Which method has the fewest algorithms or is the easiest to remember?

[asportking]

you do f2b then cmll or 2 look cmll (2 Algs) finish f2l and finally ether ell or ell and epll.

1. I'm pretty sure that requires more algs than normal Roux (or at least the "variation" PandaCuber mentioned).
2. It's slower than Roux. Roux + CFOP is a really slow combination. From what I understand, one of the benefits of solving F2B instead of F2L is that LSE +CMLL can mostly be executed faster than OLL+PLL/CLL+ELL.

 

[ben1996123]

8355/roux

 

[jskyler91]

8355/roux

I would also say roux, I wish I had started off there and it only takes 9 algs to be pretty quick and only 42 for the advanced version.

 

[waffle=ijm]

lol advanced roux isn't just doing full CMLL. it's learning LSE shortcuts, blockbuilding techniques and looking sexy while you solve.

 

[5BLD]

you do f2b then cmll or 2 look cmll (2 Algs) finish f2l and finally ether ell or ell and epll.

Not original at all. It's a version of noobroux. Which sucks and doesn't teach you fundamental LSE techniques.

I'd it's good to go for Roux. Sure 8355 is good and fits best the criteria mentioned however Roux/Petrus will introduce fundamental techniques early on. Soon it won't just be about low no of algs, and when you make the change you gotta be prepared.

By the way, (edit:as waffle mentioned) being 'advanced' at least for Roux isn't about the no of algs you know. It's about constantly expanding your knowledge and using shortcuts...

 

[Tim Major]

Lol you wrote 'witch'. Its which.
Roux. Spam sune for oll. jperm for pll. then moo.

Better to learn Antisune and Niklas. Antisune+Niklas=J-perm.
And you learn 2 CMLLs/4 with mirrors this way. Plus, shorter alg.

Biggreen could probably sub 12 with this method.

 

[Kirjava]

you can do any CMLL with a niklas/sune combo

 

[PandaCuber]

lol advanced roux isn't just doing full CMLL. it's learning LSE shortcuts, blockbuilding techniques and looking sexy while you solve.

I was reading this thinking, 'Im an advanced cuberrr!...wait...' lol. Almost there,

From what I understand, one of the benefits of solving F2B instead of F2L is that LSE +CMLL can mostly be executed faster than OLL+PLL/CLL+ELL.

Most CMLL can be solved sub 1.5 and Gilles Roux states in his page that LSE can done in under 3.
And the benefits for F2B is movecount. How a cross can be made in 8 moves, well one 'goal' in Roux, is to get the first block under 8 moves as well (my goal)
Other benefit is no rotation,.. We could go on and on.
'
Either way, you only need a few algs to be decent at roux

 

[tasguitar7]

you can do any CMLL with a niklas/sune combo

I'm suddenly on a mission to find all of them this way.

On topic, wouldn't the answer obviously be Heise considering it has 0 algs? If you want standard speed methods only then Roux, but out of all "advanced" methods Heise undeniably requires the fewest algs/most intuition.

 

[emolover]

I'd it's good to go for Roux. Sure 8355 is good and fits best the criteria mentioned however Roux/Petrus will introduce fundamental techniques early on. Soon it won't just be about low no of algs, and when you make the change you gotta be prepared.

Another option would be Petrus F2L with 8355 LL which requires 2 algs total and still will teach the fundamental techniques of block building which is always a good thing to learn.

 

[LeighzerCuber]

For what puzzle? I'm just going to assume 3x3. This method requires no algorithms and is considered advanced.

This is super exciting. I never knew there was a method requiring no algorithims. Yay!

 

[Kirjava]

I'm suddenly on a mission to find all of them this way.

I was going to post a list of them somewhere once, but it's trivial finding them all.

On topic, wouldn't the answer obviously be Heise considering it has 0 algs? If you want standard speed methods only then Roux, but out of all "advanced" methods Heise undeniably requires the fewest algs/most intuition.

If you don't consider XYX'Y' an alg (which you should), Heise has algorithms - you just have to create them yourself. Just because it requires no sequences to be memorised (actually, it does require one) doesn't mean it makes use of no algorithms.

 

[Escher]

'Algorithm' is a pretty arbitrary term, any logical structure with which to do anything useful on the cube can be called an algorithm, the most obvious being XYX'Y'...

I haven't really read the thread, the sexy method from Erik Akkersdijk is fairly easy to learn and has a few things to learn, it's basically 8355.

Of course the cube can be solved with 'no algorithms', you just have to get a great big skip of a non-blockbuilding last step.

Btw, I'm pretty sure every last layer position can be solved with 3 Niklas/Sune variants (and an <M, U> setup).

 

[buelercuber]

Sexy method has one algorithm , sexy move

http://cube.crider.co.uk/beginner.php

 

[Godmil]

Again, he specifically says he's interested in Advanced methods. However you choose to define advanced it doesn't include 8355 or Sexy.

Also Heise was mentioned in post #3.

 

[Ickathu]

Probably roux. When I use roux, I do everything except Corner Permutation and Orientation intuitively, and those 2 stages can be done intuitively as well, I just prefer algs.