[Help Thread] Roux Discussion and Help

How do you recognize the M slice edge cycles quickly/while performing L/R reduction? They all seem to look similar to me. What I'm doing right now is looking at the U/D faces and if there's one square of opposite color on U, put it in it's proper position on the D layer (e.g. by performing M' U2 M U2), but that only covers some of the cases.
 
Teoidus said:
How do you recognize the M slice edge cycles quickly/while performing L/R reduction? They all seem to look similar to me. What I'm doing right now is looking at the U/D faces and if there's one square of opposite color on U, put it in it's proper position on the D layer (e.g. by performing M' U2 M U2), but that only covers some of the cases.
Click to expand...

I use Athefre's approach. However, I wouldn't recommend this since it can add on a couple of extra moves occasionally. Instead, I think what most fast people use is to track the BU edge. Try to identify which edge will go to BU as you're inserting ULUR: that should give you enough information to know which cycle it is.

Question for people who use BU prediction: does this work for misoriented centres too? I'm having a hard time recognising these cases with Athefre's way and I'm thinking about switching.
 
Okay, Athefre's approach seems like a more formal/efficient version of what I've been trying to do; testing it out and it seems pretty straightforward.

Not quite clear on BU prediction though; does that mean while placing ULUR i track the edge that belongs in BU, or I track the edge that was at BU when starting ULUR?
 
TDM said:
I use Athefre's approach. However, I wouldn't recommend this since it can add on a couple of extra moves occasionally. Instead, I think what most fast people use is to track the BU edge. Try to identify which edge will go to BU as you're inserting ULUR: that should give you enough information to know which cycle it is.

Question for people who use BU prediction: does this work for misoriented centres too? I'm having a hard time recognising these cases with Athefre's way and I'm thinking about switching.
Click to expand...

yes, it works just as well for misoriented centers.

for Teoidus: neither, you predict the color of the BU sticker at the end of UL/UR. that's a little confusing, have you seen my video on it? https://www.youtube.com/watch?v=9AlGuW9AT1o
 
So I started cubing when there was almost no one doing methods other than CFOP (Fridrich at the time), and I now average right about 15 seconds for CFOP. I want to switch to Roux really bad, and I now consider starting with CFOP to be my biggest regret of my cubing journey. A few weeks ago, I decided that I was going to change to Roux full time, and I did un-timed solves for a while. When I decided to start timing my solves, I was incredibly frustrated that my solves were more than double the time of my CFOP times, and after 6 solves, I had switched back to CFOP and gave up with Roux.

I want to try switching to Roux again, and I need some tips on how not to be so frustrated at being slower. Should I even time my solves? Any help would be much appreciated.
 
theawesomecuber said:
So I started cubing when there was almost no one doing methods other than CFOP (Fridrich at the time), and I now average right about 15 seconds for CFOP. I want to switch to Roux really bad, and I now consider starting with CFOP to be my biggest regret of my cubing journey. A few weeks ago, I decided that I was going to change to Roux full time, and I did un-timed solves for a while. When I decided to start timing my solves, I was incredibly frustrated that my solves were more than double the time of my CFOP times, and after 6 solves, I had switched back to CFOP and gave up with Roux.

I want to try switching to Roux again, and I need some tips on how not to be so frustrated at being slower. Should I even time my solves? Any help would be much appreciated.
Click to expand...

I think you may be able to get some good tips from TDM.

Essentially, just do Roux solves constantly and try to block out any CFOP habits that keep breaking in.

I wouldn't bother timing for the first few solves and more just appreciate the method for what it is rather than the speed it can bring.

That said, you can still incorporate some CFOP into you second block and thatay give you an edge because you are already at home with those cases and the lookahead (somthing i still struggle with even though i started with Roux and am now almost sub-14 (i was ~15 until the start of this week then i started practising- shows you just what you can accomplish when you pit your mind to it :o)) so you may already have an advantage compared to sone of the other people who start Roux straight after beginner's (or even those who learned it before beginner's like kirjava)

But all in all, just appreciate the method for what it is and don't worry too much about time just yet.
 
theawesomecuber said:
So I started cubing when there was almost no one doing methods other than CFOP (Fridrich at the time), and I now average right about 15 seconds for CFOP. I want to switch to Roux really bad, and I now consider starting with CFOP to be my biggest regret of my cubing journey. A few weeks ago, I decided that I was going to change to Roux full time, and I did un-timed solves for a while. When I decided to start timing my solves, I was incredibly frustrated that my solves were more than double the time of my CFOP times, and after 6 solves, I had switched back to CFOP and gave up with Roux.

