[Help Thread] Roux Discussion and Help

Easy scramble. (I didn't make this one up)
U2 L' R U B2 F L D L' D' F' D' R2 F U2 F2 D B2 L2 U R U' B U' R

Inspection: z' y (red top yellow front)
FB: D R U' B (4)
SB: U' R' U M' U R (6)
CMLL: U F R U R' U' R U R' U' F' R U R' U' R' F R2 U' R' U' R U R' F' (26 (Wow, only if I knew full CMLL...))
LSE: U' M' U M U M' U M' U M' U2 M' U' M2 U' M' U2 M' (18 ( :( ))
54 STM 14.61 PB 3.6 TPS
(Then a few solves later... 10.43...( 5.1 TPS))
 
Last edited:
YddEd said:
Easy scramble. (I didn't make this one up)
U2 L' R U B2 F L D L' D' F' D' R2 F U2 F2 D B2 L2 U R U' B U' R

Inspection: z' y'
FB: D R U' B (4)
SB: U' R' U' M' U R (6)
CMLL: U F R U R' U' F R U R' U' F' R U R' U' R' F R2 U' R' U' R U R' F' (26 (Wow, only if I knew full CMLL...))
LSE: U' M' U M U M' U M' U M' U2 M' U' M2 U' M' U2 M' (18 ( :( ))
54 STM 14.61 PB 3.6 TPS
(Then a few solves later... 10.43...( 5.1 TPS))
Click to expand...

Something is wrong there; I don't get even the blocks. See:
alg.grarron.us
 
YddEd said:
Fixed.
Click to expand...

Still not working.

Edit: instead of z' y' as inspection, I think you wanted to type z' y. Now your blocks are working, but your CMLL is still wrong somewhere.

Please learn to use alg garron (or any similar software) to check your solutions before post it.
 
Last edited:
YddEd said:
Easy scramble. (I didn't make this one up)
U2 L' R U B2 F L D L' D' F' D' R2 F U2 F2 D B2 L2 U R U' B U' R

Inspection: z' y (red top yellow front)
FB: D R U' B (4)
SB: U' R' U M' U R (6)
CMLL: U F R U R' U' R U R' U' F' R U R' U' R' F R2 U' R' U' R U R' F' (26 (Wow, only if I knew full CMLL...))
LSE: U' M' U M U M' U M' U M' U2 M' U' M2 U' M' U2 M' (18 ( :( ))
54 STM 14.61 PB 3.6 TPS
(Then a few solves later... 10.43...( 5.1 TPS))
Click to expand...

wow, this was a crazyily easy scramble, after i did it, i saw i did the first two blocks and LSE almost just like you did.

scramble: U2 L' R U B2 F L D L' D' F' D' R2 F U2 F2 D B2 L2 U R U' B U' R

FB: u R U' B (4)
SB: U' r' U r R U R' (7)
CMLL: U l' U2 y R U2 R' U2 R' y U2 R (10)
LSE: U' M' U' M' U' M U' M U' M' U2 M U' M2 U' M U2 M (18)

move count: 39; 10.53 seconds; 3.7 moves/second
 
Hi, I'm learning CN Roux (since I've read that the best time to learn CN is when you are just starting).
But then, I read about non-matching 2nd block.

- I'd like to ask, is non-matching 2nd block feasible for a speedsolve?
Like, can you see a sub10 user going full CN non-matching 2nd block?
Fully neutral, not just opposite neutral (I think I've seen on the forums somewhere a vid of one of the top roux users doing a 13 second opposite-neutral NMCMLL solve... Kirjava I think)

- What problems do you expect to encounter?
I know [NM]CMLL recognition would be a pain (as if plain old CMLL recognition wasn't hard enough).
What about EO recognition?

Also, from what I've seen of top roux solvers, it seems their LSE is near instant.
Watching it was like "ok, block, block, CMLL, whoa tons of MU for like less than a second and its done".
Would you imagine that this "advantage" would be lost due to the recognition problems? (that is, if there is an EO problem)

And also, I really haven't checked (or read it through more carefully), but are there more cases/algs to memorize?
I'd expect it to be 4x more, but I'm not really sure.

With all the problems, is speedsolving with fully neutral NM blocks on the "theoretically possible, but apparently not really doable" level (ala ZBLL memo and recog)?

- Let's say, what the hell, I just go for it.
Any tips or helpful resources?
Like, it seems there is already a NMCMLL recog system developed (the ones in the NMCMLL linked in Gilles' pages)?
Also, I remember reading something about "hyperorientations"? (is it the same thing as the one above?)
Anything else you could recommend?

---

Also, as for why I am doing this, it's not for speed (since I read that it saves like, an average of less than 2 moves?)
I just want to be as color neutral as possible.
I thought, if I wanna go neutral, I'd really want to be completely neutral, all the way, and not just switch later.
And also seems like fun.
 
Last edited:
- non-matching block feasible?
I think it is, but only if you're really good at NMCLL recog, or hyperoreintation

- problems?
like you said, NMCLL is pretty tough, but the EO isn't TOO bad. Considering that only one of the edges won't appear to immediately follow the good edge/bad edge rule, and it will happen to be the last L colour edge, I can't see it being that hard, with practice of course.

There aren't more cases, they can just appear in different ways; for instance, you could have the regular Sune alg with the solved top corner being EITHER of your bottom colours, whereas in a matching-2nd block, those corners would be the same.

