Before you chase the clock
CFOP stands for Cross, First Two Layers, Orientation of the Last Layer, and Permutation of the Last Layer. This page is a training route: it explains what to practise, when to add a new skill, and how to tell whether a phase is becoming reliable. The algorithm dictionary is the separate reference for exact cases, formulas, and animated executions.
Start here after you can finish a cube without instructions. Keep the beginner method as a fallback while you replace one stage at a time. Your first target is not a lucky personal best; it is a repeatable solve with fewer pauses.
Step 1 – Train the Cross
Build the first-layer cross on the bottom so the top layer remains visible for F2L. During inspection, locate all four cross edges, decide their relative order, and predict where at least one first-pair piece will finish. Execute only after you have a plan.
- Use unlimited inspection and describe the four-edge plan aloud.
- Solve only the cross, stop, and check every side color against its center.
- Repeat the same scramble until the cross is smooth, then move to a new scramble.
Move on when you can plan the whole cross before turning on most scrambles. An eight-move cross is a useful long-term target, but complete planning matters before move count.
Step 2 – Build Intuitive F2L
F2L pairs a first-layer corner with its matching middle-layer edge, then inserts both pieces into one slot. Learn the piece relationships before memorising cases: identify whether the stickers face the same or opposite directions, create a pair in the top layer, and protect solved slots while inserting it.
Isolate one pair
After the cross, solve a single corner-edge pair and reset. This removes the pressure of a full solve while you learn how pieces connect.
Reduce rotations
Choose an empty slot as working space and practise seeing rear slots. Count rotations after each attempt and remove them one at a time.
Develop lookahead
Turn slowly enough to track the next pair while finishing the current one. Continuous turning is the goal; speed follows later.
Use a metronome or deliberate one-turn-per-second pace. If you pause, restart the pair more slowly instead of rushing the remaining moves.
Move on when common pairs feel understandable rather than guessed and you can finish F2L without returning to the beginner middle-layer procedure. The dedicated F2L case library is still planned; for now, use the available CFOP reference for terminology and keep this phase intuitive.
Step 3 – Recognise OLL Edges
Begin last-layer training with 2-look OLL. Its first look forms the top cross from one of four states: solved, line, L, or dot. Recognition comes before execution—pause, name the shape, orient it consistently, and only then perform the response.
- Generate a last-layer state and identify the edge shape without turning.
- Point to the front face you will use before beginning the algorithm.
- Record recognition mistakes separately from execution mistakes.
Move on when you identify line, L, and dot without rotating the whole cube. The animated beginner OLL cases show the available formulas and viewing angles without turning this training page into a second dictionary.
Step 4 – Add OLL Corners
The second OLL look orients the four corners after the cross is complete. Learn one shape at a time. For each shape, train a fixed viewing angle, accurate execution, and recovery from an incorrect U adjustment before adding another case.
Start with the two shapes you encounter and recognise most easily. Mix old and new cases in short sets so a fresh formula does not displace a reliable one. A case is ready for timed solves when you can recognise it from multiple U angles and execute it correctly ten times in a row.
Do not start full OLL because you reached a particular solve time. Expand only when 2-look recognition is automatic and the extra learning will not destabilise F2L practice.
Step 5 – Place PLL Corners
PLL moves already oriented pieces to their final positions. In 2-look PLL, first inspect the side stickers of the corners. Decide whether one matching pair is present or the corners require a diagonal swap, then align the case to the angle you trained.
- Ignore the top stickers and scan the four side faces for a matching corner pair.
- State “adjacent” or “diagonal” before touching the cube.
- After execution, verify corner placement before moving to the edges.
Move on when the decision takes one inspection of the sides rather than several cube rotations. If the visual cue is unclear, compare it with the labelled beginner PLL case animations.
Step 6 – Finish PLL Edges
With the corners placed, determine how the last-layer edges must move. First look for a solved edge and place it at the trained reference side. If none is solved, distinguish the opposite-swap and adjacent-swap patterns. Finish with the smallest final U adjustment needed to align every side.
Practise PLL as a complete phase: recognise, align, execute, and adjust. Time the whole phase rather than the finger movement alone, because a fast formula cannot recover time lost to uncertain recognition.
Complete five correct PLL finishes in a row. Any recognition, execution, or final-adjustment error resets the count. Once accurate, repeat the set while recording the phase split.
Transitioning to Full CFOP
Keep 2-look OLL and PLL until their recognition is stable under a timer. Learn full PLL first: the case set is smaller, the patterns are easier to compare, and removing a look usually produces an immediate benefit. Add full OLL gradually after your cross and F2L practice remain consistent.
Build each training block around one bottleneck. Review twelve solves, identify the phase with the most pauses or errors, and spend the next session on that phase. The final time is evidence, but cross planning, F2L pauses, recognition delay, execution accuracy, and final adjustment tell you what to train next.