I want to try switching to Roux again, and I need some tips on how not to be so frustrated at being slower. Should I even time my solves? Any help would be much appreciated.
Click to expand...

I see it this way: it took you some years, I assume, to get to a 15 sec average with CFOP (at least 3-4 years judging for your comp results). So, you cannot expect to get the same times with Roux after a couple of weeks. But it won't take nearly as much as the first time with CFOP because now you have a lot of extra dexterity and knowledge of the cube that you did not have when you 1st started with CFOP.

So, if you really want to switch, just focus on the new challenge, enjoy every little achievement, and before you realize you'll be close to where you were with CFOP. I'm just barely sub-20 and I also tried Roux some months ago. It was discouraging to get times around 40 seconds and I started to get into other events like 2x2, 4x4, pyra, skewb, Mega... I just couldn't cope with it all so I sort of left Roux on the side but I enjoy the method a lot and definitely plan on learning it well at some time in the future. So, right now I'm doing so many different things that I'm not improving much at 3x3 but I'm having much more fun with every achievement on every event, even though I'm a patzer at all of them. So, I've come to the conclusion that enjoying what I do is more important than being half a second faster or whatever. Do you enjoy Roux? Then, you have your answer.
 
theawesomecuber said:
So I started cubing when there was almost no one doing methods other than CFOP (Fridrich at the time), and I now average right about 15 seconds for CFOP. I want to switch to Roux really bad, and I now consider starting with CFOP to be my biggest regret of my cubing journey. A few weeks ago, I decided that I was going to change to Roux full time, and I did un-timed solves for a while. When I decided to start timing my solves, I was incredibly frustrated that my solves were more than double the time of my CFOP times, and after 6 solves, I had switched back to CFOP and gave up with Roux.

I want to try switching to Roux again, and I need some tips on how not to be so frustrated at being slower. Should I even time my solves? Any help would be much appreciated.
Click to expand...

I was in a similar position to you a while ago: I was averaging low 13 with CFOP but knew Roux was faster and wanted to switch. I do agree that, as you've found, motivation to continue practising is the hardest part of switching methods.

One thing I found helped was knowing that I would be faster if I kept going. I didn't doubt that Roux was faster than CFOP, so I knew that in the long run I'd not only get down to my old speed, but then keep going and become even faster. In fact, I didn't actually think the difference was that big when I started. As I've got faster with Roux, I've only seen the gap between it and CFOP grow.
Another thing was competitions. There are quite a few UK cubers who use Roux, so talking to them was helpful. Also seeing Alex Lau win Euros and then UK Championships while being second in the world was showing me that Roux definitely did have potential, even if it wasn't widely used. While I don't know how many Roux users there are where you live, hopefully seeing people like Alex Lau's past results gives some inspiration.

However, when you start, you're slow. That's a fact. When I first started practising Roux seriously, I didn't completely switch method and abandon CFOP. I know that's what people suggest, but it demoralises you. People speedcube for fun - what's the point if you're not enjoying it? So instead I regularly solved with both methods. Some CFOP, some Roux. I think it's good to keep this up until you get to around 25, maybe 20, seconds.

Once you reach this stage you start enjoying solving with Roux a lot more, and so it's probably best to stop doing CFOP at this point. This is when you need to start focusing your practice onto each step, because you'll actually find you're bad at all of them. With CFOP, you might find it hard to see where to improve next. You might not be able to improve your LL TPS, for example, and maybe your F2L lookahead is good too. With Roux on the other hand - in my experience at least - it's very easy to see where your weak spots are.

Most of the speed you have is from your TPS from CFOP, and so there will be huge weaknesses in all your steps - probably multiple problems in all four of them. Start working on one of these at a time. For example, take SB efficiency. If you think your solutions are still too CFOP-ey, find some example solves - either on YouTube or the RCDB. When you see the person do something weird or cool, look at their solution carefully. Do you understand how that solution works? Can you apply a similar idea to solve a slightly different case?

