I think it's time we take a break from trying to create new 3x3 methods, it's been quite a long time since somebody proposed one that actually went somewhere. We should start working more on methods for different cubes, like 4x4 or big cubes.
I wouldn't necessarily say that methods don't go anywhere. A big problem is the huge time commitment needed to explore, tweak and master any new method, and while many people are willing to quickly 'hash out' a new method, very few are willing to make the huge investment of time to master it. I proposed LMCF 1.5 years ago and I'm still practicing and modifying it in many ways and making great progress but it is all irrelevant until I can register some official results. WaterRoux, ECE/SSC have potential but no one has invested the time to actually get good at them (yet). Same for a number of the other proposed methods.
In my own experiments I have found that [L,l,U,M] sets are WAY faster than the generally preferred [R,r,U,M], at least for me as I use left ring for M moves and a very unusual method of fingering L moves (I hold left hand below the cube and use right index on L, just the same way as you hold left ring for M). This style of fingertricking makes [L,l,U,M] sets very fast, although in theory you can reverse the finger trick style and use the same style for [R,r,U,M] if you can do M moves with right ring finger. Right now I am mostly working on improving lookahead from corners to E2L transition and also extending the transition to solve more U layer edges, forcing the remaining E2L into [L,l,U,M] drastically improving the ergonomics of the E2L phase while maintaining the same low movecount from before.
After watching Feliks' walkthrough of his 4.73 where he saw the cross and 3 pairs (4+2+2+2 = 10 pieces) in the inspection, I am wondering which method allows you to see the most moves in the inspection? My personal best was 10 pieces when I saw a CLL skip (8) + 1st E2L pair (2) in the inspection. We theorized in WaterRoux that on rare occasions you could see the whole first block (5) + the remaining 6 corners [11 total] in the inspection.
Can anyone share (A) largest number of pieces ever seen in the inspection, and (B) method that sees the most in the inspection in general? LMCF allows a 2x2 start so you normally see 8 pieces in the inspection; most expert CFOP solvers see the cross and 1 pair [6] and rarely the XCross and 1 pair [8], and I suppose extremely rarely, 10 pieces like Feliks saw. I've heard that Alex Lau has seen both Roux blocks in the inspection (true or false?) that would be 10 pieces as well.
Roux with Tyrannical Caterpillar could be pretty insane and even Roux with just CMLL/standard LSE is really good. But I do believe that LLOB is still a method with serious potential. But if I'm wrong, at least I had some fun with it and inspired some people to help progress cubing methods.
