• 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!

Beginner's Guide to Choosing a Speedsolving Method

Raviorez

Member
Joined
Nov 22, 2013
Messages
65
Need help switching 3x3x3 methods

Hello everybody,

I'm in trouble because I'm considering to chance of method but I really don't know if I will chance/want to chance!!
Maybe Roux or ZZ?
Need help to make a choice!

Some stuff about me (may be helpfull to help me):
I average about 35 seconds with CFOP (2 look OLL en 2 look PLL) but I really get stuck with F2L!! But I think that CFOP is a quite good method for me, I really don't know and I'm very confused!!
Also I think I do too many rotations... I'm quite good at learning algorithms but I'm very lazyyy to learn them :/
I really wanna be good at BLD solving (maybe there's a method that helps for BLD solving?).
I really like the big cubes so is there a method that's better for big cubes?

Thanks in advance...
 

pipkiksass

Member
Joined
Jan 1, 2013
Messages
1,081
YouTube
Visit Channel
Maybe Roux or ZZ?
Need help to make a choice!

I'll address each of the following based on these two options vs. CFOP.
I average about 35 seconds with CFOP (2 look OLL en 2 look PLL) but I really get stuck with F2L!! But I think that CFOP is a quite good method for me, I really don't know and I'm very confused!!

Read this thread. If you like CFOP, stick with CFOP. Are you looking at a new method to get faster, or just for interest? A new method won't miraculously make you faster, but moving to 1-look PLL will make a big difference if you decide to stick with CFOP.

Also I think I do too many rotations... I'm quite good at learning algorithms but I'm very lazyyy to learn them :/
Rotations are no MASSIVE problem. I rotate too much and I average 18ish with CFOP. There's a lot of fuss made about rotationless algs, but some of them are awkward and not THAT much quicker than rotating.

In terms of algs, Roux has less algs than CFOP, if that's a deciding factor?

I really like the big cubes so is there a method that's better for big cubes?
No, most people use CFOP, or variations on it in almost all big cube methods. This includes Roux users, because M turns are a large part of Roux, and can't be done on big cubes. I've heard a lot of ZZ users say it's excellent for OH due to the ergonomic nature of the method, and can see how this probably applies to big cubes as well, but only the 3x3 stage after reduction.

In short - CFOP has the most support and resources, due to its large userbase. From what you say, the only reason I can see for switching would be the lower movecount and alg count for Roux. I'd recommend dabbling with Roux and seeing how you like it. :D
 

TDM

Member
Joined
Mar 7, 2013
Messages
7,006
Location
Oxfordshire, UK
WCA
2013MEND03
YouTube
Visit Channel
Both Roux and ZZ have steps similar to F2L. With ZZ, you're more restricted than CFOP, but that doesn't really matter because there's no rotations. In Roux, there are also no rotations, and there's much more freedom than both ZZ and CFOP. Roux has less algs than CFOP, and so does ZZ if you use either OCLL-PLL or COLL-EPLL. If you want to learn blindsolving, learn PLL; PLL will be used in both ZZ and CFOP, but not Roux. You'll have to learn the algs for BLD separately. However, if you use M2 for BLD (which you will do for most of the time until you're very fast), then no method gives you an advantage. As for bigcubes, Roux isn't that good, and nor is ZZ. CFOP is best for bigcubes, imo (but others will disagree with this) Roux is best for 2H, and ZZ is best for OH. Ultimately, you should decide which would be better for your preferable events. Or you can do what some people do and use different methods for different events; until recently I was using CFOP for 2H and ZZ for OH (I now use ZZ for both and CFOP for bigcubes).

If you do use ZZ, try being y-axis neutral. With Roux, be either x-axis neutral or y-axis neutral. Or x2y2. The earlier you start being colour neutral with a method, the easier it will be. I think it's very useful to be y-axis neutral with ZZ, but I'm not sure about which colour neutrality is best with Roux (obviosuly full CN is, but it isn't practical).
 

DeeDubb

Member
Joined
Mar 9, 2014
Messages
1,432
Location
South Korea
WCA
2014WHIT07
YouTube
Visit Channel
Roux having low algorithms isn't necessarily a good thing for speed. Nothing is executed faster than a well-practiced algorithm, so the fact that Roux has less algorithms means less automated parts and more thinking during your solves. Eventually, the steps get easier and more automatic, but early on, the freedom will slow you down a lot.
 

