[Help Thread] Roux Discussion and Help

[abomb2o11]

Another progress report because why not.
I have finally gotten a sub-minute average of 12, and only the last two were over a minute, and they were just 1:01.xx.
My f2b take about 20 seconds which I don't think is too bad for me at the current moment.
Been procrastinating learning the rest of the 2-look algs for cmll, also intimidated by all the full cmll algs to learn.
My fixing bad edges and completing the "walls", whatever that step is called are falling behind.
EO is good, figured out most of the cases by myself.

 

[waffle=ijm]

completing the "walls"

TIL.

also, don't bother learning full CMLL until much much much later. Just do two look for now. 20 for F2B for you is okay...which means the bulk of your solves is coming from LSE in general. And since you mentioned you haven't haven't learning the 2-Look, probably from here.

I'd remedy the 2-Look CMLL and then take your time actually understanding LSE.

 

[markfiend]

Just hit a new PB of 36.22 woooo!

A piece of advice for other Roux noobs: I have been learning one-look CMLL in a kind of piecemeal fashion: I have two-look CMLL completely memorised and now I'm gradually learning other CMLL algs.

Does anyone know any statistical analyses of which corner cases are more likely to crop up? I haven't been keeping count but my impression is that sune and antisune crop up most, followed by the cases with two correctly oriented corners (D, E, F), then the rest (A, G, H)

 

[JediJupiter]

Just hit a new PB of 36.22 woooo!

A piece of advice for other Roux noobs: I have been learning one-look CMLL in a kind of piecemeal fashion: I have two-look CMLL completely memorised and now I'm gradually learning other CMLL algs.

Does anyone know any statistical analyses of which corner cases are more likely to crop up? I haven't been keeping count but my impression is that sune and antisune crop up most, followed by the cases with two correctly oriented corners (D, E, F), then the rest (A, G, H)

I was under the impression they were all equally likely.
Congrats on the pb though!

 

[PenguinsDontFly]

I was under the impression they were all equally likely.
Congrats on the pb though!

The H/double headlights cases are the most probable, because they are all U'/U/U2 symmetrical.

 

[JediJupiter]

The H/double headlights cases are the most probable, because they are all U'/U/U2 symmetrical.

Uh, could you explain why that makes them more probable?

 

[PenguinsDontFly]

Uh, could you explain why that makes them more probable?

Usually, each set is 6 algs. H has 6 cases, but 2 are U2 symetric, making them 2/42 probable (each for those 2)

 

[rybaby]

Usually, each set is 6 algs. H has 6 cases, but 2 are U2 symetric, making them 2/42 probable (each for those 2)

H cases are not the most likely: https://www.speedsolving.com/wiki/index.php/CLL_algorithms_(3x3x3)

 

[PenguinsDontFly]

H cases are not the most likely: https://www.speedsolving.com/wiki/index.php/CLL_algorithms_(3x3x3)

I meant that the 2 mirror cases U2 symetric are the most common since they occur 4/48 times. They are the most useful, as you solve 4 cases with 2 algs. it really doesnt matter since they are all worth learning.

 

[rybaby]

I meant that the 2 mirror cases U2 symetric are the most common since they occur 4/48 times. They are the most useful, as you solve 4 cases with 2 algs. it really doesnt matter since they are all worth learning.

The wiki page says 4:162. But yeah, order doesn't matter.

 

[JediJupiter]

I meant that the 2 mirror cases U2 symetric are the most common since they occur 4/48 times. They are the most useful, as you solve 4 cases with 2 algs. it really doesnt matter since they are all worth learning.

Ah right I thought you meant H cases in general too.

 

[JediJupiter]

Yeah! Sub-20 ao20, and a new pb single! After doing more than 20 solves, my average increased to over 20. My first 15 solves are always better than the rest.

 

[PenguinsDontFly]

The wiki page says 4:162. But yeah, order doesn't matter.

Nice sub 8 pb. I still havent beat my pb from november. 9.37 lol. I need a 4s block cmll skip and normal lse and sub 8 single lol.

 

[lerenard]

I'm currently averaging a little over 30 seconds with Roux. My first block is fine (9 moves on average), I use COLL, and I'm working on LSE with slow solves, but the second block is killing me. I already know stuff like only use R, r, M, and U moves, but is there any special order to build it? I've been building the first pair I see and then making a square out of it and an edge followed by my last pair. My biggest issue is finding the pieces I need quickly. How many pieces should I be tracking at this point in order to consistently sub30?

 

[TDM]

I'm currently averaging a little over 30 seconds with Roux. My first block is fine (9 moves on average), I use COLL, and I'm working on LSE with slow solves, but the second block is killing me. I already know stuff like only use R, r, M, and U moves, but is there any special order to build it? I've been building the first pair I see and then making a square out of it and an edge followed by my last pair. My biggest issue is finding the pieces I need quickly. How many pieces should I be tracking at this point in order to consistently sub30?

This is how I make the first block too. I found that it helped to do lots of slow turning solves, both timed and untimed. It only made my times go up a bit at first, but as I learned to look ahead better my times dropped from 19 to 16 in a couple of months, which is fast progress for me (I don't improve very quickly ). You can gradually improve your TPS until you can turn reasonably fast during your second block and still not need to pause.

It could also be useful to learn a couple of easy CMLLs e.g. F R U R' U' F'.

 

[lerenard]

This is how I make the first block too. I found that it helped to do lots of slow turning solves, both timed and untimed. It only made my times go up a bit at first, but as I learned to look ahead better my times dropped from 19 to 16 in a couple of months, which is fast progress for me (I don't improve very quickly ). You can gradually improve your TPS until you can turn reasonably fast during your second block and still not need to pause.

