How to Learn the CFOP Method
Reviewed by CuberPal Editorial Team · Updated 2026-07-07 · Learn CFOP learning path
Quick answer
CFOP solves the cube in four steps: Cross (bottom edges), F2L (first two layers in pairs), OLL (orient the last layer), and PLL (permute the last layer). Learn it in this order: efficient cross → intuitive F2L → 2-look OLL and 2-look PLL (about 16 algorithms) → full PLL → full OLL. Most cubers make the switch from the beginner method in 2–4 weeks.
CFOP (also called Fridrich) is the method used by the overwhelming majority of the fastest speedcubers in the world. If you can already solve the cube with the beginner layer-by-layer method, CFOP is your next step — it removes wasted moves and scales all the way to world-record speed.
The mistake most people make is trying to learn all 78 algorithms at once. Here is the order that actually works.
The four steps of CFOP
- Cross — solve the four bottom-layer edges. No algorithms; pure planning. Goal: 8 moves or fewer, eventually planned entirely during inspection.
- F2L (First Two Layers) — pair up each corner with its matching edge and insert them together. Intuitive at first; 41 standard cases exist but you discover most solutions yourself.
- OLL (Orientation of the Last Layer) — make the top face one color. Full OLL is 57 algorithms; the 2-look version needs just 10.
- PLL (Permutation of the Last Layer) — move the top pieces to their final spots. Full PLL is 21 algorithms; 2-look needs 6.
The learning order that works
Phase 1 (week 1): Cross + intuitive F2L. Do slow, untimed solves figuring out pair insertions yourself. Your times will get worse before they get better — everyone goes through this dip; push through it.
Phase 2 (weeks 2–3): 2-look OLL (10 algorithms) and 2-look PLL (6 algorithms). This "4-look last layer" completes a functional CFOP solve and typically lands you at 45–60 seconds.
Phase 3 (months 2–3): Full PLL, a few algorithms per week with daily review. This alone usually brings you into the 25–35 second range.
Phase 4 (when you are pushing sub-20): full OLL, better F2L solutions, and lookahead. See our sub-20 guide for that stage.
How to practice each stage without forgetting
New algorithms decay fast: learn five PLLs on Monday and by Friday two are gone. The fix is spaced repetition — review each algorithm at growing intervals (1 day, 3 days, a week), and drill the ones you fumble more often than the ones you know.
Pair that with case recognition practice: seeing a last-layer pattern and instantly knowing which algorithm it is matters as much as finger speed.
Learning CFOP with CuberPal
CuberPal includes a structured CFOP course that follows exactly this roadmap, so you never wonder what to learn next:
- 1Open the Learn tab and start the CFOP track: dedicated modules for Cross, F2L, OLL, and PLL with progressive lessons (there is a Beginner Method course too if you are not there yet).
- 2Work through F2L lessons that teach efficient pair insertions without unnecessary rotations.
- 3Browse the full algorithm catalogues — 41 F2L, 57 OLL, 21 PLL cases — each with a 3D cube preview and tap-through playback.
- 4Mark algorithms as learned to track your progress through each set, and let spaced repetition schedule your daily reviews.
- 5Once you complete the switch, record a solve for AI analysis to see your new phase times and what to improve next.
Frequently asked questions
How long does it take to learn CFOP?▾
A functional CFOP solve (intuitive F2L + 4-look last layer) takes most people 2–4 weeks. Full PLL adds another month or two. Full OLL is an ongoing project most cubers finish on the way to sub-15.
Should I learn CFOP or Roux?▾
Both reach elite speeds. CFOP has far more learning resources, more standardized algorithms, and is easier to get help with — which is why most beginners choose it. CuberPal focuses on CFOP and also includes a Roux overview with a CMLL reference if you want to explore.
How many algorithms does CFOP require?▾
A minimum of 16 (2-look OLL + 2-look PLL) to solve with CFOP, 78 for "full CFOP" (57 OLL + 21 PLL). F2L is usually learned intuitively rather than memorized.