How Fast Can Intuitive F2L Get?

[irontwig]

Alg based f2l is inflexible, while intuitive can involve stuff like petrus, heise, multislotting, xcross, etc.

 

[trying-to-speedcube...]

It surprises me that the following hasn't been mentioned yet:

Alg-based F2L is based on the fact that the other three slots ARE SOLVED. Now grab your cube, do R U R2 U R2 U2 R'

An algorithm would say something like this: y U' R U2 R' U y' R' U' R

Now suppose that the FR slot is not yet solved. Here comes the advantage of intuition:

Alg-based people will do: y U' R U2 R' U y' R' U' R, nothing else than they have learnt.

Intuitive solvers will simply do: R U2 R2 U' R, which is simply faster (this can of course be further expanded by doing R U' R' after it, even when the FR slot is solved, but for some reason this doesn't get listed by algorithm lists)

 

[Lord Voldemort]

For the above example, R U2 R2 U' R'2 U' R' is the first thing that came to my mind... I suspect that that is what the algorithms would say too. Maybe a better example? (I do use intuitive F2L, by the way)

EDIT: I didn't see the read the rest of your post, sorry. But I find it hard to believe that R U2 R2 U' R'2 U' R' isn't on alg lists. It seems a bit obvious...

I do advocate for intuitive F2L though. It becomes like an algorithm over time, and you can apply them to ANY slot without having to learn a new set of 41 algorithms.

 

[qqwref]

Algorithmic F2L isn't about learning a "set" of 41 algorithms and then applying them blindly. It's obvious that people who do that aren't going to be fast, just like people who try to do the cross by solving one piece before finding the next one. For me algorithmic F2L is about learning an efficient way to deal with each case (or perhaps several depending on whether you have other F2L pairs open) immediately. The important thing is to understand what the algorithm is doing, so that you can do it from any angle and on any color (and even on megaminx). But for the most part I did not spend time figuring out my F2Ls myself, instead I started out by learning an efficient way to solve each case (and watching the pieces to figure out WHY it worked), which I think made me improve with F2L a lot faster at the start. Instead of figuring out a move sequence as I went, I was given one and then figured out why it worked. Algorithmic F2L is only bad if you learn the sequences blindly and can't apply them from other angles, just like intuitive F2L is only bad if you don't put in the effort to find efficient ways of doing each case.

The end result is the same, though. Someone who learns intuitive F2L properly (so they have efficient move sequences) and someone who learns alg F2L properly (so they really understand what their sequences are doing) will end up the same, both should be just as efficient and just as fast at recalling what to do for each case.

 

[JohnnyA]

Intuitive F2L ends up the same thing as alg F2L. Intuitive F2L users need to develop the lightning-quick response and look ahead ability that alg F2L gives, and alg F2L users need to develop the flexibility that Intuitive F2L gives. I would personally use algorithms to work out concepts of the F2L and do a mix of intuitive and alg-using. Nobody really memorizes algs for the base cases anyway. For multislotting and special cases, alg users have the upper hand because of the fast recognition that they can achieve.

EDIT: the above post is exactly what I'm trying to express, except that I think that both reveal the same results.

 

[abr71310]

I haven't learned any real F2L algs, and I average around high 20s, low 30s sometimes, with the occasional sub-25 and some 20.xy solves (YET TO BREAK THE SUB-20!!!)... I use intuitive F2L, 2-look OLL (some cases for 1-look learned) and full PLL...

So I'd say for Intuitive F2L + cross, about 13 - 17 seconds, depending on how fast I recognize pairs and lookahead.