Skip to main content
BEGINNER'S GUIDE

CFOP for Beginners: How to Start Learning Speedcubing's Most Popular Method

By Trevor · Published May 4, 2026 · 15 min read

CFOP is the method used by the vast majority of serious speedcubers, including most of the world's top competitors. If you can already solve a Rubik's cube but you're still using a beginner method, this guide explains what CFOP is, how it's structured, what you actually need to learn, and in what order.

The word "CFOP" intimidates a lot of people. They hear about 57 OLL algorithms and 21 PLL algorithms and assume it requires months of memorization before you can use it. The reality is more practical: you can start using CFOP with a small subset of algorithms, get measurably faster within a few weeks, and add more algorithms gradually over months. The full algorithm set is a destination, not a prerequisite.

I learned CFOP the hard way, self-taught, in fits and starts, when I started cubing back in 2016. I began with a beginner method like everyone does, then gradually swapped in CFOP pieces over months, in roughly the order below. This is the sequence I wish someone had handed me on day one; it would have saved me a lot of wasted effort.

What CFOP Stands For

CFOP is an acronym for the four stages of the solve:

C Cross → F2L First Two Layers → OLL Orient Last Layer → PLL Permute Last Layer
The four stages of CFOP, executed in sequence on every solve.

Who Invented CFOP?

CFOP was primarily developed and popularized by Czech speedcuber Jessica Fridrich in the 1980s. That's why you'll often hear it called the "Fridrich Method." She didn't invent every individual algorithm, many had been developed by other solvers, but she systematized the four-stage approach and published it online in the early days of the internet, which is how it spread globally.

Today CFOP is used by the overwhelming majority of sub-20-second cubers and essentially all sub-10-second cubers. It has been the dominant speedcubing method for over 40 years for good reason: it is among the most efficient methods known for solving a 3x3 in the fewest moves using humanly executable algorithms.

The Beginner's Version of CFOP: 4-Look Last Layer

Advanced CFOP solvers use a "2-look OLL + 1-look PLL" approach, which requires knowing all 57 OLL algorithms and all 21 PLL algorithms. That's a total of 78 algorithms, a daunting learning project.

But beginners don't need to start there. The standard beginner's version of CFOP uses what's called 4-look last layer:

With 4-look last layer you need roughly 16 algorithms total (some overlap). This is learnable in 2–4 weeks and will get most people from 45 seconds down to the 25–35 second range. It's the correct starting point.

The Four Stages in Depth

STAGE 1

The Cross

The cross involves placing four white edge pieces (or whichever color you choose as your cross color) so they form a plus sign on the bottom face, with each edge's side color matching the center of the adjacent face. This is the only CFOP stage with no fixed algorithm set, it requires genuine problem-solving every solve.

Most beginners learn to solve the cross using intuition and a few basic patterns. An efficient cross takes 6–8 moves and executes in 1–3 seconds. If your cross is taking more than 4 seconds consistently, that's where time improvements start, see the Complete Guide to Cross Planning for the structured skill progression.

What to learn first: Practice building the cross on a scrambled cube without a tutorial. Start with the bottom face and figure out how to get each white edge piece into its correct slot using whatever moves feel intuitive. The goal at this stage is to develop a basic feel for how the cube works, not to execute a fixed sequence.

STAGE 2

First Two Layers (F2L)

F2L is the most time-consuming stage in a typical solve and the highest-leverage area for improvement. You solve the four corner-edge pairs (one for each slot between the cross edges) and insert them into the middle layer simultaneously, working from the top face downward.

The key insight is that each corner piece belongs with a specific edge piece, the corner and edge that share the same two non-white colors form a "pair." Your job is to spot that pair, get the two pieces together at the top of the cube, and then insert the assembled pair into its correct slot without disturbing what you've already solved.

For absolute beginners: Start by learning F2L intuitively rather than memorizing algorithms. Watch how 4–6 common pair positions work, understand the logic, and build your own solutions. Algorithmic F2L is faster but intuitive F2L is better understood, and understanding is what leads to genuine speed improvement.

