The M2 method is the next step for most blindfolded solvers after Old Pochmann. You still send one edge home at a time, just like before, but the swap algorithm is no longer a 14-move T-perm: it is a single M2 move. This guide covers the whole edge part of M2: the buffer, how the swap works, a table of setup moves for every letter, the four special letters, parity and one complete example solve. Every letter has its own picture.

This guide is for people who can already solve blindfolded with the Old Pochmann method and know the Speffz letters well. Corners are still done with Old Pochmann, so the edges are the only part you need to redo. Every algorithm here has been run on the EasyRubic cube simulator, including the even-letter rule and parity, which were checked over 300 random scrambles.

What is M2?

M2 was put forward by Stefan Pochmann, the same person who came up with Old Pochmann. According to Cubing History, he published the full method on 24 February 2006 in a discussion group about blindfolded solving, then put M2 and the R2 corner method on his website on 3 June 2007 (Cubing History).

J Perm describes M2 as "conceptually the same as Old Pochmann": there is still a buffer, a target spot and a swap algorithm, only the swap algorithm is much shorter (J Perm). He also notes that many intermediate solvers use M2 for edges and Old Pochmann for corners. That is the approach this guide follows too.

The SpeedSolving Wiki mentions something worth knowing up front: M2 has a few exceptions, so it is not recommended as your very first blindfolded method (SpeedSolving Wiki). If you have never finished a single blindfolded solve, start with Old Pochmann.

Why is M2 faster?

In Old Pochmann, every edge letter is a setup move, a 14-move T-perm, then the setup move undone. In M2, most letters are just three setup moves, one M2 move and three moves to undo the setup. Counting in the table below, most letters need 7 moves, and even the longest one needs only 11.

Fewer moves means less time turning, and fewer chances to slip up when you can't see the cube.

The buffer, the target spot and the M move

Hold the cube with white on top and green in front, just like every picture in this guide. Each picture shows two cubes: the left one is before the algorithm, the right one is after. Only the pieces that matter are coloured.

  • M is the middle slice between the left and right faces, and it turns the same way as L. M2 means turning that slice half a turn. For a refresher, see Rubik's Cube notation for beginners.
  • The buffer is the bottom front edge (yellow-green). In Speffz, its yellow sticker is letter U and its green sticker is letter K.
  • The target spot is the top back edge (white-blue). Its white sticker is letter A.

Both of these edges sit in the M slice, so a single M2 swaps them (J Perm). You read the edge sequence exactly as in Old Pochmann, except that you start from sticker U instead of sticker B: look at the colour on sticker U, write down the letter of that colour's home sticker, then look at the sticker you just wrote down.

One swap: letter A

Letter A is already at the target spot, so all you need is M2:

  1. You start with (input)
  2. After the formula

Look closely at the before picture: besides the buffer and the A edge, the top front and bottom back edges have also swapped, and the four centres on the top, bottom, front and back are out of place too. That is the "extra" work M2 does. You don't need to fix it after each letter: the next M2 puts them back. This is why there are four special letters and an even-letter rule later on.

For every other letter: do a setup move to bring that sticker to spot A, do M2, then undo the setup move. No setup move touches the buffer, and no edge algorithm in this guide touches the corners.

Setup moves for every letter

There is no setup for U and K (they are the buffer itself). C, I, W and S sit in the M slice, so they have their own algorithms, covered further down. Each picture below shows all three parts together: setup, M2, undo.

Top right and top left edges

These are the two letters on the top face that are not in the M slice.

  • Letter B: R U R' U'
  1. You start with (input)
  2. After the formula
  • Letter D: L' U' L U
  1. You start with (input)
  2. After the formula

Middle layer edges

The four middle layer edges have eight stickers. The stickers on the front and back faces (J, L, T, R) use the pattern U, one R or L turn, then U the other way. The stickers on the left and right faces (F, P, H, N) use B.

  • Letter J: U R U'
  1. You start with (input)
  2. After the formula
  • Letter L: U' L' U
  1. You start with (input)
  2. After the formula
  • Letter T: U R' U'
  1. You start with (input)
  2. After the formula
  • Letter R: U' L U
  1. You start with (input)
  2. After the formula
  • Letter F: B L2 B'
  1. You start with (input)
  2. After the formula
  • Letter P: B' R2 B
  1. You start with (input)
  2. After the formula
  • Letter H: L' B L B'
  1. You start with (input)
  2. After the formula
  • Letter N: R B' R' B
  1. You start with (input)
  2. After the formula

Memory tip: this table works in left-right mirror pairs. Take the setup for J, change R to L and reverse the direction of every move, and you get the setup for L. The same goes for T and R, F and P, H and N.

