MehtaMH
From Speedsolving.com Wiki

MehtaMH is a speedsolving method designed for big cubes. It was proposed by Matthew Hinton as a way of applying the Mehta method to higher order puzzles. MehtaMH solves blocks in pseudo layers in similar fashion to OBLBL, by connecting rows of centre pieces.
Steps
 First 2 Centres (F2C): Solve two opposite centers in the U and D layers (same as Yau F2C).
 First Block (FB): While keeping the solved centers on U and D, build a 1 by n1 by n block like in Mehta but with more pieces.
 Pseudo Block Layers: Solve pseudo (nonmatching) layer(s) of three rows of centre pieces and two unpaired edge pieces on top of the FB, until only one layer and the Rface of centre pieces remain unconnected.
 Last Block Layer: Intuitively connect the remaining pseudo layer and Rface centre rows. This requires some skill and a good knowledge of centre commutators on cubes larger than 4x4x4. On 6x6x6 and up, solve only the corners of the F and Rface centres and then use algorithm to permute the cedges (edgelike pieces of the centres).
 Pair and Orient U/D Edges: Use simple edge pairing to solve two pairs of edge pieces at a time until all the U and Dlayer edges have been connected. Use triggers to orient these edges as you pair them. It may be useful to do a y rotation before this step so you can use the <RUFL> moveset for insertions.
 3x3x3 Stage: Rotate all the layers so that the centres are solved. You have now reduced the cube to the 3x3x3 EOBelt state in Mehta. Use any Mehta path to complete the solve. You may have permutation parity on even order cubes, which can be solved with one algorithm or you can learn the extra PLL/L5EP parity algsets.
Naming
The MehtaMH naming convention has the following forms:
 MehtaMH{N} means the method and algorithms/commutators for a cube of order N (e.g. MehtaMH4 is for the 4x4x4)
 MehtaMH+ is the method for cubes larger than 7x7x7, including 8x8x8 centre comms. This is the most advanced variation.
 MehtaMHX is used to describe the method in general or the knowledge of all the variations.
Pros + Cons
Pros
 Few or no rotations.
 Better lookahead to edge pairing and 3x3x3 stage than Reduction.
 Reduces to the advanced Mehta method
 3 algorithmic steps instead of 2, allowing for higher TPS in 3x3x3 stage.
 Ergonomics: About 40% of the 3x3x3 stage is <RUD> gen. 17% is always <MU> or <RU> gen.
 ABFs get easier on large puzzles.
 Edges in edge pairing stage all have U or Dface stickers for easy recognition
 No oriental (OLL) parity
Cons
 Bad lookahead in pseudo blockbuilding, specifically recognizing the correct 'wing' edge pieces.
 Canâ€™t use freeslice, and twopairatatime edge pairing is inefficient.
 High algorithm count for a big cube method.
 Some 6CP and APDR algorithms have many R2 moves in a row. However, an advanced solver would be able to optionselect to avoid bad cases.
 Adjusting both faces at the end has a higher chance to get a +2 or DNF.
 Transition between algorithmic steps is difficult to master.