Standard F2L has 41 cases. You don't need to know all 41 algorithmically to start getting faster. Learning the 15–20 most common cases first and handling the others intuitively is a perfectly valid approach for cubers averaging 30–40 seconds.

STAGE 3

Orientation of the Last Layer (OLL)

After completing F2L, the top layer contains all eight last-layer pieces but they're in various orientations, some corners have yellow facing the side, some edges have yellow facing the side. OLL's job is to orient everything so yellow faces up, creating a solid yellow top face.

Full OLL has 57 cases. For beginners, 2-look OLL reduces this to 10 algorithms:

  • 3 algorithms to orient the 4 edges first (getting a yellow cross on top)
  • 7 algorithms to orient the 4 corners once the edges are oriented

These 10 are the exact cases covered in the Top 10 OLL guide. Master these first. Full 1-look OLL (all 57 cases) is an intermediate-to-advanced milestone, not a beginner requirement.

STAGE 4

Permutation of the Last Layer (PLL)

After OLL, the top layer is fully oriented (all yellow faces up) but the pieces aren't in their correct positions, corners are in the wrong corners, edges are in the wrong slots. PLL moves them into their final correct positions to complete the solve.

Full PLL has 21 cases. For beginners, 2-look PLL uses 6 algorithms:

  • 3 algorithms to cycle the 4 corner pieces into their correct corners
  • 3 algorithms to cycle the 4 edge pieces into their correct slots

These 6 algorithms are the starting point. From there, adding the remaining 15 PLL cases one at a time is straightforward, most cubers complete full PLL within 1–3 months of starting CFOP. The PLL Recognition Shortcuts guide covers how to identify cases quickly once you know all 21.

What You Actually Need to Learn (in Order)

Here's the practical learning sequence that works for the vast majority of beginners transitioning from a layer-by-layer beginner method to CFOP:

Week 1–2
Cross practice. Do 20+ solves per day, focusing only on building the cross efficiently. Work toward consistently solving the cross in under 10 moves using inspection before your hands touch the cube. Don't move to F2L until your cross is reasonably intuitive.
Week 3–6
Intuitive F2L. Learn the basic logic of F2L pair-insertion intuitively. Start by solving each pair with any moves that work, then gradually recognize patterns. Expect your times to get worse before they get better, the transition from beginner layer-by-layer is normal.
Week 7–8
2-Look OLL (10 algorithms). Learn the three edge-orientation algorithms first (they're short), then the seven corner-orientation algorithms. Drill each until you can execute it in under 2 seconds.
Week 9–10
2-Look PLL (6 algorithms). Learn the corner permutation algorithms first, then the edge permutation algorithms. By the end of week 10, you should have a complete 4-look CFOP solve.
Month 3+
Expand algorithm count and drill recognition. Work toward full PLL (21 cases), then full OLL (57 cases). Simultaneously, focus on recognition speed, knowing an algorithm and recognizing when to use it are different skills, and recognition is usually the bottleneck.

How Fast Will You Get?

This is the question everyone asks and nobody gives a straight answer to. Here's an honest breakdown:

These ranges assume consistent daily practice (20–30 minutes per day). They're medians, not guarantees, some cubers move faster, some slower. What the data consistently shows is that algorithm count matters far less than recognition speed and F2L efficiency at the intermediate level. Most cubers would improve faster by drilling the 10 OLL cases they know until they're automatic, rather than rushing to learn 20 more.

A common beginner mistake: Trying to learn full OLL and PLL immediately. The standard advice, "learn all 78 algorithms before you can really use CFOP", is wrong and discourages more beginners than it helps. Start with 4-look, get comfortable, then expand. You'll be faster at 4-look after two months than most people are at partial full-algorithm CFOP after the same time.

