[Help Thread] Roux Discussion and Help

[IAmAPerson]

x/y color neutrality:
anything on bottom + anything on left (full color neutrality); highly impractical and way too complicated for Roux.

When I was a Roux solver, I was fully CN. I didn't know that was hard/abnormal? I still solve CN whenever I solve Roux.

 

[JediJupiter]

When I was a Roux solver, I was fully CN. I didn't know that was hard/abnormal? I still solve CN whenever I solve Roux.

It's harder for lookahead, whatever suits you best.

 

[IAmAPerson]

It's harder for lookahead, whatever suits you best.

I'm also fully CN with CFOP, Petrus, 2x2, Pyra, Skewb, 4x4 Redux, and I'm also y/z2 CN with ZZ. :/

 

[JediJupiter]

I'm also fully CN with CFOP, Petrus, 2x2, Pyra, Skewb, 4x4 Redux, and I'm also y/z2 CN with ZZ. :/

I'm CN with everything but roux, and with roux I'm fixed block. I don't want to have to remind myself what piece I'm looking for, because right now I notice that piece before I need it and I rarely have to waste time looking for pieces.

Maybe that doesn't actually happen with CN, but it seems like the more colours you use, the more you have to actively search for pieces.

 

[Knut]

For Roux you only really have the need to be x2 y neutral as that gives you the option to use any pair on the cube, while at the same time allowing you to keep one of two colors on the D face. And it makes recognition so much easier.

 

[theROUXbiksCube]

For Roux you only really have the need to be x2 y neutral as that gives you the option to use any pair on the cube, while at the same time allowing you to keep one of two colors on the D face. And it makes recognition so much easier.

x2 y is life

 

[dboeren]

So I bought a 4x4 yesterday at the Atlanta comp and I'm learning to solve it through the reduction method.

Make centers, pair dedges, then do the 3x3 solve using Roux. When I get to the "fix bad edges" step I see that there are an odd number of bad edges. Doh! So, now I need some sort of parity fix. I can keep going and apply some long gnarly algorithm at the very end to fix it, but I expect that if I can fix it NOW while I have less to preserve the algorithms will be shorter and I will save even more time because my L6E won't look as confusing.

I did a search for "roux 4x4 parity" but didn't really find anything. Does anyone know some good algorithms I can use to fix parity at this step? I'm not necessarily expecting to fix everything, mainly just to restore an even number of bad edges. I realize that may still leave me some parity at the end, but it should be less and I suspect it will only be one type (swaps and not flips)?

Or, is this whole thing a bad idea and I should scrap it and learn a 4x4 specific method from the ground up?

Thanks!

 

[theROUXbiksCube]

So I bought a 4x4 yesterday at the Atlanta comp and I'm learning to solve it through the reduction method.

Make centers, pair dedges, then do the 3x3 solve using Roux. When I get to the "fix bad edges" step I see that there are an odd number of bad edges. Doh! So, now I need some sort of parity fix. I can keep going and apply some long gnarly algorithm at the very end to fix it, but I expect that if I can fix it NOW while I have less to preserve the algorithms will be shorter and I will save even more time because my L6E won't look as confusing.

I did a search for "roux 4x4 parity" but didn't really find anything. Does anyone know some good algorithms I can use to fix parity at this step? I'm not necessarily expecting to fix everything, mainly just to restore an even number of bad edges. I realize that may still leave me some parity at the end, but it should be less and I suspect it will only be one type (swaps and not flips)?

Or, is this whole thing a bad idea and I should scrap it and learn a 4x4 specific method from the ground up?

Thanks!

Parity is a pain in the ass and there is no shorter one just because of no preservation of the M slice.

 

[NeilH]

I had a pop on the 3rd solve, but otherwise this is a pretty normal Roux average for me. (I average about 25 with Roux.) I know I need improvement in everything, but please let me know based on this video what I should work on now. Thanks

 

[Renslay]

So I bought a 4x4 yesterday at the Atlanta comp and I'm learning to solve it through the reduction method.

Make centers, pair dedges, then do the 3x3 solve using Roux. When I get to the "fix bad edges" step I see that there are an odd number of bad edges. Doh! So, now I need some sort of parity fix. I can keep going and apply some long gnarly algorithm at the very end to fix it, but I expect that if I can fix it NOW while I have less to preserve the algorithms will be shorter and I will save even more time because my L6E won't look as confusing.

I did a search for "roux 4x4 parity" but didn't really find anything. Does anyone know some good algorithms I can use to fix parity at this step? I'm not necessarily expecting to fix everything, mainly just to restore an even number of bad edges. I realize that may still leave me some parity at the end, but it should be less and I suspect it will only be one type (swaps and not flips)?

Or, is this whole thing a bad idea and I should scrap it and learn a 4x4 specific method from the ground up?

Thanks!

I'm using the Meyer Method, which is a Roux-variant of the reduction method.

For parity, I use the 3-flip algorithm recommended in the video:
(Rw' U' Rw' U2') (Rw' U2 Rw' U' Rw U2) (r U2 Rw U' Rw' U2 Rw' U2') (Rw' U' Rw')
alg.cubing.net
You can enforce a good EO with this, so you almost always get a 3-bad-on-top-1-bad-on-bottom case.

