• Welcome to the Speedsolving.com, home of the web's largest puzzle community!
    You are currently viewing our forum as a guest which gives you limited access to join discussions and access our other features.

    Registration is fast, simple and absolutely free so please, join our community of 40,000+ people from around the world today!

    If you are already a member, simply login to hide this message and begin participating in the community!

[ER] Alexander Lau - 7.37 3x3x3 avg

Escher

Babby
Joined
Jul 23, 2008
Messages
3,374
WCA
2008KINN01
YouTube
Visit Channel
Alex <3

The only reason I learned lots of things for 2x2 is because fast people used it AND I saw potential in how the method worked. Plus the puzzle is really simple so it doesn't take much to know how useful an approach will be.

Roux is a beautiful method through how it divides stages of the solution. CLL+L6E is pretty much the best approach there is for algorithmically but intuitively solving a final set of pieces for the cube. CLL when used on a cube with F2L solved is obviously less efficient than allowing for <M,U> disruption (or manipulation), and manipulating the RB for a decent CLL case isn't too difficult. Roux's L6E approach is human-optimal, in the sense that optimal solutions done by computation are almost always move-cancelled version of the Roux approach, and both generation and retracing of these solutions are fairly easily done. The efficiency of F2B as an approach to set up CLL+L6E is obvious, and are equally well broken down. If there was a 'perfect' method for humans Roux is far closer than CFOP is to that end.

CFOP inherently has a lower skill-ceiling because of how the cross and then F2L create restrictions - the trade-off between simplicity and efficiency is just less well done in CFOP. LL as an approach is also limited in that it's pretty much inherently non-intuitive and progression is capped by 'how many of these thousands of situations can you memorise?', either through experience and deducing PLL at the beginning of/during OLL or through raw memo and repetition. Expansion of f2l is equally difficult - it takes a lot of thought and experience to get a 27 move average f2l in a slow solve, let alone speedsolve. CFOP works well because the trade-off towards simplicity allows fast progression, somewhat simple lookahead, and rewards reactions over decision making.

Feliks overcomes plenty of CFOPs weaknesses by abusing his colour neutrality and the fact we allow inspection time. By guaranteeing an efficient/ergonomic start and reaching the end of first pair without having to make many decisions, a lot of the awkwardness of cross and FP is skipped. After enough experience, it is relatively simple to know which of the final 3 pairs one should solve first to improve the 3rd/4th pair cases, and a little manipulation of the 3rd pair usually creates an easy 4th pair -> LL transition, at which point reaction times + practise creates competitive edge.

Basically I believe Roux > CFOP as a general method, unless you are CN, in which case speedcubing is not developed enough to distinguish between the two, where CFOPs weaknesses are bandaged by raw development and Roux's resources are limited in terms of competitive pool.

Unless you think about speedsolving solutions' methodology a lot sometimes all it can take is a display of skill from someone who is VERY VERY good at the alternative to see these truths without working through them.
 

DeeDubb

Member
Joined
Mar 9, 2014
Messages
1,432
Location
South Korea
WCA
2014WHIT07
YouTube
Visit Channel
Alex <3

The only reason I learned lots of things for 2x2 is because fast people used it AND I saw potential in how the method worked. Plus the puzzle is really simple so it doesn't take much to know how useful an approach will be.

Roux is a beautiful method through how it divides stages of the solution. CLL+L6E is pretty much the best approach there is for algorithmically but intuitively solving a final set of pieces for the cube. CLL when used on a cube with F2L solved is obviously less efficient than allowing for <M,U> disruption (or manipulation), and manipulating the RB for a decent CLL case isn't too difficult. Roux's L6E approach is human-optimal, in the sense that optimal solutions done by computation are almost always move-cancelled version of the Roux approach, and both generation and retracing of these solutions are fairly easily done. The efficiency of F2B as an approach to set up CLL+L6E is obvious, and are equally well broken down. If there was a 'perfect' method for humans Roux is far closer than CFOP is to that end.

