How should I improve my times now?

[Whitebeard250]

I currently focus on my Cross and F2L - I need ~25 seconds for both (5s cross, 20s F2L), I hope that I can improve this times soon. (I do 30 to 150 solves per day at the moment). Also I'm not sure if 4LLL would make me faster than the LBL-method for the last layer (top-layer) of the cube.

If you were you I'd just start learning them now, they shouldn't interferer with your F2L progress and it's definitely a faster LL than LBL last layer. I'd also start looking into look-ahead for F2L.

 

[Cooper97]

and it's definitely a faster LL than LBL last layer.

The problem is that my finger tricks are not the best, and that's okay because it feels cool to solve the cube a little bit more slowly, but the beginner LBL last layer algorithms are faster for this because they are shorter. Last-Layer with my LBL last layer method are ~5 algorithms with an average of 4 to 7 moves per algorithm compared to the 4LLL algorithms which are average up to 12 moves per algorithm. So the 4LLL algorithms are longer and I need more time to do them.

 

[Rpotts]

So just to clarify, what do you do for last layer now? There is no "standard beginner's method", so what you do might be different than what I, and other posters, are thinking of.

Most people are pointing you to learn 4LLL, which I recommend as well. Learning 4LLL consists of 16 algorithms, some of which you certainly already know. Typically 4LLL is split up into 2-look OLL and 2-look PLL, where you first orient all the last layer edges to form the yellow (if you start on white) cross, then orient the corners to make a yellow face, then permute the corners, then permute the edges to solve the cube.

First you orient edges and make a yellow cross using one of 3 algorithms -

F (R U R' U') F' // "L case" - F sexy F'
F (U R U' R') F' // "Line" - F inverse sexy F'
F (R U R' U') F' f (R U R' U') f' // "Dot" - F sexy F' f sexy f'

Then you orient corners to make a yellow face using one of 7 algorithms -

R U R' U R U2 R' // "Sune"
R' U' R U' R' U2 R // "Backsune" or Antisune
R U R' U (R U' R' U) R U2 R' // "Doublesune" or H - Sune twice
R U2' R2' U' R2 U' R2' U2' R // "Pi" or "The Wheel" - note the 's on R2' / U2'
R U R' U R U2' R2' U' R U' R' U2 R // "Headlights" or U - Sune then Backsune
r U R' U' r' F R F' // "T" - fat sexy then fat sledgehammer
F' r U R' U' r' F R // "L" or "Bowtie" - same as "T" but with the F' as the first move, instead of last.

Then you permute corners to leave only edges using one of 2 algorithms -

(R U R' U') R' F R2 U' R' U' R U R' F' // T Perm, swaps the two right side corners
F (R U' R' U' R U R') F' (R U R' U') (R' F R F') // Y Perm, swaps diagonal corners

Then you permute edges and solve the cube using one of 4 algorithms -

R2 U (R U R' U') R' U' R' U R' // Ub Perm - clockwise cycle of 3 edges
R U' R U R (U R U' R') U' R2 // Ua Perm - counterclockwise cycle
M2 U' M2 U2' M2 U' M2 // H perm
M' U' M2 U' M2 U' M' U2 M2 // Z perm

Alright I don't know why I chose to go to such great effort there to do that but once I got started I didn't want to not finish it. Those are the algs I would recommend learning when learning 4LLL for the first time. You probably know some of them already, or you may know other algs to solve some of the same cases. More info can be found on the Wiki

Good luck

 

[Rok]

A good techinique to reduce time after doing surplus practice is by watching videos(or real people) of people who are faster than you. This has a psycological effect and will improve your times. But it will work only when you have done a lot of practice. www.youtube.com/watch?v=f2oK_GmoWhk it is a sample.

 

[Solvador Cubi]

Hi Cooper97,

I assembled some 3x3 methods that can be used to solve the cube in under a minute without memorizing of a bunch of algorithms. However, both of these methods still have useful steps and algs that are used for more advanced CFOP.

The intermediate method is called FIXLL and has a 4 Look Last Layer. My goal for it was to use the fewest number of algorithms to memorize instead of the 16 mostly unrelated ones. It uses only 9 main algorithms to complete the last layer in at most 4 steps.

Here's a link to the one-page quick reference guide:
http://solvexio.cf/app/#/FIXLL_OnePage

I hope some of that will help you!
Solvador Cubi

 

[Cooper97]

So just to clarify, what do you do for last layer now? There is no "standard beginner's method", so what you do might be different than what I, and other posters, are thinking of.

The beginner method I currently use for the last layer is this one:

1. Make a cross on the last-layer (like in 4LLL)
2. Orient the cross with R, U, R', U, R, U, U, R', U
3. Permutate the corners to their correct spot with U, R, U', L', U, R', U', L
4. Orient the corners using R', D', R, D on each corner.

I need 8 to 15 seconds in average for this steps. I'm not sure if 4LLL would be faster for me, but I like this beginner method because of the less algorithms it needs (2 for the cross + 3 mentioned above).

Here's a link to the one-page quick reference guide:
http://solvexio.cf/app/#/FIXLL_OnePage

Thanks, I will take a look at it.

Most people are pointing you to learn 4LLL, which I recommend as well.

Yes, I currently learn them too so that I can compare beginner last-layer with the 4LLL method, but it takes a lot of time for me to learn them and to speed up my look-ahead for them.

 

[pglewis]

I need 8 to 15 seconds in average for this steps. I'm not sure if 4LLL would be faster for me, but I like this beginner method because of the less algorithms it needs (2 for the cross + 3 mentioned above).

4LLL (or 2LLL) will likely be significantly slower at first because it's new recognition, new algs to get into muscle memory, and extra time for recall vs. the ones you can already do without much thinking time. I'm a proponent of just picking up a few algs at a time and incorporating them until you're comfortable using them in a speedsolve, add a few more, repeat. You don't have to attack full OLL/PLL/whateverLL in one-shot. It sounds like you're already comfortable with the beginners' method so I'd either learn a single OCLL or PLL alg, then when you see that one show up you have a short-cut at your disposal.

Better lookahead is always a valid answer for improving too, but the sooner you start learning new things to make your last layer more efficient, the sooner you can actually use those "new" algs faster.