M2 difficult to understand



I am trying to learn M2 for blindsolving the cube with the M2/Old Pochmann method. I am not able to understand it nor to perform by just following the instructions.

I use the Speffz letter scheme A-X and try to do the following:

(A): M2
(B): R’ U R U’ M2 U R’ U’ R
©: U2 M’ U2 M’
(D): L U’ L’ U M2 U’ L U L’
(E): B L’ B’ M2 B L B’
(F): B L2 B’ M2 B L2 B’
(G): B L B’ M2 B L’ B’
( H ): L’ B L B’ M2 B L’ B’ L OR u’ L’ u M2 u’ L u
(I): D M’ U R2 U’ M U R2 U’ D’ M2
(J ): U R U’ M2 U R’ U’
(L): U’ L’ U M2 U’ L U
(M): B’ R B M2 B’ R’ B
(N): R B’ R’ B M2 B’ R B R’ OR u R u’ M2 u R’ u’
(O): B’ R’ B M2 B’ R B
§: B’ R2 B M2 B’ R2 B
(Q): B’ R B U R2 U’ M2 U R2 U’ B’ R’ B
®: U’ L U M2 U’ L’ U
(S): M2 D U R2 U’ M’ U R2 U’ M D’
(T): U R’ U’ M2 U R U’
(V): U R2 U’ M2 U R2 U’
(W): M U2 M U2
(X) : U’ L2 U M2 U’ L2 U

My problem is that I don’t understand the buffer position. It is so easy for the corners with the Old Pochmann method. The buffer there is always A. But for M2 the buffer seems to be U, is that right? Do I have to bring U to A BEFORE I perform the permutations above? Whatever way I try to solve the cube I get it wrong.


Hi, l don’t use speff’s notation so l’ll just refer to pieces as DF (edge on Down and front Face). M2 just sends pieces from the Buffer spot to the Target spot. DF to UB

Let’s say the piece (J) is on the buffer spot, in order to send it to it’s rightful place (FR) you must do some set up moves so that when you execute M2 , you have to replace the piece currently at UB with the piece at FR so that the M2 might permute them. the set up move for this is U R U’ then exchange the pieces with M2 and redo the set up moves U R’ U’.

If it’s still not clear, consider going to “speedsolving.com” you’ll get detailed responses from there.


Zane has a very good tutorial on YouTube. I never used M2 and took the TuRBo to speedoptimal 3cycles route, but I know some of the problems you may be facing. Most people do not notice that you have to know if you are on an odd or even number of targets to properly shoot to the M slice since the M2 restores itself every even target so pay close attention when Zane mentions that.

There are two ways that people take note of the change of the M slice in a solve. They either see which part of the letter pair the target is or they memorize FU as BD during memo because that is where it will actually be during the cycle. Pick one of these two methods and stick with it.

You may also have a Flipped Edge that is in place. I visually take not of that and would set up your buffer which is DF and the flipped piece to the following alg after I have finished my edge cycles. The alg: M’U M’U M’U2 MU MU MU2

If your cycle ends, then shoot to an unsolved piece. This is called breaking into a new cycle.

Memorization and fluidity are the most important aspects of BLD by far so don’t worry about learning a more difficult execution method too soon. Good luck.


How many hours of practice does it take from square one before you were able to do a blind cube in under 5:00?


This will be very different from person to person based upon prior experience. You know how to use memory techniques and that is the big barrier that most people need to learn. It takes time for your fingers to learn cases and most people will learn 2 or 3 algs a day when they are beginning. Try not to use memory techniques to learn algs and just let your fingers get used to it. I was sub 5:00 within my first 15 solves but I didn’t know what I was doing at all at first, especially with the memory part. Here’s a procedure that could help you get your first sub-5 as fast as possible.