Side stickers of the top and bottom edges

These stickers sit on the left or right face, on an edge that belongs to the top or bottom face.

  • Letter E: B L' B'
  1. You start with (input)
  2. After the formula
  • Letter M: B' R B
  1. You start with (input)
  2. After the formula
  • Letter G: B L B'
  1. You start with (input)
  2. After the formula
  • Letter O: B' R' B
  1. You start with (input)
  2. After the formula

These four letters use the same pattern: B, one L or R turn, then B the other way. E and G differ only in the direction of L; M and O differ only in the direction of R. M is the mirror of E, and O is the mirror of G.

Yellow stickers on the bottom face

  • Letter V: U R2 U'
  1. You start with (input)
  2. After the formula
  • Letter X: U' L2 U
  1. You start with (input)
  2. After the formula

X is the mirror of V. The same goes for the two letters on the top face: the setup for D is the mirror of the setup for B.

Letter Q: the blue sticker of the target spot itself

Q is the other sticker of the A edge. That edge is already in the right spot but flipped, so the setup is longer. From the SpeedSolving Wiki:

  • Letter Q: U B' R U' B
  1. You start with (input)
  2. After the formula

The four special letters: C, I, W, S

C and I are the two stickers of the top front edge; W and S are the two stickers of the bottom back edge. Both of these edges sit in the M slice, so any setup that brought them to spot A would break up the M slice. J Perm notes that M2 has exactly four letters that are special cases like this. The four algorithms below come from the SpeedSolving Wiki.

  • Letter C: U2 M' U2 M'
  1. You start with (input)
  2. After the formula
  • Letter W: M U2 M U2
  1. You start with (input)
  2. After the formula
  • Letter I: D M' U R2 U' M U R2 U' D' M2
  1. You start with (input)
  2. After the formula
  • Letter S: M2 D U R2 U' M' U R2 U' M D'
  1. You start with (input)
  2. After the formula

The even-letter rule

After every M2, the top front and bottom back edges trade places. So after an odd number of M2s, the piece sitting at top front is really the bottom back piece, and the other way round. The rule:

  1. Count each letter's position in the edge sequence: 1st, 2nd, 3rd...
  2. A letter in an odd position (1, 3, 5...): do that letter's own algorithm.
  3. A letter in an even position (2, 4, 6...): if it is C, do the algorithm for W; if it is W, do the one for C; I gets the one for S, and S gets the one for I.

If you memorise letters in pairs, as in Rubik's Cube blindfolded memorisation, the second letter of each pair is the one in an even position. Every other letter (anything that isn't C, I, W or S) is done as usual, whatever its position.

Example: an edge sequence with only two letters, J and C. J is in position 1, so it is done as usual. C is in position 2, so you do the algorithm for W. The picture shows both steps together:

  1. You start with (input)
  2. After the formula

Two letters, two M2s (the algorithm for W counts as one), so afterwards the centres and the M slice edges are all back in place.

Parity: when the edge sequence has an odd number of letters

If the edge sequence has an odd number of letters, then after the edges the centres and the two edges at top front and bottom back are still out of place. The corner sequence will also have an odd number of letters, and the Old Pochmann corners will swap the top back edge with the top left edge one extra time.

J Perm only says that the M2 parity algorithm is different from the Old Pochmann one, without giving it. Here is the approach we checked on the simulator: finish the edges, then shoot three more letters, A, D, A, before the corners. Put together, that makes one algorithm:

M2 L' U' L U M2 U' L' U L M2

  1. You start with (input)
  2. After the formula

This algorithm puts the centres and the two M slice edges back in place, and at the same time swaps the top back edge with the top left edge. The corners after that, with an odd number of letters, swap those two edges back exactly once, so in the end everything is solved. The order is: edges, parity, corners.

