How to solve a Rubik's Cube blindfolded: the Old Pochmann method
Solving the 3x3 blindfolded does not take a special memory or lots of algorithms. With the Old Pochmann method you only need three PLL algorithms, an alphabet that names every sticker and a few setup moves. Here is the complete method: the letters, the table for every sticker, parity and an example solved step by step.
R U R' U' R' F R2 U' R' U' R U R' F'
On this page
In short
- Blindfolded, you first study the cube with your eyes open, then cover your eyes and solve it without seeing it. In competition the time includes both phases.
- Old Pochmann moves the pieces one at a time. A fixed position, the buffer, is swapped with another fixed position, the target, always with the same algorithm: the T perm for edges and a modified Y perm for corners.
- To send the buffer piece anywhere else, you do a setup that brings that spot to the target, then the algorithm, then the setup in reverse.
- Every sticker has a letter. What you memorize is a list of letters: about 12 for edges and about 8 for corners in a normal scramble.
- If the number of edge letters is odd you have parity, and you fix it with an Ra perm between the edges and the corners.
What blindfolded solving is and how it is timed
Solving with your eyes covered (blindfolded, shortened to BLD, or 3BLD for the 3x3) has two phases. In the first you look at the scrambled cube and memorize where every piece is. In the second you put on a blindfold and solve it from memory, without looking at it again.
WCA competitions do not have the 15 seconds of inspection used in other events: the timer starts and then you lift the cover over the cube, so memorization time counts. While memorizing you can pick up the cube, but you cannot turn it or take notes. Before the first move your blindfold has to be on properly, and the judge also places a screen between your face and the cube. If the cube is not solved when you finish, the attempt is a DNF.
Since January 1, 2026, 3x3 blindfolded rounds are best of 5 attempts. Your best attempt is what ranks you, and the mean of the 5 also counts for rankings and records.
To practice at home, all you need is a sleep mask that blocks out the light and a timer. You can use the portal's timer and, to check the algorithms with your eyes open, the simulator.
You do not need to be fast with your eyes open. Old Pochmann only uses three algorithms and simple moves. What you need to train is accuracy: a single wrong move halfway through leaves the cube unsolved.
How the Old Pochmann method works
Stefan Pochmann came up with the method in 2004. At first it used a single algorithm, the T perm with setup moves, and he extended it that same year. The version people start with today, known as Old Pochmann, solves the pieces one at a time: first all the edges, then all the corners.
Everything revolves around two positions on the cube:
- Buffer: the position the pieces are sent from. For edges it is the top right edge (UR). For corners, the top back left corner (ULB, also written UBL).
- Target: the position the algorithm sends the buffer piece to. For edges it is the top left one (UL). For corners, the bottom front right one (DFR), specifically its sticker on the right face.
The algorithm swaps the buffer piece and the target piece. To send the buffer piece somewhere else, you first do a few preparation moves (the setup) that bring that spot to the target without touching the buffer. You do the algorithm and undo the setup. The buffer piece has reached its spot and there is a new piece in the buffer, which is the next one to move.
| Algorithm | Used for | What it swaps |
|---|---|---|
T permR U R' U' R' F R2 U' R' U' R U R' F' | Edges | Edges UR and UL. It also swaps corners UFR and UBR. |
Modified Y permR U' R' U' R U R' F' R U R' U' R' F R | Corners | Corners ULB and DFR. It also swaps edges UL and UB. |
Ra permR U' R' U' R U R D R' U' R D' R' U2 R' U' | Parity | Edges UL and UB, and corners UFR and UBR. |
The extra swaps each algorithm makes do not spoil anything. The two corners the T moves go back to their places with the next T, and the two edges the Y moves, with the next Y. Only when the number of letters is odd is one pair left swapped, and that is what parity takes care of.
Before you start, you should be able to solve the cube with your eyes open, for example with the beginner's method, and know the notation. The setups use two wide moves: l turns the left layer and the middle layer together, in the same direction as L, and d turns the bottom layer and the middle layer, in the same direction as D. They are also written Lw and Dw.
The method is slow: a full solve usually takes more than 300 moves. In return, you can understand all of it, and the faster methods are built on the same ideas.
The Speffz lettering scheme
To memorize, every edge and corner sticker gets a letter. The most widely used scheme is called Speffz: 24 letters for the edge stickers and another 24 for the corner stickers, from A to X.