  1. Buy a Speed cube online (not a rubik’s brand). I think a Shengshou Aurora,Dayan Guhong or Dayan Zhanchi would be a good choice. This really should be your first step since Rubik’s brand’s are very bad in comparison. It will also be less expensive.
    Time:~2-3 days
  2. Learn the Layer by layer method to solve it while looking. This took me one night to get proficient at when I started. It will take about a week until you are comfortable with it.
    Time:1 day
  3. Give each edge sticker and each corner sticker a letter and practice tracing where one piece would have to go to the next. Really get used to calling that place the letter.
    Time: 2-3 days
  4. Learn M2/OP since they are the easiest methods for execution and they can be fast. These are very straight forward methods.
    Time:2-5 days
  5. Learn about parity, breaking into a new cycle and solving edge flips, corner twists.
    Time:2 days
  6. Make sure that you can come up with an image for every letter and think about how many images per location that you will use.
    Time:2-10 minutes
    7:Do 10 or more sighted solves where you memorize the cube and solve it to see if you are making mistakes with your algorithms. This is important because it is easier to find mistakes with your method this way.
    Time:2 days
  7. you have everything you need to be sub-5. You don’t need to turn fast at all for sub-5.Memo is about 20 letters so that should be easy for you to remember. You want to “think ahead” during the solve and plan the next moves while you are finishing the current one.
  8. Get a sub-5
  9. Think about if and what you want to improve

This will not take very much time per day but It is better to learn a bit at a time instead of all of it in one day. Don’t use mnemonics to learn algs. Think of it like guitar and use your natural muscle memory. You can also overlap the days of step 3-5. Anyway, I think this is something that could be learned in a couple weeks if you already know mnemonics.


Cool, thanks. I can solve a cube already, with a very simple method that is probably the layer method that you’re talking about, but it takes a few minutes and is obviously nothing like what speed cubers do. Maybe I’ll get to work on this in April, but certainly not sooner :slight_smile:


I respect the opinion, that one should learn all these algorythms by muscle memory. It might be the best approach for most people but it drives me nuts. I would never be able to learn it that way because I would simply give up out of boredom. I learned the beginner speed cube 100% with memory techniques. That was fun, slow and perfect for me. Now it is mostly muscle memory but without training explicitly for that.

I am pushing the idea of learning blind solving for a couple of months now without doing it. But I think I will approach it just like the beginnert method and memorize everything. That is much more fun for me. :slight_smile:


Flauwy, have fun learning with mnemonics.

 Cubers have our own way of thinking of algs to learn them and it is more than just muscle memory. We actually have our own Jargon to describe most of the types of moves that you see in Algs and we have to water down our language when talking to people that don't know the language. 
 Most fast algorithms are made as a combination of fingertricks or triggers(a short number of moves that is easy to execute. like RUR', U'L'U, R'FRF',etc These are chunked together in the learning process). Also the short and fast algs have names and we have other attributes that make an alg memorable too. For example, do we remove an F2L pair and reinsert it in a different way?,Is it a Commutator?, <RUD> (generator notation means that it is only using those faces)?, Is it a Mirror, Inverse, or setup to something you already know?   We use ideas like that to break an alg down into something that is very easy to remember while our fingers' muscle memory catches up. An effective memory method for learning algs would take those into concern instead of going a move at a time. 

The reason that I think that memorizing the cases with images is a bad Idea for BLD is because the set of moves should=the image that describes the cycle instead of having to use images to recall the set of moves that another image is supposed to represent. It will be too much interference during recall.


This is how the beginner’s method seems to us.
1.make cross (intuitive)
2.rotate and spam sexy
4.F setup to sexy, f setup to sexy, or both
5.spam sune
6.spam nikolas
7.rotate and spam sexy

actual things to learn:
sexy=RUR’U’ pulls an f2l pair out and back over the slot
sledgehammer= R’FRF’
anti’s are the above backwards
sune= (RUR’) U (RU2R’) a lot like sexy, take a pair out and put it in a different way.
Nikolas= R (U’L’U) R’ (U’LU) a commutator means do part 1 and 2 and then anti 1 and anti 2

I have thought someone how to do the cube in about 10 minutes this way.