How does the M2 work?

[Lotsofsloths]

I read the page and saw the example solve and stuff, but how does it work?
no algorithms, only setup moves then preforming M2, its quite amazing actually...
Also, are the setup moves intuitive?
It could be a possible method for me

 

[Mike Hughey]

The setup moves are actually very intuitive - I've never bothered to memorize them because they're almost visually obvious, and I can execute them faster than almost anything else I've tried, even though I rarely ever do M2. The only problem with M2 is dealing with the difficult edges UF and DB; you may wind up with misoriented edges that you have to go back and fix (but there are optimizations you can do to help with this, which are definitely more advanced). I still think M2 is a little more difficult than some of the other methods for beginners. It's a good advanced method, but if I were you I'd probably stick with pjk's method (I think you said you were trying that) until you get it down first. Still, it couldn't hurt to at least try to understand how it works.

As for how it works, it's a 3-cycle between the UB and DF pieces and the piece you're trying to replace. So you cycle your DF piece to the desired location, that piece goes to UB, and UB goes to DF. But since you've done M2, that means the piece at UB is now at DF, so you can just use it again to shoot to your next location. If you're going to do M2, it's very helpful to understand that this 3-cycle is what's actually going on.

 

[Lotsofsloths]

As for the piece by piece method that uses J perm and T perm, that never seems to work for me!
I'm not sure if its my memo or execution, I get corners with ease, I preform parity when necessary, but my edges suck, they are always in the wrong spot except a few!(Today I got 6 edges incorrect)


Wow, something just clicked
(Not about M2 method)
Instead of 3 cycling 3 edges with the Permutation, use this algorithm they use with Roux method, only 4 moves and cycles 3 edges,
U2 M U2 M'
and It's inverse

Just thought of it so its a rough idea..
,

 

[joey]

(Not about M2 method)
Instead of 3 cycling 3 edges with the Permutation, use this algorithm they use with Roux method, only 4 moves and cycles 3 edges,
U2 M U2 M'
and It's inverse

Just thought of it so its a rough idea..
,

Umm, you basically use that in M2. except its U2 M' U2 M'.

 

[cuBerBruce]

Wow, something just clicked
(Not about M2 method)
Instead of 3 cycling 3 edges with the Permutation, use this algorithm they use with Roux method, only 4 moves and cycles 3 edges,
U2 M U2 M'
and It's inverse

Just thought of it so its a rough idea..

Daniel Beyer mentioned this idea back in 2006:
http://games.groups.yahoo.com/group/blindfoldsolving-rubiks-cube/message/1568

I have been using essentially this idea for solving edges. See:
http://games.groups.yahoo.com/group/blindfoldsolving-rubiks-cube/message/1641

Also see my recent "FMC BLD" solution for an example solve with this method.

 

[alexc]

The idea of M2 is to solve one piece at a time by shooting from a buffer. The buffer is either DF or FD, whichever you prefer. Stefan Pochmann, who I'm sure you know invented the method, use DF. Erik uses FD. I use FD. It is just your personal preference.

So, say that FD goes to RU. The idea is to bring RU to BU (Only if your buffer is FD. If it is DF bring the sticker to UB.) with setup moves that don't disturb the other M slice edges, perform M2, undo the setup moves. This results in one solved edge. So, to do this I would use R' U R U' as the setup moves. So, the whole alg would be R' U R U' M2 U R' U' R. I do all of the setups intuitively except the M slice ones.

I think you already know it is impossible to switch two pieces and not change anything else on the cube. Therefore the method has some side effect. It is a rotation of the M slice centers and switches the UF/FU and DB/BD edges. This only affects how you solve those two edges. If the centers are wrong, you need to shoot to the opposite place, so UF becomes DB and FU becomes BD and vice versa.

So, to answer some of your questions, yes, the setup moves are all intuitive for me, except for the M slice I memorized algs. I am very glad I switched to this method, it is very easy and extremely fast!!!

 

[Lotsofsloths]

I still don't get how you would fix the centers and the UF/FU and DB/BD edges.
I basically need an easy effective method, that I can sub 10min BLD solve it.

 

[Dene]

I use Pochmann's Y+T perms method, extremely easy. I got a new Personal Best (of maybe, 5 solves?) of 8:29 yesterday with it.

 

[Lotsofsloths]

Dene, you can BLD?
You never told me that!
I currently use the one where you use the J perms and the T perm.
How would you use the Y Perm and the T perm?
They don't switch the same corners?