Always hold the cube the same way: yellow on top and green in front, as in the rest of the guides. That puts red on the left, orange on the right, blue at the back and white on the bottom. The letters go face by face in this order: top (A-D), left (E-H), front (I-L), right (M-P), back (Q-T) and bottom (U-X).
You look at each face head on. The top face with the back face above it, and the bottom face with the front face above it. Corners start at the top left one and edges at the top one, and you go clockwise. Edges and corners use the same letters but are never mixed up: edge A is the yellow sticker of the UB edge, and corner A is the yellow sticker of the ULB corner.
Edge letters
| Edge | Colors | Letters |
|---|---|---|
| UB | yellow, blue | A: yellow (top); Q: blue (back) |
| UR | yellow, orange | B: yellow (top); M: orange (right) |
| UF | yellow, green | C: yellow (top); I: green (front) |
| UL | yellow, red | D: yellow (top); E: red (left) |
| FL | red, green | F: red (left); L: green (front) |
| DL | red, white | G: red (left); X: white (bottom) |
| BL | red, blue | H: red (left); R: blue (back) |
| FR | green, orange | J: green (front); P: orange (right) |
| DF | green, white | K: green (front); U: white (bottom) |
| BR | orange, blue | N: orange (right); T: blue (back) |
| DR | orange, white | O: orange (right); V: white (bottom) |
| DB | blue, white | S: blue (back); W: white (bottom) |
Corner letters
| Corner | Colors | Letters |
|---|---|---|
| UBL | yellow, red, blue | A: yellow (top); E: red (left); R: blue (back) |
| UBR | yellow, orange, blue | B: yellow (top); N: orange (right); Q: blue (back) |
| UFR | yellow, green, orange | C: yellow (top); J: green (front); M: orange (right) |
| UFL | yellow, red, green | D: yellow (top); F: red (left); I: green (front) |
| DFL | red, green, white | G: red (left); L: green (front); U: white (bottom) |
| DBL | red, blue, white | H: red (left); S: blue (back); X: white (bottom) |
| DFR | green, orange, white | K: green (front); P: orange (right); V: white (bottom) |
| DBR | orange, blue, white | O: orange (right); T: blue (back); W: white (bottom) |
Each letter names a sticker, not just a piece. That way a single letter tells you which spot a piece goes to and in which orientation. The buffer edge (UR) has the letters B and M; the buffer corner (ULB), A, E and R. Those letters are never memorized.
Learn the letters with a solved cube in your hand: point at a sticker and say its letter, and the other way round. Once you get them right without thinking, move on to memorizing scrambles.
Edges: T perm with the UR buffer
The T perm swaps the top right edge (UR, the buffer) with the top left edge (UL, the target). The top sticker of the buffer goes to the top sticker of the target, which is D. That is why the edge target is the letter D.
How to solve one letter
- Take the next letter and look up its setup in the table in the next section.
- Do the setup. The sticker for that letter ends up in the target, top left, facing up.
- Do the T.
- Undo the setup: the same moves in reverse order and in the opposite direction. The setup
d2 Lis undone withL' d2.
Edge setups only use L, D, l and d moves. None of them touches the UR buffer or the two corners the T moves (UFR and UBR). If a setup moved those corners, the T would swap two other ones and the next T would no longer put them back.
Example: the letter J
J is the green sticker of the front right edge (FR). The setup is d2 L: d2 takes that edge to the back left and L lifts it into the target with green facing up. Then comes the T, and you undo the setup with L' d2.
The target letters (D and E) also come up. D needs no setup: you do the T straight away. E has the setup L d' L.
Edge setup table
This table has the setup for every edge sticker. B and M are not in it because they are the buffer. For each letter: setup, T and undo.
