[Help Thread] Roux Discussion and Help

[GuRoux]

How often do other x2/y (or y2/x) people one look first block? I do about 60% of the time, and do it 100% of the time when I keep myself from freaking out with "only" 8 seconds of inspection left.

with y2 i do it 100% of the time. 75% of the time i track RD.

 

[rybaby]

with y2 i do it 100% of the time. 75% of the time i track RD.

Do you trace RD from the start? I have mostly just been slowing down a bit on the second half of FB and paying attention to pieces not in the FB. You can probably see this strategy a bit in the ao5 I posted on Sunday (not sure how well I transitioned in those solves though).

 

[GuRoux]

Do you trace RD from the start? I have mostly just been slowing down a bit on the second half of FB and paying attention to pieces not in the FB. You can probably see this strategy a bit in the ao5 I posted on Sunday (not sure how well I transitioned in those solves though).

probably 50% of the time i know where RD will be before i start, other times i just track it.

 

[tpt8899]

Would it be a bad idea for me (around 40-55 seconds) to (attempt) to be full CN?

 

[GuRoux]

Would it be a bad idea for me (around 40-55 seconds) to (attempt) to be full CN?

if there is anytime to do it it is now, but it is agreed among the roux community that CN is way can never be utilized to the full extent since you can't look at 24 blocks in 15 seconds. i think about anywhere between 2 and 8 starting blocks is standard. 8 probably is the best.

 

[rowan]

if there is anytime to do it it is now, but it is agreed among the roux community that CN is way can never be utilized to the full extent since you can't look at 24 blocks in 15 seconds. i think about anywhere between 2 and 8 starting blocks is standard. 8 probably is the best.

I agree that you cannot inspect 24 blocks in 15 seconds, I do not do this. Just to clear up a little bit on how color neutral users use their inspection, because I don't think it's well understood: About 2/3 of my solves have at least one pre-made 1x1x2 block in them, if this pair isn't trash, I base my block off it, which reduces the blocks I'm looking at from 24 to 4. If the pair is useless, I can figure this out easily enough and move to another block with either a second pre-made pair or an easily built pair and once again I'm looking at only four potential first blocks at any given time, I think the most I'll ever consider in inspection is eight. Maybe if I have an awful case and I'm struggling to decide, I'll have looked at twelve blocks. From here I'm inspecting how I want to extend the pair, this is the primary advantage I see in being color neutral because I can pick the most optimal way to build on this pair instead of being limited by a bottom color, I always go for the FB with the fewest moves.

In the other 1/3 of solves, I usually look for one move 1x1x2 blocks, which happens a majority of the time, and apply the same process I normally do, so on and so forth from there. I'm doing the math on these breakdowns (it's a little difficult to do by hand), but so far I'm finding about 80-90% of cases have prebuilt 1x1x2 blocks or 1-move 1x1x2 blocks. If anyone is interested enough to figure this out with some help from a computer, that'd be much appreciated. So for color neutral users, some bad cases can be especially bad, but there are good cases we can take advantage of in unique ways, and the vast majority fall in this category.

Do you guys really inspect all eight of the possible blocks you can build? Or do you realize pretty quickly which blocks look nicest and optimize how you're going to build it? Because that's what I do, but instead I can expand the 1x1x2 or 1x2x2 in any direction, being able to place any pair with a 1x2x2 I think is one of the bigger advantages here, that only full color neutrality can grant.

 

[GuRoux]

I agree that you cannot inspect 24 blocks in 15 seconds, I do not do this. Just to clear up a little bit on how color neutral users use their inspection, because I don't think it's well understood: About 2/3 of my solves have at least one pre-made 1x1x2 block in them, if this pair isn't trash, I base my block off it, which reduces the blocks I'm looking at from 24 to 4. If the pair is useless, I can figure this out easily enough and move to another block with either a second pre-made pair or an easily built pair and once again I'm looking at only four potential first blocks at any given time, I think the most I'll ever consider in inspection is eight. Maybe if I have an awful case and I'm struggling to decide, I'll have looked at twelve blocks. From here I'm inspecting how I want to extend the pair, this is the primary advantage I see in being color neutral because I can pick the most optimal way to build on this pair instead of being limited by a bottom color, I always go for the FB with the fewest moves.