CFOP inherently has a lower skill-ceiling because of how the cross and then F2L create restrictions - the trade-off between simplicity and efficiency is just less well done in CFOP. LL as an approach is also limited in that it's pretty much inherently non-intuitive and progression is capped by 'how many of these thousands of situations can you memorise?', either through experience and deducing PLL at the beginning of/during OLL or through raw memo and repetition. Expansion of f2l is equally difficult - it takes a lot of thought and experience to get a 27 move average f2l in a slow solve, let alone speedsolve. CFOP works well because the trade-off towards simplicity allows fast progression, somewhat simple lookahead, and rewards reactions over decision making.

Feliks overcomes plenty of CFOPs weaknesses by abusing his colour neutrality and the fact we allow inspection time. By guaranteeing an efficient/ergonomic start and reaching the end of first pair without having to make many decisions, a lot of the awkwardness of cross and FP is skipped. After enough experience, it is relatively simple to know which of the final 3 pairs one should solve first to improve the 3rd/4th pair cases, and a little manipulation of the 3rd pair usually creates an easy 4th pair -> LL transition, at which point reaction times + practise creates competitive edge.

Basically I believe Roux > CFOP as a general method, unless you are CN, in which case speedcubing is not developed enough to distinguish between the two, where CFOPs weaknesses are bandaged by raw development and Roux's resources are limited in terms of competitive pool.

Unless you think about speedsolving solutions' methodology a lot sometimes all it can take is a display of skill from someone who is VERY VERY good at the alternative to see these truths without working through them.

Very well thought out interpretation. I agree with pretty much everything said here.

One thing to add is that may be holding most Rouxers back (certainly I feel like it's the case for me) is I believe that Roux requires a lot more practice, especially with L6E. L6E leads to a ridiculously high number of different cases compared to OLL/PLL, and very few people can seemlessly flow through at a TPS/efficiency rate that can compare to OLL/PLL. I think L6E separates average Roux users from great ones.
 

biscuit

Member
Joined
Sep 8, 2014
Messages
1,811
Location
Kansas City
WCA
2015WEBS01
YouTube
Visit Channel
I'm not saying that Feliks should switch to Roux. But I do think that newer cubers should switch. How long has CFOP been around? 20 years? How long has Roux been around as a real speed solving method? 3-4 years TOPS? Once enough people switch and become fast I have no doubt that it will be a better method. Sure there will be fast CFOP solvers but many people have said that Roux is a "more pure" and "better" method. Maybe Alex is at it's cap. It's possible but I don't think so.
 

GuRoux

Member
Joined
May 6, 2013
Messages
1,712
Location
San Diego, California
WCA
2014TANG03
YouTube
Visit Channel
I'm not saying that Feliks should switch to Roux. But I do think that newer cubers should switch. How long has CFOP been around? 20 years? How long has Roux been around as a real speed solving method? 3-4 years TOPS? Once enough people switch and become fast I have no doubt that it will be a better method. Sure there will be fast CFOP solvers but many people have said that Roux is a "more pure" and "better" method. Maybe Alex is at it's cap. It's possible but I don't think so.

i think roux is around 8-10 years, not really sure.
 

Myachii

Member
Joined
May 24, 2014
Messages
884
Location
England
WCA
2013MORR03
YouTube
Visit Channel
It doesn't matter which method is better, because as we can see you can get amazing times with both Roux and CFOP and many other methods (anyone know how many non-roux/cfop solvers are in the top 100?)

To be honest I think this is a huge hit for MoYu. Not only is the 3x3 WR Single set with a ZhanChi, it has also stood for 772 days (since March 2013). If my research is correct, the MoYu Huanying was the first speedcube released by the MoYu division of YJ. Since then, the only record broken with a MoYu puzzle has been the 3x3 Average.

I think it's kinda ironic how nearly everyone (including myself) is using MoYu's 3x3's as their main puzzle, yet it hasn't yet managed to defeat the noble Zhanchi in terms of WR Single status, and a Guhong v2 has just gained 2nd in the world average for 3x3.

