[Help Thread] Roux Discussion and Help

[aNub]

Yes, for two-handed solving, Roux is just as fast as CFOP. Your improvement depends entirely on how much practice you do, so of course, if you enjoy a method more then you'll improve faster with it. Although the methods matters when you get down to world class times, it's still debated which method is better so the difference is likely so close that it's irrelevant.

However, you may want to consider other events. CFOP is better for big cubes and Roux is better for one-handed solving; however, Roux is still okay on 4x4/5x5 with magnets and 3x3 stage is a small part of 5x5+ solves, making Roux's bigcube weakness less of an issue. But most Roux users will use CFOP on bigcubes anyway, because it is definitely better on those puzzles. It's not too much of a hassle to use different methods for different puzzles.

Are you a Roux or CFOP user, and what times do you get on 3x3?

 

[TDM]

Are you a Roux or CFOP user, and what times do you get on 3x3?

I'm primarily a Roux user. I average low 10 with Roux and although I don't do CFOP sessions any more (only the occasional solve when I see an easy cross/bad block), I think I average low-mid 11.

 

[Solvador Cubi]

I'm thinking about learning the Roux method. I enjoy M slice turns and methods with few algs!

When I start, what are the average move counts I should expect for each step?

15 First 1x2x3 block
15 Second 1x2x3 block

8 Orient Corners
12 Permute Corners

7 Orient L6E - 9 cases
6 Permute UL/UR - 5 cases, keep edges oriented
4 Midge Permutation: 4 cases

Then can I anticipate a total of around 67 stm ?


Then as I progress (better block building, 1-look U corners, etc.)
what are decent move counts at that time?


thanks!

-= Solvador Cubi

 

[TDM]

I'm thinking about learning the Roux method. I enjoy M slice turns and methods with few algs!

When I start, what are the average move counts I should expect for each step?

15 First 1x2x3 block
15 Second 1x2x3 block

8 Orient Corners
12 Permute Corners

7 Orient L6E - 9 cases
6 Permute UL/UR - 5 cases, keep edges oriented
4 Midge Permutation: 4 cases

Then can I anticipate a total of around 67 stm ?


Then as I progress (better block building, 1-look U corners, etc.)
what are decent move counts at that time?


thanks!

-= Solvador Cubi

If you're just learning Roux, then I would expect:
FB: 10-15
SB: 20
CMLL: If you know COLL then you can use this here, but if you're using 2-look then your movecounts are accurate
LSE: There's a lot of variation in each step, so I would just aim for around 16 for the whole of LSE and not worry too much about individual substeps for now.

So yeah, your total movecount should be around what you suggested to begin with, but it should quickly improve (especially blocks and CMLL)

For advanced Roux solvers, movecounts are approximately:
FB: 8
SB: 15
CMLL: 11
LSE: 14
Total: 48

 

[Marbles]

Hey everyone, this is my first post here.
I am currently learning full CMLL with the help of Kian's resources (thank you; about 30 Algs in) and have a simple question about CMLL recognition.
How do you differentiate the two O cases (adjacent vs. diagonal swap) when the Headlights happen to be in the back?
At the moment, I am torn between:
1. Simply doing a U' and then checking left side. In case it is an adjacent swap, the headlights are then already in the correct position for the alg. In case it is a diagonal swap, I waste a move.
2. Checking whether both opposite stickers on the left and right are the same colour or not. If they are, i do the diagonal swap alg, if not I do U' into the adjacent swap alg. This never wastes moves, but takes me quite some time to check.

What do you do / think is faster in the long run? I am aware that already looking for corner orientation / permutation while doing SB is optimal, but I don't always manage to do that completely.

Thank you and have a great cubing day
Max

 

[shadowslice e]

Really you should be looking at two (adjacent) sides when recognising so when the headlights are in the back you can tell the difference by looking at one of the sides (but not necessarily AUFing) and seeing that the colours are not opposite colours

 

[TDM]

Hey everyone, this is my first post here.
I am currently learning full CMLL with the help of Kian's resources (thank you; about 30 Algs in) and have a simple question about CMLL recognition.
How do you differentiate the two O cases (adjacent vs. diagonal swap) when the Headlights happen to be in the back?
At the moment, I am torn between:
1. Simply doing a U' and then checking left side. In case it is an adjacent swap, the headlights are then already in the correct position for the alg. In case it is a diagonal swap, I waste a move.
2. Checking whether both opposite stickers on the left and right are the same colour or not. If they are, i do the diagonal swap alg, if not I do U' into the adjacent swap alg. This never wastes moves, but takes me quite some time to check.

What do you do / think is faster in the long run? I am aware that already looking for corner orientation / permutation while doing SB is optimal, but I don't always manage to do that completely.

Thank you and have a great cubing day
Max

It's important here to note the distinction between pairs of "opposite" colours and "adjacent" colours.

Opposite colours: colours which are on opposite faces of the cube, e.g. red+orange or blue+green
Adjacent colours: colours which are on adjacent faces, e.g. blue+red, orange+blue, etc.