| Letter | Sticker | Setup | Undo |
|---|---|---|---|
| A | yellow on UB | l2 D' L2 | L2 D l2 |
| C | yellow on UF | l2 D L2 | L2 D' l2 |
| D | yellow on UL | none | none |
| E | red on UL | L d' L | L' d L' |
| F | red on FL | d' L | L' d |
| G | red on DL | L' d' L | L' d L |
| H | red on BL | d L' | L d' |
| I | green on UF | l D' L2 | L2 D l' |
| J | green on FR | d2 L | L' d2 |
| K | green on DF | l D L2 | L2 D' l' |
| L | green on FL | L' | L |
| N | orange on BR | d L | L' d' |
| O | orange on DR | D' l D L2 | L2 D' l' D |
| P | orange on FR | d' L' | L d |
| Q | blue on UB | l' D L2 | L2 D' l |
| R | blue on BL | L | L' |
| S | blue on DB | l' D' L2 | L2 D l |
| T | blue on BR | d2 L' | L d2 |
| U | white on DF | D' L2 | L2 D |
| V | white on DR | D2 L2 | L2 D2 |
| W | white on DB | D L2 | L2 D' |
| X | white on DL | L2 | L2 |
You do not have to learn the table by heart. The setups follow a few ideas:
- Middle layer edges (F, H, J, L, N, P, R, T): if the edge is not already on the left, a
dmove takes it there, and anLmove lifts it into the target. - E and G, the red stickers of UL and DL: three moves,
L,dandLin some direction, so the piece goes in with the right sticker facing up. - White stickers on the bottom (U, V, W, X): a
Dmove puts the edge bottom left andL2lifts it. - Edges UB, UF, DF and DB (A, C, I, K, Q, S): an
lmove takes them through the middle layer to the bottom,Dputs them bottom left andL2lifts them. - O, the orange sticker of DR, is the only one with four moves.
Before doing a setup, check that it does not turn the top or the right face. If it did, it would move the buffer.
Corners: modified Y perm with the ULB buffer
Corners follow the same idea with another algorithm. The modified Y perm is the Y perm without its first move (F) and its last one (F'). It swaps the buffer corner, top back left (ULB), with the bottom front right one (DFR). The top sticker of the buffer goes to the orange sticker of DFR, which is P. The corner target is the letter P.
Corner setups only use R, F and D. None of them touches the buffer corner or the UL and UB edges the Y moves. All of them have two moves at most.
| Letter | Sticker | Setup | Undo |
|---|---|---|---|
| B | yellow on UBR | R D' | D R' |
| C | yellow on UFR | F | F' |
| D | yellow on UFL | F R' | R F' |
| F | red on UFL | F2 | F2 |
| G | red on DFL | D2 R | R' D2 |
| H | red on DBL | D2 | D2 |
| I | green on UFL | F' D | D' F |
| J | green on UFR | R2 D' | D R2 |
| K | green on DFR | F D | D' F' |
| L | green on DFL | D | D' |
| M | orange on UFR | R' | R |
| N | orange on UBR | R2 | R2 |
| O | orange on DBR | R | R' |
| P | orange on DFR | none | none |
| Q | blue on UBR | R' F | F' R |
| S | blue on DBL | D' R | R' D |
| T | blue on DBR | D' | D |
| U | white on DFL | F' | F |
| V | white on DFR | F' R' | R F |
| W | white on DBR | R2 F | F' R2 |
| X | white on DBL | D F' | F D' |
Example: the letter X
X is the white sticker of the bottom back left corner (DBL). The setup D F' brings it to the target with white facing right. Then comes the modified Y, and you undo the setup with F D'.
Pay attention to the direction of each setup move. D F' and D F bring the same corner to the target, but with a different sticker on P: the first solves X, the second another letter of the same corner.
Cycles, flipped edges and twisted corners
You do not turn anything while memorizing: you follow the path of the stickers. Look at the top sticker of the buffer. Its color and the color of the other sticker on that piece tell you which spot it goes to: that is the first letter. Then you look at the sticker now sitting in that spot, see where it goes, and that is the second letter. And so on.
When you reach the buffer: the cycle break
Sooner or later you reach a sticker of the buffer piece. If there are pieces left to solve, you need a cycle break: pick a piece that is not in its place and does not have letters yet, and write down the letter of one of its stickers. From there you follow the path as before, and when you get back to that same piece you write down the letter of the sticker you land on, and the cycle closes.
Pieces that are already in place and correctly oriented get no letter. Memorization ends when every piece has its letters or is solved.
Flipped edges
An edge can be in its place but flipped. You treat it as a cycle break and both of its sticker letters come up. For example, if the top front edge (UF) is flipped, you write down C I. Those two letters flip that edge and the buffer too. If two edges are flipped, their four letters fix both of them and the buffer: C I U K fixes flipped edges UF and DF.
Twisted corners
A corner twisted in its place works the same way. You write down the letter of one of its stickers, look at which color is on it and write down the sticker of that same corner where that color belongs. C M twists the top front right corner (UFR) and the buffer too. Two corners twisted in opposite directions come out as four letters, for example C M D F.