This amazing achievement by Alex in my opinion is more likely to turn people back to their Zhanchi's to see if they can get any decent times on them than it is to convert people to the Roux method.
 

TDM

Member
Joined
Mar 7, 2013
Messages
7,006
Location
Oxfordshire, UK
WCA
2013MEND03
YouTube
Visit Channel
To be honest I think this is a huge hit for MoYu. Not only is the 3x3 WR Single set with a ZhanChi, it has also stood for 772 days (since March 2013). If my research is correct, the MoYu Huanying was the first speedcube released by the MoYu division of YJ. Since then, the only record broken with a MoYu puzzle has been the 3x3 Average.
In addition, Feliks has switched his main to a Gans, so if he does break any WRs in the future, it probably won't be with a MoYu cube, until they release something better.
 

Tim Major

Platinum Member
Joined
Aug 26, 2009
Messages
5,381
Location
Melbourne, Australia
WCA
2010MAJO01
i think roux is around 8-10 years, not really sure.

But it was mostly irrelevant as a SPEED method. When I started it was pretty much universally thought that CFOP was the best current speedsolving method. Roux was on the same level as Petrus. Biggreen was the only fast Roux solver and he wasn't competitive with WR times. Only in the last few years it's been getting suggested to beginners alongside CFOP, so it likely means that a top 100 with more Roux solvers will just take time, not further innovation.

CFOP inherently has a lower skill-ceiling because of how the cross and then F2L create restrictions - the trade-off between simplicity and efficiency is just less well done in CFOP.
I disagree, I think the tradeoff is done differently. CFOP has different periods of restriction, many of which are superior to Roux. As fast as Alex is at <MU>, TPS simply doesn't compare to CFOP's move restrictions <RUF> for example. CFOP is less efficient (assuming a high level Roux solver), but also less complex/adlib. CFOP solves are so similar that there is practically no thinking time. This allows much higher TPS than Roux.

Feliks overcomes plenty of CFOPs weaknesses by abusing his colour neutrality and the fact we allow inspection time. By guaranteeing an efficient/ergonomic start and reaching the end of first pair without having to make many decisions, a lot of the awkwardness of cross and FP is skipped. After enough experience, it is relatively simple to know which of the final 3 pairs one should solve first to improve the 3rd/4th pair cases, and a little manipulation of the 3rd pair usually creates an easy 4th pair -> LL transition, at which point reaction times + practise creates competitive edge.

Basically I believe Roux > CFOP as a general method, unless you are CN, in which case speedcubing is not developed enough to distinguish between the two

Expand on this please, I think you are overstating CN's advantage. You don't need colour neutrality to inspect cross+1. Also inspection time is far more influential on Roux than CFOP.

The parts I didn't quote I don't necessarily have anything to say about, but you make some assumptions/broad statementd that you imply are fact. I'm not sure this method advantage is nearly as clear as you think
 
Last edited:

mark49152

Premium Member
Joined
Oct 29, 2012
Messages
4,719
Location
UK
WCA
2015RIVE05
YouTube
Visit Channel
One thing to add is that may be holding most Rouxers back (certainly I feel like it's the case for me) is I believe that Roux requires a lot more practice, especially with L6E. L6E leads to a ridiculously high number of different cases compared to OLL/PLL, and very few people can seemlessly flow through at a TPS/efficiency rate that can compare to OLL/PLL. I think L6E separates average Roux users from great ones.
Couldn't the same be said about block-building versus CFOP F2L?
 

mark49152

Premium Member
Joined
Oct 29, 2012
Messages
4,719
Location
UK
WCA
2015RIVE05
YouTube
Visit Channel
Let me try again by stealing DeeDub's words.

Roux requires a lot more practice, especially with block-building. Block-building leads to a ridiculously high number of different cases compared to CFOP F2L, and very few people can seamlessly flow through at a TPS/efficiency rate that can compare to CFOP/F2L. Block-building separates average Roux users from great ones.
 
Top