When you have an adjacent swap case with the headlights in the back, the two colours on the front (the FUL and FUR stickers) will be opposite colours, so the case will look similar to the diagonal swap case. However, the two colours on the right (RUF and RUB stickers) will be adjacent. This is what's different from the diagonal swap case, where the two stickers on the R face are adjacent colours. This is how most people will recognise it, as this is faster than both of the options you suggested (it doesn't require any extra AUFs or even looking at the L side).

 

[Marbles]

Thank both of you. For some reason, I missed that you can differentiate the cases by looking at only FUL / FUR and RUF / RUB.

 

[1001010101001]

I have been practising FB for an... uncomfortable amount of time and now all I can do is sub-5 it. How am I supposed to do it? I try to do it optimally instead of starting with a square.

 

[TDM]

I have been practising FB for an... uncomfortable amount of time and now all I can do is sub-5 it. How am I supposed to do it? I try to do it optimally instead of starting with a square.

Try just doing a square and then another pair. This is what top solvers do for at least 90% of cases. For now, don't worry too much about influencing the pair during the square if you're finding it too difficult.

 

[Jlvs2run]

I have been practising FB ... How am I supposed to do it?

This block trainer set on Roux FB can be helpful.

Try just doing a square and then another pair. This is what top solvers do for at least 90% of cases.

That's clear. I misread it, sorry.

 

[TDM]

Do you mean to place the first matching edge square that you see immediately?

Not quite sure what you mean by "first matching edge square". I use my inspection to see which 1x2x2 square is best, and then I track the pair which goes with it to form a complete 1x2x3.

(because of the colour neutrality I use, x2/y, there's only ever one possible such pair which completes my FB)

 

[Roux_Over_CPOF]

Hi! I am a roux solver for 3x3, and wondering if anyone knew about some intermediate roux subsets to share, or post a comment about a subset (for roux) that you made yourself! Good luck!

 

[1001010101001]

Hi! I am a roux solver for 3x3, and wondering if anyone knew about some intermediate roux subsets to share, or post a comment about a subset (for roux) that you made yourself! Good luck!

You can use UFUD or build both blocks at the same time.

 

[Wyan]

so here’s the scoop. I have been learning and practing roux for 2 months now and have an average of 45s and a pb of 36... I have concluded that this is not very good, at all, and that it is most likely my intuitve blockbuilding skills that are trash. I know there is no way to “teach” this, but if a few people could tell me how they got fast at recognizing colors, and/or a helpful orientation of the cube / a video to help get my mind in gear when doing the blocks? I do NOT know CFOP.

PS: is a lengthy time on this usually caused by innefficiency with moving? Because i have very few pauses, and inspections are attempted during movements.

 

[1001010101001]

so here’s the scoop. I have been learning and practing roux for 2 months now and have an average of 45s and a pb of 36... I have concluded that this is not very good, at all, and that it is most likely my intuitve blockbuilding skills that are trash. I know there is no way to “teach” this, but if a few people could tell me how they got fast at recognizing colors, and/or a helpful orientation of the cube / a video to help get my mind in gear when doing the blocks? I do NOT know CFOP.

PS: is a lengthy time on this usually caused by innefficiency with moving? Because i have very few pauses, and inspections are attempted during movements.

Count your moves while you do untimed solves. Get the movecount under 60 and then practice lookahead, then when you are smooth do 50 moves and lookahead, and so on.

 

[Wyan]

Count your moves while you do untimed solves. Get the movecount under 60 and then practice lookahead, then when you are smooth do 50 moves and lookahead, and so on.

This includes FB and SB, together? thanks!

 

[Marbles]

Hey everyone!
I noticed that every now and then I get the opportunity to finish the first two layers instead of doing 4b (Solving UL + UR) by doing a M U2 M' (or M' U2 M), giving me a U Perm to finish the solve. Do you do this as well / think this is a good idea? Or should the Roux UL / UR Solution be more efficient? I guess if finishing the first two layers give me an H or Z perm, it is not worth it?

To set up a situation similar to the one I describe, simply do a U Perm (e.g. (R U') (R U) (R U) (R U') R' U' R) and then M U2 M'.

Thank you in advance

 

[xyzzy]

Hey everyone!
I noticed that every now and then I get the opportunity to finish the first two layers instead of doing 4b (Solving UL + UR) by doing a M U2 M' (or M' U2 M), giving me a U Perm to finish the solve. Do you do this as well / think this is a good idea? Or should the Roux UL / UR Solution be more efficient? I guess if finishing the first two layers give me an H or Z perm, it is not worth it?

To set up a situation similar to the one I describe, simply do a U Perm (e.g. (R U') (R U) (R U) (R U') R' U' R) and then M U2 M'.

Thank you in advance

M2 U M U2 M' U M' U2 M' will solve that case, fyi. (This is just MU U perm (M2 U M U2 M' U M2) cancelling into insert (M U2 M').)

General consensus is that this is not worth it; if you look at how the MU U perm algs work, or the common MU H and Z perm, algs, they're all essentially doing 4b then 4c. You're not skipping 4b/4c by setting up to EPLL; you're just doing it less efficiently.

 

[obelisk477]

Hi! I am a roux solver for 3x3, and wondering if anyone knew about some intermediate roux subsets to share, or post a comment about a subset (for roux) that you made yourself! Good luck!

I genned 6-flip CMLL a while back