I searched for a few days looking for the best speedcubing methods and so far I like Petrus and Roux because they seem to have little memorization but are still fast.

Which one is faster? Which one has less algorithms to learn (if you use Jelinek's algorithms for the last layer on Petrus). Which one is the best to learn?

I'm currently using a corners first method but it has a lot of F turns and I don't know how to do those and B turns fast. It's also not very fast because you place one edge at a time.

If I go with Petrus I think I should use those algorithms Josef Jelinek made for the last layer. If I use Roux I will still be using middle layer turns. I can do 3 middle layer turns a second if that helps.

P.S. Don't tell me to go with Fridrick. I'm not learning that. Maybe later. I'm also not going to do Waterman.

check speedcubing.com

Roux has a sub 14 average with his own system

Alright, I checked the site and Lars posted his average and it was 21 seconds and the two people that posted their record using Roux were about 25 seconds. Most of the people using Petrus were in the 30s. I guess I'll go with the Roux method. I'll learn it pretty fast too I believe since I'm already using a corners first method. Too bad I can't use my current method for getting top corners...

could somebody pleas explain how to solve the edges and centers using roux's method. I cannot understand the moves on his site and they dont seem to work. If someone could explain them or give me alternatives i would greatly appreciate it.

email Gilles himself. He is very nice and I'm sure could direct you to some answers, if not answer them himself.

I don't know what you have against Friedrich (CFOP). It has a proven record of being super fast with very little thinking. I was able to get under 35 sec average with a 4 look LL only using 6 algorithms.

I'm not trying to bash any other methods, I know theres lots of ways to get a cube solved quickly, it's just that there's so much more information on Friedrich method all over the net.

I would recommend Roux if you can do slice moves quickly, or Petrus if you have 4 sets of eyes.

By the way, Lars is really nice also as long as you don't mention "the tree of Death"

good luck!

Gosh, can't anyone spell Fridrich correctly?

What I don't like about the Fridrich method is all of the algorithms you have to learn. I'm not going for record times yet (I might eventually get the urge to though) so I decided that the Roux method would be good to learn because it looks like it is just about as good as Fridrich (and doesn't have many algorithms) and might actually be as good if more people would try it.

I believe I can do slice moves fast. I started practicing that after I posted this topic and I timed myself using my microwave (I have nothing else to use ) and I was doing 5 a second. My cube isn't very smooth though (25th anniversary crap) so I don't know how fast I really could go.

About solving the edges in Roux's method, doing just one of those algorithms doesn't solve all of them at once. You have to find the pattern you have that matches one of the pictures he has and do the algorithm for it. Then, if they aren't all permuted, find the pattern you have and do the algorithm. Then once they are fixed, continue to step 4b. If that doesn't work for you, just email Roux like they said.

What is the "tree of death"?

Maybe you're taking into account the F2L algs. These can totally be done intuitively if you just look at the pieces. Anyway I only use..

2 edge orient algs
7 orient corners
21 PLL

it isn't that many algs to be sub 30.

My friend Clancy was doing some Roux solves and it was really neat to see. He explained it to me but I haven't messed with it very much. See if you can find a video of Bob Burton doing a Z or H Permute, he does some blazing fast M's.

The 'Tree of Death' is an evil tree in Lars backyard, I recommend nobody ever try to climb it.

good luck finding your method, if you mess with a few different ways you may end hybriding (is that a word?) something.

-mike

Yeah, I was adding in the corner edge pairs algorithms and it does make it seem like a lot. I was messing with corner edge pairs yesterday and I found that I didn't need the algorithm sheet I printed out to do it. It was actually pretty easy.

About "hybriding", I've been trying to figure out how to get the two 3x2x1 blocks very fast but the only thing I've came up with is to either do two opposite edges on the bottom and then do all four corner edge pairs or to make a 3x2x1 block then do the other one like in the Petrus method. I've also been trying to find a way to use my old way of getting top corners and it still be as fast as Roux.

I'm just really lost on what I should do.

EDIT: I just watched some of Bob Burton's videos but I can't really tell what is going on when I watch speedcubing videos.