 

[Dene]

Oh, sorry, I've only been learning for just over a week lol.
I used Joel's page to learn BLD cubing. Basically, you use Y perm without the starting F and finishing F', for corners. T perm for edges. Go through it, it is EXTREMELY easy. I did my first BLD solve after only 6 days after reading that page.

 

[malcolm]

So i'm not the only one from NZ who can do blind =) But im faster =P

Another trick you can do to avoid shooting to the bad places is to change your buffer, since M2 is just 3 cycles really. I use this for 4x4x4 BLD with R2, which i intend to try soon, i've just been doing sighted solves, but i suck at the centers.

 

[Lotsofsloths]

Oh, sorry, I've only been learning for just over a week lol.
I used Joel's page to learn BLD cubing. Basically, you use Y perm without the starting F and finishing F', for corners. T perm for edges. Go through it, it is EXTREMELY easy. I did my first BLD solve after only 6 days after reading that page.

Hm...I'll check it out, my corners are pretty good though, just my edges suck, both memo and execution

 

[Dene]

So i'm not the only one from NZ who can do blind =) But im faster =P

Another trick you can do to avoid shooting to the bad places is to change your buffer, since M2 is just 3 cycles really. I use this for 4x4x4 BLD with R2, which i intend to try soon, i've just been doing sighted solves, but i suck at the centers.

How fast? I intend to be sub-5 by the end of this week, and sub-3 in a month.

 

[malcolm]

2:21 is my PB =D Mostly im sub-3. I'm in the process of switching from Macky edges to freestyle, and commutators for corners, so i should be sub-2 soon. For getting from 8 mins down to sub 3, just work on your memo, you can cut minutes off easily. I memo in 45-1:15 usually.

 

[alexc]

I was extremely slow (like 6-10 minutes) with pure and free style 3 cycle. Then, I switched to M2, and BAM my times dropped to consistent 3, sub 3 on a good one for me now, 4 on a bad one. Hope to be at about 2 by Princeton.

 

[amateurguy]

I basically need an easy effective method, that I can sub 10min BLD solve it.

Some beginner's methods have great potential to be fast.

I use PJK's method that only uses T-perm for CP and EP (I learnt orientation from random vids on YouTube). My personal best is under 5 and a half minutes. An average of sub-6 is very possible even though the method might seem inefficient.

With enough practice the execution phase can be done in under 3 minutes. Seriously. And the cool part is with this method, I sometimes don't pause to think between steps because I know my next setup and my next move (sort of like lookahead in F2L during normal solves).

But yeah, we're in the same boat. I still don't quite get M2. Plus I'm not ready to memorize more stuff (22 pieces versus 12 on T-perm) without a good memory method to start with.

 

[joey]

Some beginner's methods have great potential to be fast.

Depends on what you call fast.

With enough practice the execution phase can be done in under 3 minutes.

I executed in 1:42 on my first try.

Plus I'm not ready to memorize more stuff (22 pieces versus 12 on T-perm) without a good memory method to start with.

Where is this 22 pieces thing coming from? You memorise the same wether you use M2 or T-perm.

 

[alexc]

Where is this 22 pieces thing coming from? You memorise the same wether you use M2 or T-perm.

I think he does orientation with the T perm, making only 12 places to shoot to. Really though, the real strength of the Pochmann methods is 1, no orientation, and 2, easy setups and execution.

 

[joey]

Where is this 22 pieces thing coming from? You memorise the same wether you use M2 or T-perm.

I think he does orientation with the T perm, making only 12 places to shoot to. Really though, the real strength of the Pochmann methods is 1, no orientation, and 2, easy setups and execution.

Who was it that invented M2 again, oh right, it was Mr. Pochmann himself. Therefore M2 still only needs 12 edges.

 

[amateurguy]

Where is this 22 pieces thing coming from? You memorise the same wether you use M2 or T-perm.

I think he does orientation with the T perm, making only 12 places to shoot to. Really though, the real strength of the Pochmann methods is 1, no orientation, and 2, easy setups and execution.

Who was it that invented M2 again, oh right, it was Mr. Pochmann himself. Therefore M2 still only needs 12 edges.

Yes, I do orientation separately from T-perm, which is why my execution time is considerably slower.

I should've rephrased my sentence correctly. In the method I use now, I only have 12 (or 10) different edge positions and setup moves to know. M2 and other orientation-permutation methods require me to know 22 different positions instead (an edge has 2 orientations, ignoring buffers).

Sure M2 needs you to only memorize 12 edges but you still have to remember how to shoot to 22 different locations. And I mainly avoided T-perm with orientation because I have trouble setting up with the edge's orientation in mind.

My apologies if I've used the wrong terms/jargon etc.