[Help Thread] Roux Discussion and Help

[GuRoux]

1) predict BU and work from there

setup: M' U2 M' U M2
is there some type of trick you use to avoid doing M2 U U2 M U2 M?

 

[TDM]

M2 U //First part of ULUR
M2 // "the M2 that places the UL and UR edges"
U' // "put the white edge diagonally across from the other white edge". By white edge, CheesecakeCuber should mean the DB and DF edges (So they are actually yellow edges if you're doing white top.
M U2 M' // "bring them together on the U layer"
M2 // Intuitive finish (and obvious cancellation)

Sorry, I made a typo in my setup. How about now?
(also I do white on D and yellow on U, but if you do white on U then just swap 'white' for 'yellow' in my post)

 

[Tao Yu]

Sorry, I made a typo in my setup. How about now?
(also I do white on D and yellow on U, but if you do white on U then just swap 'white' for 'yellow' in my post)

M2 // ULUR
U' // put the white edges opposite each other

MU2M //Bring them together on the U layer
U2M2U2 //Intuitive finish

A shorter way though is M2UM2U2MU2M. The "put the white edges opposite each other" step sometimes makes your LSE one move longer. In my opinion, it's better to just track the DB and DF edges during ULUR and then just solve the rest intuitively.

 

[GuRoux]

Not sure I understand this... How would you deal with this case?
Setup: M' U2 M' U2 M2 U' M2

Here's what I would do:
M2 // ULUR
U' // put the white edges opposite each other
... but then what?

i think after M2 U', it is intuitive. it's just a dot case + M2.

 

[rybaby]

isn't that kind of hard? after 4b, you can only know the solved edge if it is at UF. any where else on the cube and you have to look at the bottom or do M'(s).

Yeah, sometimes it's a bit frustrating. I honestly don't even think about it when I'm solving; it feels pretty natural to recognize the cases. I'll look into that BU stuff and see if I can understand how others do it.

 

[TDM]

i think after M2 U', it is intuitive. it's just a dot case + M2.

M2 // ULUR
U' // put the white edges opposite each other

MU2M //Bring them together on the U layer
U2M2U2 //Intuitive finish

A shorter way though is M2UM2U2MU2M. The "put the white edges opposite each other" step sometimes makes your LSE one move longer. In my opinion, it's better to just track the DB and DF edges during ULUR and then just solve the rest intuitively.

Thanks. Yeah, the problem was that it made the LSE longer, which confused me; I was thinking this as a way of recognising and doing the same moves more easily rather than an entirely new way to solve them. I just misunderstood the method.

E: also I learned how to do ULUR better thanks to looking at some of Athefre's example solves in the thread about this method, so thanks Athefre!

 

[GuRoux]

Thanks. Yeah, the problem was that it made the LSE longer, which confused me; I was thinking this as a way of recognising and doing the same moves more easily rather than an entirely new way to solve them. I just misunderstood the method.

E: also I learned how to do ULUR better thanks to looking at some of Athefre's example solves in the thread about this method, so thanks Athefre!

personally i have been trying to take up the predicting BU way of recognition. That would solve it like you normally do.

 

[TDM]

personally i have been trying to take up the predicting BU way of recognition. That would solve it like you normally do.

I've been doing Roux quite a lot recently, so I'll practise both and see which I prefer. Thanks for the help.

 

[PJKCuber]

I think I'll just use Waffo's tutorial for Fuu

 

[GuRoux]

I think I'll just use Waffo's tutorial for Fuu

oh, that's what rybaby does. personally i don't think it's good because you have to do a move getting the solve piece to the correct position.

 

[DeeDubb]

oh, that's what rybaby does. personally i don't think it's good because you have to do a move getting the solve piece to the correct position.

Yeah, the U2 at the beginning is a bit of a waste. You should learn to recognize all the cases with your UL/UR edges in the front and back (since that's the position they will be in when you finish placing them)

 

[GuRoux]

Yeah, the U2 at the beginning is a bit of a waste. You should learn to recognize all the cases with your UL/UR edges in the front and back (since that's the position they will be in when you finish placing them)

as far as i can see, the best recognition method is predicting BU and tracking D layer edges.

 

[DeeDubb]

as far as i can see, the best recognition method is predicting BU and tracking D layer edges.

I think that's coming along naturally for me. I think it's something that will get better with time. I find myself peeking at the bottom less and less these days. I can't really describe my method, because it's pretty subconscious. So, my method is PRACTICE PRACTICE PRACTICE

 

[GuRoux]

I think that's coming along naturally for me. I think it's something that will get better with time. I find myself peeking at the bottom less and less these days. I can't really describe my method, because it's pretty subconscious. So, my method is PRACTICE PRACTICE PRACTICE

that's pretty cool. you can actually predict it by knowing any of the hidden colors, so i guess that's what you're doing.

 

[obelisk477]

About a month at 23 seconds. I won't switch directly, I'll practice both methods to see which one seems more easy. Both of them are nice. I want to get good at big cubes and 3x3 and BLD.

I was stuck at a 22.5 second global average for 4 YEARS, and I suddenly dropped to 18 this summer and am still progressing. You've been cubing since May of this year (2 months, compared to my 7 years). Putting hard work into cubing is the only way to get better, and one month stuck at an average is certainly not hard work.

 

[rybaby]

oh, that's what rybaby does. personally i don't think it's good because you have to do a move getting the solve piece to the correct position.

It's kinda hard to describe, but a lot of times I actually am able to avoid an extra move. I could look into another recognition method, but it's kinda hard to change. I stumbled upon Brusinque's tutorial a few days ago, so I might try that. I don't know, but I bet drilling my recognition for those could save me 0.25-0.5 s depending on how well I can predict 4c during 4b.

 

[Tao Yu]

I think that's coming along naturally for me. I think it's something that will get better with time. I find myself peeking at the bottom less and less these days. I can't really describe my method, because it's pretty subconscious. So, my method is PRACTICE PRACTICE PRACTICE

This pretty much happened to me as well. However, looking at my solves, I think I'm tracking DB and DF most of the time.

I don't actually see myself tracking BU anytime soon. Tracking DB and DF seems to be working out just fine for me.

 

[Renslay]

Yeah! Some new PBs!

I didn't solve a cube for almost a week. I just sit down and did some solves... Didn't expect that:

Avg50: 14.41 (Damn close!)
Avg12: 12.89 (Woaaah! New PB by 0.6 seconds!)
Avg5: 12.28 (Woah! New PB too!)
Best: 11.44

Avg12 times:
12.39, 13.18, 12.87, 14.04, (18.18), 13.42, 13.52, 12.17, 12.12, 12.55, (12.66), ((12.07))

The mindblowing is the consistency in the Avg5 (last 5 solves). All of them are 12.xy, with a standard deviation of 0.24...!
My reaction: https://www.youtube.com/watch?v=05CelOOq8nQ

 

[CheesecakeCuber]

Not sure I understand this... How would you deal with this case?
Setup: M' U2 M' U2 M2 U' M2

Here's what I would do:
M2 // ULUR
U' // put the white edges opposite each other
... but then what?

You might have already figured this out but here's my explanation:
(I scrambled yellow top blue front)
M2 //UL UR edges
Now like you said, white edge is at DF so put the other white edge across like you did:
U'
You have to recognize that you can basically turn this Fuu case into the "H case" that Waffle (I think) calls it. So the fastest thing to do now is: M U2 M' U2 M2 U2 M2

 

[TDM]

M U2 M' U2 M2 U2 M2

Why not M U2 M U2 M2 U2 to avoid a whatevertheM-sliceequivalentofanAUFis?