Corners: still Old Pochmann

None of the edge algorithms in this guide touch any corner, so the corner sequence you memorised at the start is still correct after the edges. Do the corners just as in the Old Pochmann method guide: buffer A, swap spot P, the shortened Y-perm and the corner setup table from there.

There is just one change: edges first, corners last. The shortened Y-perm swaps the top back edge with the top left edge, and the top back edge is the M2 target spot. Doing the corners last means that extra swap either cancels itself out (even sequence) or has already been taken care of by the parity algorithm (odd sequence).

A full solve, start to finish

The cube in the picture below was scrambled so that both sequences are short. Reading from the buffer:

  • Edges: sticker U has the green colour of the green-red edge, write J. Sticker J has the white colour of the white-green edge, write C. Sticker C has the yellow colour of the yellow-orange edge, write X. Sticker X has the yellow colour of the buffer itself, so stop. Edge sequence: J, C, X.
  • Corners: sticker A has the green colour of the yellow-orange-green corner, write L. Sticker L has the white colour of the corner buffer itself, so stop. Corner sequence: L.

Doing the solve:

  1. J in position 1: the algorithm for J.
  2. C in position 2 (even): the algorithm for W.
  3. X in position 3: the algorithm for X.
  4. The edge sequence has 3 letters, an odd number: do the parity algorithm.
  5. Corner L: setup D, shortened Y-perm, D'.

The picture shows that whole solve, from the scrambled cube to the solved one. Between steps 2 and 3, the U2 and U' that come right after each other are merged into one U:

  1. You start with (input)
  2. After the formula

M2 practice plan

  1. Know M2 and M by heart. Practise M2 with your ring finger or middle finger until it feels smooth, because you will do it dozens of times in every solve.
  2. Go through the table, eyes open. Take a solved cube, do one letter and compare it with the picture. Do that same letter once more, and the cube should be back to how it was.
  3. Get the hang of the four special letters, and ask yourself "is this letter in an even or odd position?" every time you come across C, I, W or S.
  4. Solve the edges, eyes open. Scramble the cube, read the edge sequence from sticker U, and do it. When the sequence is even, every edge and centre should be in place after the edges.
  5. Add the Old Pochmann corners and parity, still with your eyes open.
  6. Put on the blindfold once you can solve with your eyes open without looking back even once.

The most common mistake for people new to M2 is forgetting the even-letter rule. When a solve goes wrong, check the letters C, I, W and S in your sequence first. Once you can solve it reliably, time yourself on the 3x3 timer to track your progress.

Video

The video below, from the Daniel Kidd channel on YouTube, is a blindfolded solve using exactly the combination in this guide: M2 for edges, Old Pochmann for corners.

41.55 3x3 Blindfolded Solve! (M2/OP, PB2)Video · Daniel Kidd · YouTube

Sources

Frequently asked questions

What is the M2 method?

It is a method for solving the edges of a Rubik's Cube blindfolded, put forward by Stefan Pochmann in 2006. The buffer is the bottom front edge, the target spot is the top back edge, and each swap is just one M2 move plus a setup.

Should blindfolded beginners start with M2 straight away?

No. The SpeedSolving Wiki says M2 has exceptions and is harder for beginners. Get to the point where you can solve with Old Pochmann first, then switch the edges over to M2.

What is the difference between M2 and Old Pochmann?

The idea is the same: one buffer, one target spot, one piece sent home at a time. Old Pochmann swaps with a 14-move T-perm, while M2 swaps with a single M2 move, so each letter is much shorter. In return, M2 has four special letters and the even-letter rule.

When does M2 have parity?

When the edge sequence has an odd number of letters. Finish the edges, do the algorithm M2 L' U' L U M2 U' L' U L M2, and only then do the corners.

What do I use for the corners?

This guide uses Old Pochmann for corners, which J Perm says is what many intermediate solvers choose. Stefan Pochmann also has his own corner method called R2, in the same style as M2.

Does M2 work on a 4x4?

Yes, as a variation. According to Cubing History, Erik Akkersdijk applied the idea to 4x4 edges in 2007, and the SpeedSolving Wiki says the version for 4x4 edges uses r2 instead of M2.

EasyRubic Editorial

Friendly notes for anyone who wants to solve their first Rubik’s Cube.

Ready for your first turn?

Go from reading notation to each step of a solved cube.

Start with step 0

Learn the 3x3 step by step · 3x3 online timer