Equipment You Need

You don't need specialized equipment to start learning CFOP, but your hardware does matter eventually. Here's the honest picture:

The Most Common Beginner Mistakes When Learning CFOP

Watching too many tutorials instead of solving. You can watch every CFOP tutorial on YouTube and still not get faster if you're not doing 50+ solves per day. The knowledge comes from the screen; the skill comes from the cube in your hands.

Learning F2L algorithms before understanding the intuition. If you start memorizing F2L case-by-case algorithmically before you understand why you're inserting a pair the way you are, you'll get stuck on unusual cases and every scramble will feel like an unknown. Learn the logic first, then add algorithms for efficiency.

Skipping the cross entirely. Most tutorials spend 30 seconds on the cross and 15 minutes on OLL. The cross is actually where many learners have the most room for improvement. A cross that takes 5 seconds is a real problem. A cross that takes 2 seconds is a meaningful competitive advantage.

Measuring progress only by single-solve times. Your best single solve is not your average. When tracking progress, always use an average of at least 5 solves (Ao5) or preferably 12 (Ao12). Single-solve times are too variable to be meaningful as a progress metric at the beginner level.

Ignoring recognition training. At some point, usually around the 30-second mark, you'll realize that you know all the algorithms but you're still pausing to identify cases. Recognition is a separate skill from execution, and it needs specific practice. The tools on this site were built specifically for this: use them alongside your regular solving practice.

The Role of Recognition Training in CFOP

The reason AI Speed Cube Trainer was built is that recognition is consistently the overlooked bottleneck in intermediate CFOP improvement. Knowing an algorithm and recognizing when to execute it are completely different cognitive skills. You can know all 57 OLL algorithms perfectly and still take 3 seconds to identify which one you need, which eliminates any advantage those algorithms give you.

Recognition training means drilling the ability to look at a last-layer position and know the case instantly, before you consciously process it. This is a pattern-matching reflex, not recall. Building that reflex requires hundreds of exposures in conditions that mirror real solving. That's what the OLL and PLL trainers on this site do: you input the pattern you see, get immediate feedback, and build the association between the visual pattern and the algorithm label.

For F2L, the same principle applies. F2L recognition (looking at the cube and instantly identifying where your corner and edge are, which case it presents, and what insertion to use) is a skill most CFOP beginners never practice explicitly. Using the F2L flashcard trainer develops this reflex in the same way.

Start building recognition now

The OLL, PLL, and F2L trainers are designed specifically for CFOP learners at every stage. Open the trainer and start with OLL, identify the case before you move.

Open the Trainer →

Where to Go from Here

Once you've got 4-look CFOP solid, the natural progression is:

  1. Cross planning: Move from improvised cross to planned cross. Read The Complete Guide to Cross Planning for the skill tree and drills.
  2. Full PLL: Expand from 6 algorithms to all 21. Most cubers complete this in 1–2 months of focused work. Start with the most common cases. U-perms, T-perm, Y-perm.
  3. Expand OLL: Add cases beyond the core 10, starting with the ones you encounter most often in real solves. Read The 10 Most Important OLL Cases first if you haven't already.
  4. F2L look-ahead: Work on eliminating pauses between pairs. This becomes the primary bottleneck once your algorithm execution is solid. Read Advanced F2L Look-Ahead Techniques.
  5. Sub-20 planning: When you're consistently averaging 22–28 seconds, the structured 30-day approach in How to Break Sub-20 in 30 Days gives you a targeted path to the next barrier.
T
Trevor
Speedcuber & Creator of AI Speed Cube Trainer

Trevor learned CFOP the slow way before switching to the recognition-first method he now teaches here; this is the guide he wishes he'd started with. Read the full story →

Further Reading & External Resources

These resources provide additional context for new CFOP learners looking for tutorials, community support, and algorithm references:

Related reading: The 10 Most Important OLL Cases Every Speedcuber Should Master First · The Complete Guide to Cross Planning · How to Break Sub-20 in 30 Days