GuRoux

Member
Joined
May 6, 2013
Messages
1,712
Location
San Diego, California
WCA
2014TANG03
YouTube
Visit Channel
Roux having low algorithms isn't necessarily a good thing for speed. Nothing is executed faster than a well-practiced algorithm, so the fact that Roux has less algorithms means less automated parts and more thinking during your solves. Eventually, the steps get easier and more automatic, but early on, the freedom will slow you down a lot.

The plus of using only one alg per solve is that more pieces are solved "intuitively," which equates to better movecounts and that you spend less time recognizing what algs to do. If you really hate high movecounts, rotations, and the restriction of F2L, roux is a good choice; though, it is a method that takes some time to understand well.
 

mDiPalma

Member
Joined
Jul 12, 2011
Messages
1,534
Try Petrus.

It has 50 less algs than CFOP, 20 less algs than Roux, same amount as ZZ, it requires less rotations than CFOP (if you do it right), it has a lower half-turn metric movecount than Roux/ZZ/CFOP, it offers the MOST freedom out of all the methods, it makes more efficient use of inspection time than CFOP/Roux/ZZ, it maximizes the time-impact of apparent "lucky scrambles", it's easier to explain Petrus' blockbuilding approach to non-cubers than CFOP/Roux/ZZ, and it's more fun (which is all that really matters).

Edit:
With regards to BLD solving, Petrus will get you accustomed to the color scheme of the cube MORE than any other method, which will drastically help speed up your memorization phase during BLD solves (when you are looking for which pieces belong in which locations).

With regards to Big Cubes, Petrus will expose you to the effects of good/bad edges MORE than any other method, which will allow you to pair up edge pieces while orienting them for a quicker F2L phase (assuming Yau method is pursued).
 
Last edited:

GuRoux

Member
Joined
May 6, 2013
Messages
1,712
Location
San Diego, California
WCA
2014TANG03
YouTube
Visit Channel
Try Petrus.

It has 50 less algs than CFOP, 20 less algs than Roux, same amount as ZZ, it requires less rotations than CFOP (if you do it right), it has a lower half-turn metric movecount than Roux/ZZ/CFOP, it offers the MOST freedom out of all the methods, it makes more efficient use of inspection time than CFOP/Roux/ZZ, it maximizes the time-impact of apparent "lucky scrambles", it's easier to explain Petrus' blockbuilding approach to non-cubers than CFOP/Roux/ZZ, and it's more fun (which is all that really matters).

Edit:
With regards to BLD solving, Petrus will get you accustomed to the color scheme of the cube MORE than any other method, which will drastically help speed up your memorization phase during BLD solves (when you are looking for which pieces belong in which locations).

With regards to Big Cubes, Petrus will expose you to the effects of good/bad edges MORE than any other method, which will allow you to pair up edge pieces while orienting them for a quicker F2L phase (assuming Yau method is pursued).

how can there only be 20 algs in petrus. I thought it was two look oll and pll, around 30 algs.
 

mDiPalma

Member
Joined
Jul 12, 2011
Messages
1,534
If you count those, you might as well count petrus EO too.

I mean the 4 algs that are used for the permutation of the M slice during L6E.

Roux still has 18 algs more than Petrus. My original number (20) was obviously an approximation.

Don't pull a symmetry/inverse/intuitive argument, because if you do, >13 PLLs and >2 OLLs get deducted from the Petrus alg-count, leaving it with < 13.
 

GuRoux

Member
Joined
May 6, 2013
Messages
1,712
Location
San Diego, California
WCA
2014TANG03
YouTube
Visit Channel
I mean the 4 algs that are used for the permutation of the M slice during L6E.

Roux still has 18 algs more than Petrus. My original number (20) was obviously an approximation.

Don't pull a symmetry/inverse/intuitive argument, because if you do, >13 PLLs and >2 OLLs get deducted from the Petrus alg-count, leaving it with < 13.

