Waffle's Roux Tutorial

[5BLD]

How do you practice LSE increasing TPS? I notice your (waffle/Thom) LSE is really smooth and 4c recog is fast.

Is there an exercise other than just solving MU over and over?

 

[MostEd]

5bld: metronome LSE&

 

[5BLD]

I dunno why but metronome makes me have a very rough turn style, when I want the exact opposite. Maybe it's that i'm not doing it right?

 

[MostEd]

try to do it slowly smooth, then gradually increase speed? without metronome then

 

[waffle=ijm]

it's only fast because A/B/C are often combo'd with each other. shortcuts yo

 

[Athefre]

An update to my 4c 3-cycle recognition. It had a problem, the M2U2M'U2 cases were automatically solved as M'U2M'U2M2U2. The solution to this is easy. If you notice that, right before finishing 4b, UB+center and FD+center are opposites (as if you have the "dots" case), after 4b adjust U the opposite way you usually would and do M2. Unless I've looked over a case that would cause this to not work.

EDIT: U2MU2M It's not worth it to examine six stickers to save that move or two.

I know this is waffle=ijm's tutorial. But, I don't know where else to post Roux stuff that I feel is important that others see.

 

[cubersmith]

I feel so bad, I keep putting off learning this.

 

[PandaCuber]

An update to my 4c 3-cycle recognition. It had a problem, the M2U2M'U2 cases were automatically solved as M'U2M'U2M2U2. The solution to this is easy. If you notice that, right before finishing 4b, UB+center and FD+center are opposites (as if you have the "dots" case), after 4b adjust U the opposite way you usually would and do M2.

I know this is waffle=ijm's tutorial. But, I don't know where else to post Roux stuff that I feel is important that others see.

Nice

 

[5BLD]

An update to my 4c 3-cycle recognition. It had a problem, the M2U2M'U2 cases were automatically solved as M'U2M'U2M2U2. The solution to this is easy. If you notice that, right before finishing 4b, UB+center and FD+center are opposites (as if you have the "dots" case), after 4b adjust U the opposite way you usually would and do M2. Unless I've looked over a case that would cause this to not work.

EDIT: U2MU2M It's not worth it to examine six stickers to save that move or two.

I know this is waffle=ijm's tutorial. But, I don't know where else to post Roux stuff that I feel is important that others see.

I discovered a very simple way to recog yesterday.
I found just glancing at the gonna-be BU sticker (is revealed during LR edge insertion) , and the centres, is enough info for the 4c case.

 

[tasguitar7]

Alright so i have been messing around and I'm trying to decide which is better to do with Roux. (I currently use 2look CMLL and average 25 seconds)

I found that if you do 2look CMLL, orientation then permuation it averages 7.5 moves for orientation, and 12.17 moves for permuation. 19.67 total

I also found algs (easy to execute so not optimal) using cube explorer for 2look CMLL, permutation then orientation, which averages 6.3 moves for permutation, and 9.625 moves for orientation. 15.925 total

I also averaged the number of moves in the full CMLL algs found on 5BLD and Pandacuber's roux tutorial. They average 10.5 moves.

Is it a good idea for me to switch to my 2look system for now or stop beating around the bush and learn full CMLL? Another reason I ask is also because I have difficulty with CMLL recognition and I was hoping using my 2look system for now would help me improve recog without overwhelming myself with algs at the same time.

 

[waffle=ijm]

Recognition comes with time and practice. The sooner you work on it. The sooner you get better.

 

[Athefre]

I discovered a very simple way to recog yesterday.
I found just glancing at the gonna-be BU sticker (is revealed during LR edge insertion) , and the centres, is enough info for the 4c case.

That's among the many recognition styles I've used. It provides the information, but for me it wasn't fast enough. There were still solves where I would have a slight delay because of the thinking time to compare edge and center colors. My goal has been to completely remove delays from LSE. I have achieved it with a minor cost of two of the 3-cycles requiring just one move more. I guess there's no easy way around that. But to me it's worth it to average just .5 moves or whatever more so that I have a perfect flow LSE.