In the other 1/3 of solves, I usually look for one move 1x1x2 blocks, which happens a majority of the time, and apply the same process I normally do, so on and so forth from there. I'm doing the math on these breakdowns (it's a little difficult to do by hand), but so far I'm finding about 80-90% of cases have prebuilt 1x1x2 blocks or 1-move 1x1x2 blocks. If anyone is interested enough to figure this out with some help from a computer, that'd be much appreciated. So for color neutral users, some bad cases can be especially bad, but there are good cases we can take advantage of in unique ways, and the vast majority fall in this category.

Do you guys really inspect all eight of the possible blocks you can build? Or do you realize pretty quickly which blocks look nicest and optimize how you're going to build it? Because that's what I do, but instead I can expand the 1x1x2 or 1x2x2 in any direction, being able to place any pair with a 1x2x2 I think is one of the bigger advantages here, that only full color neutrality can grant.

you're right, inspection isn't really looking at every possible block you can make. Even I, with two possible starting blocks, don't always look at both. if you are y x2, you can use any 1x1x2 block you find in multiple blocks. compared to y x2, full CN barely give you any advantage in first block. But it does make getting better at the rest of the sove harder.

 

[tpt8899]

Thanks for the advice guys (or gals)!

 

[rowan]

if you are y x2, you can use any 1x1x2 block you find in multiple blocks.

This is not true. Half the time they can only use it in one block, the other half the time they can use it in two blocks. Even then a color neutral user can expand upon the 1x1x2 in four ways, which increases the likelihood that the block will be beneficial for them to use. There are many cases in which a 1x1x2 pair would be a "good" case for a color neutral user but a "bad" case for a x2/y color neutral user. I can make four blocks from each pair, a x2/y color neutral user cannot. Do you see my point here?

But it does make getting better at the rest of the sove harder.

I still have no idea why people think they can make this claim. I have yet to see any evidence. Just because when someone who isn't color neutral does color neutral solves and their recognition is off, doesn't mean that someone who has been color neutral from the start has poor recognition.

 

[GuRoux]

This is not true. Half the time they can only use it in one block, the other half the time they can use it in two blocks. Even then a color neutral user can expand upon the 1x1x2 in four ways, which increases the likelihood that the block will be beneficial for them to use. There are many cases in which a 1x1x2 pair would be a "good" case for a color neutral user but a "bad" case for a x2/y color neutral user. I can make four blocks from each pair, a x2/y color neutral user cannot. Do you see my point here?



I still have no idea why people think they can make this claim. I have yet to see any evidence. Just because when someone who isn't color neutral does color neutral solves and their recognition is off, doesn't mean that someone who has been color neutral from the start has poor recognition.

Nevermind, about the always multiple ways to use a 1x1x2 block, you're right. But on an average scrambled cube there is usually at least one corer edge pair and multiple edge center pairs already build. all of which a xy2 solve ca use and some of which can go into different blocks. already there are more good options than one can consider. According to this: https://www.speedsolving.com/forum/...2x2-move-count&p=185213&viewfull=1#post185213, there is an average movecount of 6.7 for a fixed first block. xy2 probably brings it down to under 5.4 compared to CN 5.1. People claim thta recognition is harder for a CN usually from experience. Especially in the beginning, it isn't too hard to just memorize one fixed color scheme and improve. Even from transfering from fixed to y2 neutrality i felt loss of recognition, took about a week. When you transfer towards CN, you have to get use to all the different colors in all parts of the solve. You have to get familiar with all of them. That's much harder for a beginner to learn and initially improve. After that i don't really know how much harder it is to improve. But it must be harder.