I guess you're right then. It's a shame that petrus doesn't have much popularity; I always thought it is just as good as zz, roux, and cfop.
 

rybaby

Member
Joined
May 12, 2013
Messages
548
Location
Texas
WCA
2013PRZY02
YouTube
Visit Channel
You realize that the last layer of the Petrus method is not always OLL+PLL...right? Petrus last layer could be completed in 493 algs, or it could be completed in 2. Roux CMLL could also be done with only 2 algs (Sune+Niklas). I don't think algorithm count should matter much anyway, unless the number is drastic (Like ZBLL for example). Beginners can complete Roux with 2 algorithms, but ideally they would go on to learn CMLL at some point. So I think a beginner can ease themselves into either Petrus OR Roux without being overwhelmed with algorithms. IMO, you (OP with the question) should try both; see which one fits you best.

Oh and about LSE "algorithms," ideally the step is intuitive, meaning you can see how U2 M2 U2 moves pieces as it does. Eventually it's just engrained as an "algorithm," but (according to Kyle Allaire, Petrus user) Petrus' step 4 similarly becomes algorithmic after using intuition to figure it out for yourself. Insertions just become natural like LSE does. fedora
 

GuRoux

Member
Joined
May 6, 2013
Messages
1,712
Location
San Diego, California
WCA
2014TANG03
YouTube
Visit Channel
You realize that the last layer of the Petrus method is not always OLL+PLL...right? Petrus last layer could be completed in 493 algs, or it could be completed in 2. Roux CMLL could also be done with only 2 algs (Sune+Niklas). I don't think algorithm count should matter much anyway, unless the number is drastic (Like ZBLL for example). Beginners can complete Roux with 2 algorithms, but ideally they would go on to learn CMLL at some point. So I think a beginner can ease themselves into either Petrus OR Roux without being overwhelmed with algorithms. IMO, you (OP with the question) should try both; see which one fits you best.

Oh and about LSE "algorithms," ideally the step is intuitive, meaning you can see how U2 M2 U2 moves pieces as it does. Eventually it's just engrained as an "algorithm," but (according to Kyle Allaire, Petrus user) Petrus' step 4 similarly becomes algorithmic after using intuition to figure it out for yourself. Insertions just become natural like LSE does. fedora

I guess for alg number, it's a little shaky with how many substeps or multiple algs you want to learn. Either way, roux only has one step where "long" or "unintuitive" algs are used.
 

rybaby

Member
Joined
May 12, 2013
Messages
548
Location
Texas
WCA
2013PRZY02
YouTube
Visit Channel
I guess for alg number, it's a little shaky with how many substeps or multiple algs you want to learn. Either way, roux only has one step where "long" or "unintuitive" algs are used.
. pizza fedora !@#$%^&*
True. I mean, you could learn KCLL if you're an alg machine. But it's a little more ambiguous with Petrus since the last layer has no definite way of solving. It could be OLL/PLL, COLL/EPLL, CPLL/2GLL, ZBLL, ZZLL (phasing), CP/CO/EP, etc. Roux is a bit more standard since CMLL is the only algorithmic step. I guess some break it uo for 2 look, but still the possibilities for Petrus are too much for it to be definite that Petrus "has fewer algorithms."
 

mDiPalma

Member
Joined
Jul 12, 2011
Messages
1,534
~random stuff~


fedora

True bro, but I was addressing one of the constraints that the OP had in mind (low alg-count for standard use).

I personally use COLL/EPLL for Petrus, and my global average is sub-14.

I just don't appreciate it when people compare the viabilities of ZZ and Roux (simply because CFOP is too mainstream for them), and they completely omit an alternate feasible method from their comparison.
 

GuRoux

Member
Joined
May 6, 2013
Messages
1,712
Location
San Diego, California
WCA
2014TANG03
YouTube
Visit Channel
True bro, but I was addressing one of the constraints that the OP had in mind (low alg-count for standard use).

I personally use COLL/EPLL for Petrus, and my global average is sub-14.

I just don't appreciate it when people compare the viabilities of ZZ and Roux (simply because CFOP is too mainstream for them), and they completely omit an alternate feasible method from their comparison.


How advantageous is the petrus EO step? I feel like doing it regularly with oll/pll or cll/ell would do better. What are your times without EO?
 
Top