If a single flipped or twisted piece is left at the end of memorization, the buffer is flipped or twisted too. The two letters of that piece fix the buffer at the same time.
Parity: an odd number of letters
Every T also swaps two corners, and every modified Y, two edges. If the edge letters are even, those corners go back to their places. If they are odd, when you finish the edges the top right corners (UFR and UBR) are left swapped. That is parity.
On the cube, the number of edge letters and the number of corner letters are always both even or both odd. If one is odd, so is the other. It happens in roughly half of all scrambles.
The fix is to do an Ra perm right after the last edge and before the first corner:
- It puts back the two corners the last T left swapped. That way the corners are as they were in the scramble and what you memorized still holds.
- It swaps edges UL and UB. Since the corners are odd as well, the last modified Y swaps those two edges again and they end up correct.
The order is fixed: edges, Ra if the edges are odd, then corners. If you do the Ra at the end or skip it, the cube will not be solved.
You know whether there is parity during memorization: count the edge letters. If the number is odd, add something to remind you, for example an image at the end of the story.
Full example: memorizing a scramble
Do this scramble with yellow on top and green in front: L2 U F L2 D2 R' B' F2 D2 B R' B2 R U B F D2 B D2 F2
Edges
You start at the buffer (UR). Each row shows the colors you see on the current sticker, that sticker's color first, and the letter it goes to.
| You look at | You see | Letter |
|---|---|---|
| Buffer, top | orange (with green) | P |
| P | green (with yellow) | I |
| I | red (with yellow) | E |
| E | blue (with white) | S |
| S | white (with orange) | V |
| V | green (with white) | K |
| K | yellow (with blue) | A |
| A | white (with red) | X |
| X | orange (with yellow): it is the buffer piece | The cycle closes |
| Cycle break: you pick the FL edge, unsolved | F | |
| F | blue (with orange) | T |
| T | green (with red): back to the FL edge | L |
The blue and red edge is already in place, so it gets no letter. That gives 11 letters: PI ES VK AX FT L. That is odd, so there is parity.
Corners
| You look at | You see | Letter |
|---|---|---|
| Buffer, top | orange (with yellow and blue) | N |
| N | white (with green and orange) | V |
| V | red (with white and blue) | H |
| H | orange (with yellow and green) | M |
| M | white (with green and red) | U |
| U | blue (with white and orange) | T |
| T | red (with green and yellow) | F |
The seven corners that are not the buffer form a single cycle: NV HM UT F. That is 7 letters, odd, just like the edges.
Execution
- Edges, letter by letter:
d' L', T andL dfor P;l D' L2, T andL2 D l'for I; and so on up to L. - Parity: since there were 11 edge letters, one Ra perm.
- Corners, letter by letter:
R2, Y andR2for N;F' R', Y andR Ffor V; and so on up to F.
Check your memorization with your eyes open: do the scramble, solve letter by letter while looking at the cube and make sure every piece reaches its place. That way you catch reading mistakes before you cover your eyes.
How to memorize the letters
A list of 20 separate letters is hard to remember. The usual approach is to group them in pairs and turn each pair into something meaningful.
- Words and images: each pair becomes a word that starts with those letters or contains them. PI can be pirate; ES, espresso; VK, vodka; AX, axe; FT, football. Then you link the images into a short story: a pirate drinks an espresso and a vodka, picks up an axe and kicks a football at a lion (the L on its own). The stranger and more visual it is, the better you remember it.
- Audio: you repeat the letters as syllables, with a rhythm, like a phone number. NV HM UT F can be said as "navy, hum, ut, eff". It works well with short lists like the corners.
- Fixed word list: over time, every possible pair always gets the same word. That way you do not invent anything during the attempt and you memorize faster.
A common way to organise it is to memorize the edges first with images and then the corners with audio. You solve the edges first, and the corner audio, the last thing you memorized, is still fresh when you get to them.
To train your memory without the cube, write down 10 or 12 random letters, memorize them in pairs and repeat them a minute later. Gradually increase the number. To train reading the cube, memorize scrambles without solving them and compare your letters with the ones you get when solving with your eyes open.
Go over the story once before putting on the blindfold: if any part does not come out smoothly, look at the cube again. In competition you can take off the blindfold to look again, as long as you have not made the first move.