It could also be useful to learn a couple of easy CMLLs e.g. F R U R' U' F'.

I'll definitely work on that. I think my roux knowledge is still pretty low so I need to focus much more on slow solves. I already use that CMLL as well as F R U R' U' R U R' U' F'

My experience with Petrus/CFOP was that as I got more comfortable with the different cases I naturally just didn't need to watch what I was doing and so I was looking ahead, so as I get more used to roux the same thing should happen.

Doing an ao50, I found that my first block sucks with only 15 seconds of inspection :/ I'm not used to restricting myself to pieces that all share a color for my block building, and so I usually can only get my 2x2x1 with inspection, and then I have to look around for my other two pieces. Guess I just need a lot more practice.

 

[TDM]

Doing an ao50, I found that my first block sucks with only 15 seconds of inspection :/ I'm not used to restricting myself to pieces that all share a color for my block building, and so I usually can only get my 2x2x1 with inspection, and then I have to look around for my other two pieces. Guess I just need a lot more practice.

I'm not very good at planning my first block either; I can only plan the 2x2. But I don't practise my blocks much (even though I should); it could be a good idea to spend more than 15 seconds inspecting and trying to plan out the full 2x3, and practise to try to reduce the time needed until you can get it sub-15.

 

[Notasurgeon]

Hi, quick question: At the end of Roux I keep ending up (five times in a row now) with the centers on the M-slice out of place (by 90 degrees) along with exactly two edge blocks in each other's place (also in the M slice, meaning that L and R are finished). If I do (M2 U2 M2 U2) to fix them I just end up flipping the two in the bottom slice and therefore being in the exact same position only now upside down. None of the Roux guides I've read seem to address this situation and I can't figure out how to fix it on my own. Any ideas?

The only way I could figure out how to fix it was to align the centers and use a 3-cycle to flip the three in the wrong location (M' U2 M U2), which left me with all four edges in the correct place but all flipped backwards, which I fixed with two applications of an edge-flipping algorithm that I know (M' U M' U M' U2, M U M U M U2). But that's not Roux.

 

[PenguinsDontFly]

Hi, quick question: At the end of Roux I keep ending up (five times in a row now) with the centers on the M-slice out of place (by 90 degrees) along with exactly two edge blocks in each other's place (also in the M slice, meaning that L and R are finished). If I do (M2 U2 M2 U2) to fix them I just end up flipping the two in the bottom slice and therefore being in the exact same position only now upside down. None of the Roux guides I've read seem to address this situation and I can't figure out how to fix it on my own. Any ideas?

Picture please?

 

[primarycuber]

Pretty average splits for me except that 20.69 which is a pb by like 6 seconds... I don't really know what happened, I'm still in shock
What do you think, does this look ok, or are there are any too slow / too fast stages compared to each other?

[table="width: 500, class: grid"]
[tr]
[td][/td]
[td]time[/td]
[td]FB[/td]
[td]SB[/td]
[td]corners[/td]
[td]lse[/td]
[/tr]
[tr]
[td]1[/td]
[td]29.84[/td]
[td]4.26[/td]
[td]10.32[/td]
[td]5.03[/td]
[td]10.24[/td]
[/tr]
[tr]
[td]2[/td]
[td]33.69[/td]
[td]7.97[/td]
[td]12.69[/td]
[td]5.29[/td]
[td]7.74[/td]
[/tr]
[tr]
[td]3[/td]
[td]35.36[/td]
[td]8.59[/td]
[td]14.51[/td]
[td]2.82[/td]
[td]9.44[/td]
[/tr]
[tr]
[td]4[/td]
[td]31.02[/td]
[td]11.87[/td]
[td]5.62[/td]
[td]6.60[/td]
[td]6.93[/td]
[/tr]
[tr]
[td]5[/td]
[td]20.69[/td]
[td]3.81[/td]
[td]10.00[/td]
[td]2.02[/td]
[td]4.86[/td]
[/tr]
[tr]
[td]6[/td]
[td]35.33[/td]
[td]9.27[/td]
[td]10.83[/td]
[td]7.26[/td]
[td]7.96[/td]
[/tr]
[tr]
[td]7[/td]
[td]38.08[/td]
[td]6.66[/td]
[td]16.39[/td]
[td]3.39[/td]
[td]8.64[/td]
[/tr]
[tr]
[td]8[/td]
[td]36.16[/td]
[td]6.38[/td]
[td]13.77[/td]
[td]6.93[/td]
[td]9.07[/td]
[/tr]
[tr]
[td]9[/td]
[td]33.72[/td]
[td]5.69[/td]
[td]10.45[/td]
[td]8.90[/td]
[td]8.68[/td]
[/tr]
[tr]
[td]10[/td]
[td]33.62[/td]
[td]6.99[/td]
[td]10.96[/td]
[td]7.77[/td]
[td]7.90[/td]
[/tr]
[tr]
[td]11[/td]
[td]36.19[/td]
[td]9.13[/td]
[td]13.45[/td]
[td]5.87[/td]
[td]7.74[/td]
[/tr]
[tr]
[td]12[/td]
[td]28.25[/td]
[td]4.99[/td]
[td]8.71[/td]
[td]7.12[/td]
[td]7.42[/td]
[/tr]
[tr]
[td]mean[/td]
[td]32.66[/td]
[td]7.13[/td]
[td]11.48[/td]
[td]6.00[/td]
[td]8.05[/td]
[/tr]
[/table]