 

[rowan]

Nevermind, about the always multiple ways to use a 1x1x2 block, you're right. But on an average scrambled cube there is usually at least one corer edge pair and multiple edge center pairs already build. all of which a xy2 solve ca use and some of which can go into different blocks. already there are more good options than one can consider. According to this: https://www.speedsolving.com/forum/...2x2-move-count&p=185213&viewfull=1#post185213, there is an average movecount of 6.7 for a fixed first block. xy2 probably brings it down to under 5.4 compared to CN 5.1. People claim thta recognition is harder for a CN usually from experience. Especially in the beginning, it isn't too hard to just memorize one fixed color scheme and improve. Even from transfering from fixed to y2 neutrality i felt loss of recognition, took about a week. When you transfer towards CN, you have to get use to all the different colors in all parts of the solve. You have to get familiar with all of them. That's much harder for a beginner to learn and initially improve. After that i don't really know how much harder it is to improve. But it must be harder.

Yeah, I don't think there is enough advantage to warrant switching, but there is an advantage to being color neutral however slight it may be, so that's why I chose it. Also, I think it's important to note this cannot be overcome by practice whereas any supposed recognition disadvantage could be. But, I don't really want to entertain the idea of a recognition disadvantage because I disagree that one exists and I have yet to see any non-anecdotal evidence suggesting that it does.

The way that color neutral users and non-color neutral users recognize and visualize their cases I think is fundamentally different, which may explain why everyone seems to think it is so complicated. I never recognize cases within the context of colors; this would clearly not work. Perhaps some color fixed users think of their cases as "this piece will be x color" but I think of cases within relation to my L and D colors on a solve by solve basis. My recognition systems work around patterns, not particular colors.

And, if the argument is that color neutral is bad because of recognition issues, I assume that being anything other than x2/y2 would similarly effect your recognition systems? I'm assuming that people who use something other than x2/y2 also use a similarly-themed recognition system based on patterns rather than colors specifically.

 

[AirbusCube]

First timed roux solve:59.94 seconds
. I really love roux. I'm going to switch from cfop

 

[rybaby]

14 & 16 = meh
But the rest of the average was decent

 

[GuRoux]

nice solves. try to not look around so much when you do second block. if you find yourself pausing more, try slow turning.

 

[AirbusCube]

I'm sub 45 with roux ( learned yesterday) and I don't know how to improve. Is it just practice? I use f2l blocks because I switched from fridrich

 

[Username]

I'm sub 45 with roux ( learned yesterday) and I don't know how to improve. Is it just practice? I use f2l blocks because I switched from fridrich

Learn to do the f2b in a non cfop way and then its just practice

 

[Cool Frog]

I'm sub 45 with roux ( learned yesterday) and I don't know how to improve. Is it just practice? I use f2l blocks because I switched from fridrich

Not using F2L blocks would be a good start. You want to work on reducing move-count for your F2B, which would mean not using the same techniques you use in CFOP, a lot of those can be really inefficient. Do some slow, untimed solves and plan your blocks out (however long it takes) making them in as few moves as you possibly can. I would aim for less than 10 for first block and less than 20 for second, take your time to figure out how to make this possible. Consider how wide turns vs. slice turns affect your block building. In the beginning it doesn't matter how long you're inspecting, you'll naturally get better at that, your block building however, won't magically improve unless you put in the leg work.

Try to cut out unnecessary AUFs early on. These small moves add up. Also don't adjust your centers all the time, only adjust after second block (with R/r). Try to recog from different angles. Take time to do slow, untimed solves. Efficiency won't develop unless you take time out to really think about your solving.

 

[Berd]

Probably been asked a hundred times before, but the best tutorial for roux? Thanks!

 

[PenguinsDontFly]

Personally, I used donovan's tutorial. Other good ones are Alex lau's (5bld) speed tutorial, waffo's tutorial, and deedubb's tutorial.

 

[TDM]

Personally, I used donovan's tutorial. Other good ones are Alex lau's (5bld) speed tutorial, waffo's tutorial, and deedubb's tutorial.

DeeDubb's is really for complete beginners (people who can't solve a cube), so it's a bit too simple for cubers imo.
I learned from Waffo's, just because it was the only one I knew of at the time