Practice plan, common mistakes and next steps
Practice plan
- Learn the T, the modified Y and the Ra until you can do them without thinking. They are in the PLL algorithms guide and you can drill them with the algorithm trainer.
- Learn the letters with a solved cube.
- Solve only the edges with your eyes open: memorize, then do each letter while looking at the cube.
- Do the same with the corners and add parity.
- Do full solves with your eyes open, but without using what you see to decide anything: follow only the list you memorized.
- Put on the blindfold, without a timer. If you get a DNF, redo the scramble with your eyes open and find where you went wrong.
- Once most of your attempts succeed, start timing yourself.
Common mistakes
- Undoing the setup wrongly. This is the most common mistake: the inverse of
l D' L2isL2 D l', notl' D L2. - Misreading a color. In poor light red and orange are easy to confuse. Memorize in good light and always check the two or three colors of each piece.
- Miscounting the letters. If you skip one, parity comes out wrong as well.
- Forgetting the cycle break or writing down a buffer letter. B, M, A, E and R are never memorized.
- Changing the cube's orientation. The letters depend on always holding it with yellow on top and green in front.
- Doing the Ra at the wrong time. It goes between the edges and the corners, never at the end.
- Losing track of where you are. Say the letter again before each setup and do not rush.
Next steps
Once Old Pochmann works for you, there are faster methods that use the same ideas of buffer, letters and cycles. M2 solves the edges with middle layer moves, and R2 does the same for the corners. 3-style places two pieces with each algorithm using commutators, and it is the method the fastest solvers use.
Records
The 3x3 blindfolded single world record is 11.56 seconds, set by Tommy Cherry (United States) on June 28, 2026 at the Mid-Atlantic Quiet Championship in Branchburg, New Jersey. On August 15, 2026, at the Hefei August Open (China), Cherry set a mean of 5 of 13.97 seconds. Those times include memorization. The records for other events are on the records page.
Frequently asked questions
How many algorithms do you need to solve the cube blindfolded with Old Pochmann?
Three: the T perm for edges, the modified Y perm for corners and the Ra perm for parity. Everything else is setups of zero to four moves that you undo after each algorithm.
Do you need a good memory?
Not especially. A normal scramble gives about 12 edge letters and about 8 corner letters. Grouped in pairs and turned into words or syllables, they are easy to remember with a little practice.
What is the buffer?
The position from which pieces are sent to their places. In Old Pochmann it is the top right edge for edges and the top back left corner for corners. Its letters are never memorized.
What do I do if I get back to the buffer while memorizing and there are unsolved pieces left?
Do a cycle break: pick a piece that is not in its place, write down the letter of one of its stickers and follow the path from there until you get back to that same piece.
When do you do parity?
When the number of edge letters is odd. You do an Ra perm after finishing the edges, before starting the corners.
Can I use another lettering scheme or another orientation?
Yes, as long as you are consistent. The setup tables in this guide are for the Speffz scheme with yellow on top and green in front. If you change the orientation, the letters refer to other positions on the cube.
Does memorization time count?
Yes. In competition there is no inspection: the timer starts before the cube is uncovered and stops when you finish solving it blindfolded.
You can already practise in the simulator
Turn the layers with your finger, scramble the cube and try any algorithm step by step.
Keep learning
- Where do I start? Tell us what your cube looks like and we'll take you to the right lesson.
- Notation: what each letter means R, U, F, the prime and the 2, shown turning in 3D.
- Beginner's course Eight short lessons to solve the cube for the first time.
- Intermediate method (2-look OLL and PLL) The bridge between the beginner's method and CFOP.
- CFOP method Cross, F2L, OLL and PLL: the method used by almost every speedcuber.
- Roux method Blocks and the middle slice: fewer moves, a different way of thinking.
- Solving one-handed How to hold the cube and which algorithms change.
- Fewest moves (FMC) Solving the 3x3 in as few moves as possible.
- Solving the 2x2 The pocket cube, in three steps.
- Solving the Pyraminx The pyramid, with its tips and centres in 3D.
- Solving the 4x4 The Revenge cube, by reduction.
- Solving the 5x5 The Professor's Cube, with the same method as the 4x4.
- Solving the Skewb The cube that turns around its corners, with two algorithms.
- Solving the Megaminx The 12-sided dodecahedron, layer by layer.
We sell the cubes these guides were made with: GAN, MoYu and QiYi, from a first cube to competition models, plus lubricants.
See cubes in the shop →