While my knee-jerk response to "is it doable" is no, if I remember correctly, people were skeptical about being CN for CFOP for a long while, so who knows?

http://home.comcast.net/~quadricode/hyperorientations/
 
Opposite blocks are easy. I often do them in speedsolves these days.

I think if you got used to adjacent blocks and EO that it would be easy too. It's just a case of getting used to looking for UL/UR stickers for detecting EO instead of only U/D stickers.

It's absolutely doable.
 
bundat said:
Like, can you see a sub10 user going full CN non-matching 2nd block?
Click to expand...
I don't think there is a sub10 NM block Roux solver who is already sub10.

bundat said:
I know [NM]CMLL recognition would be a pain (as if plain old CMLL recognition wasn't hard enough).
Click to expand...
recognition isn't much harder than CMLL recognition is.

bundat said:
Would you imagine that this "advantage" would be lost due to the recognition problems? (that is, if there is an EO problem)
Click to expand...
I don't think EO would be a problem. You just have to track the UR edge during corners. EO recognition for the other edges is about the same.

bundat said:
And also, I really haven't checked (or read it through more carefully), but are there more cases/algs to memorize?
I'd expect it to be 4x more, but I'm not really sure.
Click to expand...
Algs are the same, there are just more cases (~85).

bundat said:
With all the problems, is speedsolving with fully neutral NM blocks on the "theoretically possible, but apparently not really doable" level (ala ZBLL memo and recog)?
Click to expand...
It is definitely possible and doable

bundat said:
Any tips or helpful resources?
Like, it seems there is already a NMCMLL recog system developed (the ones in the NMCMLL linked in Gilles' pages)?
Click to expand...
This is the page linked from the wiki: https://sites.google.com/site/athefre/roux
To begin with, you should learn a 2look system first.


What you should consider before learning NMCMLL:
The only advantage with NMCMLL in Roux is more flexibility during second block. However, in a speedsolve you have to be very fast => You don't have time to look for the best block solution.
If you are looking at reconstructions of fast people, there often is an easy 1move pair to build, but they don't see it because they just looked for other pieces. With NM blocks there are more pairs to build, but you aren't able to track more pieces during FB.
=> I don't think learning the NMCMLL system for speedsolves is very useful. You can better spend the time to learn multiple CMLLs ore LSE tricks...
For fewest moves you have enough time to check the case without knowing the system anyway.

But maybe you are able to be very fast with it and can convince me that NM blocks is great. So just go for it ;)
 
Thanks for the replies.
I've been doing it today (with a NMCMLL cheat sheet), and it really does feel doable.
It feels really weird that many different cases use the same alg (and surprisingly, most I've come across are actually OLL algs I already know o_O).
I'm still confused how to memo this tho...I'm too used to fridrich where everything is easy to recog, and you just drill the alg

At first I tried opposite NMB (but not with the guide, but instead intuitively) and I realized I could do it as fast as matching blocks (at least, using 2-look CMLL), just had to recognize opposite colors instead during orientation. (I imagine this wouldn't be as simple with 1-look's recog).

I tried doing adjacent intuitively, and it was nowhere near as simple.

EO was actually good, in that it actually taught me to track the UR edge, and made my transition into next substep (UL/UR edges) a bit smoother.

I think I found an error in the NMCMLL one at Google Sites, couldn't get E1-1 to work. Checked it a bit, finally got it to work with a U2 setup. The alg at the other site (at 110mb) works though.
EDIT: Its fixed now.
EDIT: I must've been just lucky earlier... now I'm running into errors a bit often >_<... switched to dbeyer's nmcmll... and still running into errors :( A bit frustrating... was hoping to just work on blocks and L6E while using a cheat sheet like I used to do with CFOP's LL back when I was just starting out, but ending up with a messed up cube/case is annoying. Also wish they put in the final AUF needed, to make setting up cases (by doing the algs backwards) for practice much easier.

If you are looking at reconstructions of fast people, there often is an easy 1move pair to build, but they don't see it because they just looked for other pieces.
Click to expand...
This is part of the reason I want to be fully neutral.

I used to be Y/W cross CFOP, and I would find myself so focused on the cross edge pieces (and a pair if the cross was easy enough), it was like I was blind to the rest of the cube. And also found myself so focused on pairs during F2L that I couldn't see any other possible configurations.

Went CN, now I'm seeing xcrosses everywhere, and can easily spot much shorter FreeFOP alternatives during F2L. Not so much for speed reasons though (my times actually suffered lol, probably due to lack of experience with these techniques), it just felt like I was seeing the whole cube for the first time, and I was just peeking through a keyhole in the past, and it is so much more fun.

I'm hoping being fully neutral with NMB will change my perspective enough that I might easily spot these "1 move pairs", due to said "changed perspective". And if not, well it's still fun. :)

Anyway, thanks again. :)
 
Last edited:
You must log in or register to reply here.

Similar threads

pcwiz
[Help Thread] CFOP Discussion
144 145 146
Replies
2,918
Views
770,520
abunickabhi
abunickabhi
E
  • Poll Poll
Roux CN discussion (x2/y vs full CN)
2
Replies
22
Views
3,812
abunickabhi
abunickabhi
C
[Free App] Cube Trainer — CFOP + Roux algorithm trainer with timer
Replies
0
Views
116
cellonty
C
T
New method idea: Using Pinkie Pie roux variant with CFOP instead
Replies
4
Views
408
theail03
T
Roux
Should i become full CN with Roux?
Replies
4
Views
533
Roux
Roux
Back
Top