 

[TDM]

So I bought a 4x4 yesterday at the Atlanta comp and I'm learning to solve it through the reduction method.

Make centers, pair dedges, then do the 3x3 solve using Roux. When I get to the "fix bad edges" step I see that there are an odd number of bad edges. Doh! So, now I need some sort of parity fix. I can keep going and apply some long gnarly algorithm at the very end to fix it, but I expect that if I can fix it NOW while I have less to preserve the algorithms will be shorter and I will save even more time because my L6E won't look as confusing.

I did a search for "roux 4x4 parity" but didn't really find anything. Does anyone know some good algorithms I can use to fix parity at this step? I'm not necessarily expecting to fix everything, mainly just to restore an even number of bad edges. I realize that may still leave me some parity at the end, but it should be less and I suspect it will only be one type (swaps and not flips)?

Or, is this whole thing a bad idea and I should scrap it and learn a 4x4 specific method from the ground up?

Thanks!

I do my orientation parity before CMLL. That way I can use a faster algorithm than one that preserves the corners:
Rw2' F2 Rw U2 Rw U2' x U2 Rw U2' Rw' U2 Rw U2' Rw2 U2' x'
You can look at the bad edges without having the corners solved, so recognition for parity is still the same.
If you want to do parity and preserve the corners (so if you want to do it during LSE), then change the Rw moves to slice moves:
r2 F2 r U2 r U2 x U2 r U2 r' U2 r U2 r2 U2 x'
I find doing an R2 before/after makes it slightly easier to execute:
Rw2 F2 r U2 r U2 x U2 r U2 r' U2 r U2 r2 U2 (R2 x')

(E: this flips UB, not UF, like most other parity algs do)

 

[dboeren]

I'm using the Meyer Method, which is a Roux-variant of the reduction method.

For parity, I use the 3-flip algorithm recommended in the video:
(Rw' U' Rw' U2') (Rw' U2 Rw' U' Rw U2) (r U2 Rw U' Rw' U2 Rw' U2') (Rw' U' Rw')
alg.cubing.net
You can enforce a good EO with this, so you almost always get a 3-bad-on-top-1-bad-on-bottom case.

Thanks. I was browsing the 4x4 methods page on the wiki last night and came across Meyer along with the Shadowslice variant. I'll put those on my list to learn soon.

 

[dboeren]

Parity is a pain in the ass and there is no shorter one just because of no preservation of the M slice.

Well, M slice and the U layer as well. It should be recognizable before doing the U-corners. But if there isn't a better alternative known, then there isn't a better alternative known.

 

[JustinTimeCuber]

How do I get rid of the Roux from my cube?
Sometimes I pick up a cube and it makes a Roux sound.
It's starting to smell like Roux.
Sometimes it even is Roux.
Do I need to buy a new cube?
I mean, heck it even tastes like Roux.
Why is there so much Roux in my cube?

 

[Ross The Boss]

https://www.youtube.com/watch?v=ZDaVI5MZE7A

I had a pop on the 3rd solve, but otherwise this is a pretty normal Roux average for me. (I average about 25 with Roux.) I know I need improvement in everything, but please let me know based on this video what I should work on now. Thanks

ok, inefficiency at all steps is a problem you have. sitting down and figuring out more efficient solutions to common cases in f2b and lse would help with planning and instant recognition.
you also do not have a very good M move execution. actually, your solutions hardly used any Ms, which is bad because that means you are missing out on many more efficient solutions to step 4b (solving ULUR).
something which might have spawned from this apparent reluctance to do M it the way you rotate into step 4c. that is not a good thing to do.
however, you can turn decently fast which could be an advantage once you become more efficient.

 

[PenguinsDontFly]

How do I get rid of the Roux from my cube?
Sometimes I pick up a cube and it makes a Roux sound.
It's starting to smell like Roux.
Sometimes it even is Roux.
Do I need to buy a new cube?
I mean, heck it even tastes like Roux.
Why is there so much Roux in my cube?

see! its so gooooooooooooooooooooood

 

[JustinTimeCuber]

see! its so gooooooooooooooooooooood

but it smell funny D:
It smeels liek triangel

 

[JustinTimeCuber]

4k posts wat
tis derpy

 

[NeilH]

ok, inefficiency at all steps is a problem you have. sitting down and figuring out more efficient solutions to common cases in f2b and lse would help with planning and instant recognition.
you also do not have a very good M move execution. actually, your solutions hardly used any Ms, which is bad because that means you are missing out on many more efficient solutions to step 4b (solving ULUR).
something which might have spawned from this apparent reluctance to do M it the way you rotate into step 4c. that is not a good thing to do.
however, you can turn decently fast which could be an advantage once you become more efficient.

Alright I think I'll try to pick up GuRoux or 5BLD's finger tricks and also just do untimed solves where I'll try to find efficient solutions. Thanks, this helped

 

[Ross The Boss]

but it smell funny D:
It smeels liek triangel

goddamnit. you wasted the 4000th post on this??? ugh