
A basic F2L insertion: pair the corner and edge, then insert together.
OLL -- Orientation of the Last Layer
After solving the first two layers, you orient all pieces on the top face so the top color faces upward. Full OLL uses 57 algorithms to handle every possible case in a single step. Beginners can start with 2-look OLL, which splits the step into two stages (first form the cross, then orient the corners) using just 9 algorithms.
The Sune algorithm -- one of the most common OLL cases.
See 2-look OLL: three edge shapes, then the corners
Full OLL is 57 cases, but you can orient the whole last layer in two looks with about 10 algorithms. Look 1 — make the yellow cross. Read only the four edge stickers on top; you'll see one of three shapes:
Look 2 — orient the corners. With the cross done you're in one of 7 corner cases. Two algorithms cover all of them by repetition: Sune R U R' U R U2 R' and Anti-Sune R U2 R' U' R U' R'. Learn the 7 dedicated cases later for speed; Sune-spam works today. That's the on-ramp before the full 57.
PLL -- Permutation of the Last Layer
With the top face fully oriented, PLL rearranges all the top-layer pieces into their correct positions. Full PLL consists of 21 algorithms. As with OLL, beginners can start with 2-look PLL, which separates corner permutation from edge permutation using only 6 algorithms. Once you know full PLL, the last layer is always solved in two looks at most.
The T-Perm -- one of the most versatile PLL algorithms.
Transitioning from Beginner to CFOP
You do not need to learn everything at once. Start by switching to a bottom cross -- this alone will speed up your solves. Next, learn intuitive F2L by practicing pairing corners and edges without algorithms. Then add 2-look OLL and 2-look PLL to your toolkit. Once comfortable, gradually learn the full 57 OLL and 21 PLL algorithms. Within weeks of deliberate practice, you should see your times drop dramatically.
Speed & finger-trick tips
Raw algorithm knowledge only takes you so far — these techniques, drawn from Dan Harris's Speedsolving the Cube, are what turn a correct solve into a fast one.
The last-layer ladder: 4 looks → 3 looks → 2 looks
You do not have to learn 57 OLL algorithms before you can use full PLL. The last layer is best climbed as a ladder — each rung is usable on its own, and you upgrade over months, not in one sitting.
4-look (where you start)
Orient edges → orient corners → permute corners → permute edges. This is the beginner last layer — four small algorithm families, four separate looks. It works; it's just slow.
3-look (the smart stepping stone)
Learn the 7 corner-only OLL cases to orient the top in one look on top of the yellow cross, then do full PLL (one look). Cross → cross-edges → 7-case OLL → PLL. Harris's own recommendation: front-load PLL — 21 cases beats 57, and it's the part of the last layer most worth learning early.
2-look (full CFOP)
Full 57-case OLL (one look) + full 21-case PLL (one look). The expert finish: see the case, fire the algorithm, done. Beyond it lie VHF2L, COLL and ZBLL — last-layer influence you start during F2L — but those are years away and optional.
Getting faster: the honest progression
Speed is not faster fingers — it's fluency: a solve with no pauses. New solvers turn at full tilt, then freeze after every step to hunt for the next piece, and those freezes dominate the clock. The fix feels backwards: turn a little slower, slow enough that your eyes track the pieces, run the algorithm on autopilot, and use the freed attention to find the next F2L pair before your hands finish the current one. Creep the speed back up only once solving stays continuous.
The order that works: (1) bottom cross and intuitive F2L; (2) 3-look last layer; (3) drop OLL to 2-look, then learn full PLL; (4) chase fluency — muscle-memory algorithms free your eyes for look-ahead. What separates a sub-20 solver from a 40-second one is rarely more algorithms; it's an efficient, rotation-light F2L and a pause-free solve. F2L is where the time is — the last layer is just recognition speed.
You are ready to start CFOP!
Begin with a bottom cross, then practice intuitive F2L. Add 2-look OLL and PLL as you go. The full algorithm sets come later — focus on understanding first.
Practice with the Timer