 

[PandaCuber]

Alright so i have been messing around and I'm trying to decide which is better to do with Roux. (I currently use 2look CMLL and average 25 seconds)

I found that if you do 2look CMLL, orientation then permuation it averages 7.5 moves for orientation, and 12.17 moves for permuation. 19.67 total

I also found algs (easy to execute so not optimal) using cube explorer for 2look CMLL, permutation then orientation, which averages 6.3 moves for permutation, and 9.625 moves for orientation. 15.925 total

I also averaged the number of moves in the full CMLL algs found on 5BLD and Pandacuber's roux tutorial. They average 10.5 moves.

Is it a good idea for me to switch to my 2look system for now or stop beating around the bush and learn full CMLL? Another reason I ask is also because I have difficulty with CMLL recognition and I was hoping using my 2look system for now would help me improve recog without overwhelming myself with algs at the same time.

I used to fear CMLL. I was like 'OMG 42 Cases!? I'll never learn them D:' But then I started with easy cases and cases that made it so I didnt have to do Y perms (I do slow yperms slow)
Over time recognition gets better and better.
I would say take baby steps. Start with shorter easier algs, Like H's. Then youll get the hang of CMLL.

 

[waffle=ijm]

or get the hard ones over with.

 

[IanTheCuber]

You should probably do the super easy ones first, right when you start, and work your way up the ladder until you know all of them. Them maybe you could modify the method a bit.

 

[Cheese11]

I just learn the cases that have the most move average first, so the hard ones get put of the way.

 

[oll+phase+sync]

I found that if you do 2look CMLL, orientation then permuation it averages 7.5 moves for orientation, and 12.17 moves for permuation. 19.67 total

Permutation with an 9- ,12-, and ZERO-move alg. should average just around 8,6 moves.


EDIT: Learning one aditional CLL per OC-Case dpoubles your chances for 0-move Permutation and give's the opportunity to avoid the 12 move cases => though permutation averages at around 6 moves.

 

[Sarahjdes]

I learned F2L a while ago. Before I decided to switch to Roux. Thing is, I usually end up inserting pairs. Sometimes I'd put the bottom edge along with a corner, but I only do it every 5 solves or so. It's not a reflex for me.

After watching Donovan's Roux tutorial on Youtube, I realised that it never occured to be that I could build a block "upside down". (I solve with white on bottom, since I started learning OLL/PLL, which yellow on top, so I'm used to tracking the yellow sides when I'm doing my CMLL) I'm used to build it with the white edge and "white corners", and I can't help it to build it with the actual white line on the bottom. I want to get used to building it no matter what the orientation of the block is, and then do some L turns to orient it the right way (assuming I'm building it on the L face). Would it be worth it? I feel like I could save a couple of moves and that it might be easier to track pieces after my inspection.

Any tips? I was thinking of starting by positionning my white-blue edge (assuming I'm first building my blue block) somewhere on my next to the blue center, no matter if it's looking up, down, back or front, and work from there. Of course, at some point I'll have to practice building my block starting with the blue-white-orange corner and, let's say, blue-white edge, and so on, but I think I need to start somewhere. I don't want to start going in every directions at once... Does this sound like a good plan following the Roux philosophy?

I hope this was clear... English is not my first language, and even though I don't make too many grammar/spelling mistakes, I feel like I still have problems organising my thoughts.

Oh, and I usually average sub-75 with Roux, with a 20-20-10-25, since I'm still cycling for EO. I start working on it soon.

Thanks a bunch!

 

[PandaCuber]

I Hope we are talking about the same thing. What I usually do it put my corner on the U face, no matter the orientation. bring the edge to the U face with the orientation that I choose, cause the M slice lets me. then I connect them and place them into their corresponding slot/

 

[Sarahjdes]

I Hope we are talking about the same thing. What I usually do it put my corner on the U face, no matter the orientation. bring the edge to the U face with the orientation that I choose, cause the M slice lets me. then I connect them and place them into their corresponding slot/

Yes, this is what I usually do. But I'm trying to get to bloc building without continuously inserting pairs into slots.

But I see that you are sub-19 using the inserting thing, so I guess it's an okay way to do it!

Is everyone else doing it this way? Is there anyone that builds the first block not correctly positionned, and then rotates the L or R face to get in it's "canonical position"?