Note: for each step, there are three key features. Lookahead (or recognition), TPS and transition to the next step. This is the order you should usually practice them in. It's far easier to increase your TPS after lowering movecount than the other way round.

LSE: NM centres/BU 4c recognition
CMLL: recognition from multiple angles
LSE: predicting and skipping dots cases
FB: efficiency
FB: planning within inspection
SB: efficiency
FB: transition/lookahead to SB
SB: TPS
CMLL: learn more algs, including OLLCPs and possibly CMLLs from other angles
LSE: TPS
CMLL: drill algs
FB: TPS
CMLL: transition to LSE
LSE: possibly learn the difference between M' U M' and M' U' M' for influencing ULUR?

The further down the list you go, the more unsure I am about the steps. I don't even know if some of them will be any help or not, but for now they're too far in the future for me to worry about.
However, the point of this list isn't really the specifics. It's just to give you an idea of the type of things you can be looking to improve.


I realise I've diverged from the original question and talked more about how to improve than avoiding frustration, but hopefully at least the first half of my post answers your question.
 
Can someone teach me how to solve using Roux via Skype?

So I haven't found a way to learn roux online, because I dont get pdf instructions. Is anyone willing to teach me via skype?
 
shadowslice e said:
Do you know how to blockbuild?

Because if you do,

FB
SB
OCLL (orient remaining corners like in beginner's)
Permute corners (like in beginner's)
Orient edges (basically play around with M U M')
Insert UL/UL
L4E ( you could learn algs for it)

Try also Waffle's site.
Click to expand...

I don't, and I read Waflle's instruction but i couldnt understand it.
 
ortomy said:
I don't, and I read Waflle's instruction but i couldnt understand it.
Click to expand...

Basically, for blockbuilding, take a cube and look for anyway to form a pair (corner-edge/edge-centre. Either is fine).

Now move that out of the way and form the opposite pair (so edge-corner if you formed an edge centre and vice versa)- make sure that all the pieces in the pairs have at least one colour in common. Don't destroy the first pair. Move it out of the way if you see a move will split it up.

Now form the final edge-corner pair which will go in the last place to complete the 1x2x3 first block.

Then do essentially the same thing with the second block but try only to do <R, U, r, M> moves.

Essentially, just see what ways you can form pairs or blocks in anyway you can in anyway you can think of. If it works, just remember and think, "yay, I can use that next time". If it doesn't, just move on and you'll know not to do it next time. Don't be afraid to experiment and try new ways of building the blocks.
 
shadowslice e said:
Basically, for blockbuilding, take a cube and look for anyway to form a pair (corner-edge/edge-centre. Either is fine).

Now move that out of the way and form the opposite pair (so edge-corner if you formed an edge centre and vice versa)- make sure that all the pieces in the pairs have at least one colour in common. Don't destroy the first pair. Move it out of the way if you see a move will split it up.

Now form the final edge-corner pair which will go in the last place to complete the 1x2x3 first block.

Then do essentially the same thing with the second block but try only to do <R, U, r, M> moves.

Essentially, just see what ways you can form pairs or blocks in anyway you can in anyway you can think of. If it works, just remember and think, "yay, I can use that next time". If it doesn't, just move on and you'll know not to do it next time. Don't be afraid to experiment and try new ways of building the blocks.
Click to expand...

I dont get the arrow part at the end. And I always hear "R u and m moves" but on the algs I see F moves.
 
Those algs are for solving the corners after you've built the second block (CMLL). Building the second block should only require <R,U,r,M>
 
See, this is why I just want someone to talk it with me live. So I can understand what Im doing.
 
You must log in or register to reply here.

Similar threads

pcwiz
[Help Thread] CFOP Discussion
144 145 146
Replies
2,918
Views
771,752
abunickabhi
abunickabhi
E
  • Poll Poll
Roux CN discussion (x2/y vs full CN)
2
Replies
22
Views
3,817
abunickabhi
abunickabhi
C
[Free App] Cube Trainer — CFOP + Roux algorithm trainer with timer
Replies
0
Views
124
cellonty
C
T
New method idea: Using Pinkie Pie roux variant with CFOP instead
Replies
4
Views
411
theail03
T
Roux
Should i become full CN with Roux?
Replies
4
Views
536
Roux
Roux
Back
Top