Blindfolded solving of 4x4

Welcome to the first real site with directions on how to solve a 4x4. I'm in the process of learning it, and have already solved all edges and corners blindfolded. I know how to do the centres but those take some practise as I will explain later on.
First lets compare a 4x4 and see the differences to a 3x3.
There are 3 groups: centres, edges, corners.

Centres is quite different since we don't have to solve these really on most 3x3 BLD methods. It is however handy to seek a position of the cube (with cube rotations) in the first few seconds, where there are the most pieces solved already. Another thing which is quite important is that most normal 3x3 algorithms like the y permutation used to solve corners is actually twisting some centres. To make things easier right away I decide to solve the centres first so I get rid of other nasty things. I use commutators to solve 2 centres at once
Also, like you may know it doesn't matter where a centre is on the bigger cubes (besides supercubes, but that's out of the picture for now), but as long as it is in its home face it is solved. We will need that later on too.
The edges are quite different from a 3x3, you know there are 2 of them but surprisingly they don't have any orientation, just permutation. To see what I mean take your 4x4 take a single edge out and try to put it in the other way around. It's simply not possible. I use the same principle as the M2 method for 4x4, but as find out, it has some differences.
Corners are totally the same as on a 3x3. There's nothing much to say. I solve them after the centres.

Edges

For edges I use the M2 method, now the r2 method. It's advisable to know how to use M2 on a 3x3 before trying this on a 4x4. I shoot from the rFD to the rBU position. It's interesting to see that solving the edges on 4x4 with this is mostly the same as on a 3x3 but it only takes a bit longer. You only have to remember the cubie that you will solve since there is only one way to bring that to the rBU position. Another way to remember it as the same way as on a 3x3. This is a bit awkward in the beginning but the RU and UR positions are in this case 2 different cubies. This might sound wrong to you but what I mean is that there is only one cubie that can have the U color of those 2 cubies on the back side of the cube when it's moved to rBU. So lets say we have yellow on top and orange on front, then with UR I mean the back cubie of the 2 UR/RU's, cause that one has yellow on top and on original M2 method, if I shoot to UR I position the yellow side of the yellow/blue cubie on BU. Because of this new type of identifying a cubie you can use the same algorithms as on M2 method. So to shoot to bUR, I use the same algorithm as for UR and therefore I only have to memorize UR. To shoot to fRU I use the algorithm to shoot to RU. I hope you can understand this as this is pretty hard to explain for me.
So this was one difference compared to M2 method. Another difference is the algorithms for the r/l slice edges. We see that the old algorithm: MU2MU2 never would work on 4x4 as: r'U2r'U2 (with r I mean the single slice!). So that was a bit puzzling. I've made 2 categories for the new algorithms. The first one is for the l edges (remember there are just 4 cases for that (also notice the new target of DF (lDF) )) has the principle of: 1. bring to UB, 2. flip it so it will be in BU, 3. r2, 4. flip it back, 5. undo the bring to UB move.
Example: shoot to DF. First step: l2, second step: U B' R U' (flips the cubie and brings it to BU), 3rd step: r2, 4th step UR'BU' (undo the flip), 5th step: l2. So that is: l2 U B' R U' B r2 B'UR'BU' l2. To shoot to UB you don't have step 1 and 5 of course. There are probably better ways of doing this, but this works fine. Now for the 3 r-slice edges. To shoot to UF on M2 method I use: F E RUR' E' RU'R' F' M2. To shoot to UF (you know it's rUF now I hope) use the same principle: F d RUR' d' RU'R' F' r2 (thanks to Joel van Noort for this). To shoot to DB I use the inversion of this: r2 F R U R' d R U' R' d' F'. To shoot to BU we can merely use r2 of course. Note: when you have to shoot to for instance UF and the centres are wrong you have to shoot to DB instead, BUT because the l slice edge don't switch any more after the r2 move you don't have to shoot to FU instead of BD any more(!).
So here are the algorithms. Again all algo's besides the r/l slice ones are build up: 1. Bring to BU, 2. r2, 3. undo setup moves.
Target Algorithm
LU LU'L'U r2 U'LUL'
LF U'L'U r2 U'LU
LD U'L2U r2 U'L2U
LB U'LU r2 U'L'U
UL x' UL'U' r2 UL'U' x
FL x' UL2U' r2 UL2U' x
DL x' ULU' r2 UL'U' x
BL x' L'U'LU r2 U'L'UL x
RU R'URU' r2 UR'U'R
RF URU' r2 UR'U'
RD UR2U' r2 UR2U'
RB UR'U' r2 URU'
UR x' U'RU r2 U'R'U x
FR x' U'R2U r2 U'R2U x
DR x' U'R'U r2 U'R'U x
BR x' RU'R'U r2 U'RUR'
r/l slice edges
BD l, U B' R U' B, r2 ,B'UR'BU', l'
UF F d RUR' d' RU'R' F' r2
DB r2 F R U R' d R U' R' d' F'
FU l', U B' R U' B, r2 ,B' U R' B U', l
UB U B'RU'B, r2, B'UR'BU'
BU r2
DF l2 U B' R U' B r2 B'UR'BU' l2

So that's about everything for the edges besides the parity. If you have an odd number of edges your centres are still wrong. To fix that I use: r2 D' L' F ( l' U2 l' U2 F2 l' F2 r U2 r' U2 l2) F' L D, r2 of course, then I fix it with a parity algorithm for K4 method (first some setup moves), and then undo the setup moves.

Corners

I solve the corners the same as on the a 3x3. This is done with 2 cycles with the algorithm: R U' R' U' R U R' F' R U R' U' R' F R which shoots from buffer LBU to DFR having the side effect of switching edge UB and UL back and forth. You basically setup the corner to solved on DRF, do the algorithm and undo the setup moves. For more, see the M2 method for 3x3.

Centres

Currently I'm still practising the centres. Centers are the hardest thing for big cubes I think. I stated on the M2 method for 3x3 that you can solve 2 edges at a time by using a 3 cycle instead of the normal things. This is exactly how I (try to) solve the centres. A face is build up like:
1 2
3 4

The names for the centres:
U = AQBC
Q is the buffer (the place we shoot from)
F = DEFG
R = HIJK
B = LMNO
L = PRST
D = UVWX

So the U centre which would move with the r and b slice (I'm bad at the notation of centres so I use the letters from now on) is the buffer. I use yellow on top and orange on front, if the buffer is yellow it doesn't have to mean it is solved. Like I said before we always first solve place number 1, then 2, then 3, then 4. For the U face this is a bit different because of the buffer, there the order is ABCQ. So the if spot A is not solved yet and the buffer is a yellow one, we first shoot to the A position and target 2 is where the centre of A should go.
So each time we shoot like: Q->target1->target2. This is done with 3 cycles of centres build up by commutators. An example: Q->C->K: U r'd'r U' r' d r.
I'll continue on this centre tutorial later on when I'm better at thinking up these commutators. For now have a look at: This. Commutators for big cubes explained by Daniel Beyer. Also a big thanks to him for helping me with bigger cubes BLD.