• 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 35,000+ people from around the world today!

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

Method Pros and Cons: CFOP vs Roux vs ZZ

Potato Kuber

Member
Joined
Jan 29, 2019
Messages
5
What I basically want to know is what are all the pros and cons of the three most popular methods.
I ,at the moment, use Roux but I want to know if Roux is a good method.
Write down your opinions on CFOP, Roux and ZZ.
 
Joined
Feb 23, 2019
Messages
381
Location
The FitnessGram Pacer Test is a multi stage...
Please just go searching through the forum for a little bit before asking a question. This in particular has been answered and discussed thousands of times. The answer is: nobody really knows. Most people think CFOP and Roux are about the same, with ZZ maybe being a bit slower unless you ask PapaSmurf or GenTheThief. Who knows ZZ is probably pretty similar but nobody has shown it exclusively yet. I use roux and like that; I think it is mostly a matter of preference.
 

Izaden

Member
Joined
Sep 18, 2019
Messages
39
Location
Kitchener, Ontario
These three links go over the pros and cons of the three methods in question.

One thing I will add that is not stated there is that the reason CFOP is so fast even with its high move count is the reliance on algs and muscle memory which takes away some of the pausing to look at the cube and give you a slightly higher tps since you have practiced the move sequences so many times (At high levels).

Roux IMO is on par with CFOP. The choice (again, IMO) should be based on if you want to use mostly intuitive solutions (roux) or alg based solutions (cfop)

I solve using CFOP because I initially learned lbl and don't have a high enough understanding of the cube to use something as intuitive as roux efficiently.

https://www.speedsolving.com/wiki/index.php/CFOP_method#Pros
https://www.speedsolving.com/wiki/index.php/Roux_method#Pros
https://www.speedsolving.com/wiki/index.php/ZZ_method#Pros
 

PapaSmurf

Member
Joined
Jan 4, 2018
Messages
425
WCA
2016TUDO02
YouTube
PapaSmurf Cubes
If you want a quick overview of what method I think is the best at the highest level, I'll put it in a spoiler, with some reasoning. Otherwise, read around.
CFOP: worst out of the big 3. It's basically a worse version of ZZ. Has a movecount of 55-60, 4.5 looks (if a pair is half a look and with XCross) and even with a lot of extra algs, you aren't getting an efficiency of 50 moves or less. It has some advantages of being well developed and very decent lookahead.
ZZ (EOCross and ZBLL): joint best. Has a movecount of 50-55, 4 looks (if a pair is half a pair). Requires 349 algs to be very good, but that's similar to what top level CFOPers know. Has half the F2L cases of CFOP and no rotations and minimal regrips so tps can be abused more. EO also means that ZZ has better recog for back slots than CFOP. Disadvantages of blind spots still existing, even if they're somewhat mitigated by EO and also EOCross+1 is very difficult.
Roux: joint best has the lowest movecount of about 47 (45-50). Isn't the outright best because TPS isn't as good (SB has regrips caused by overturning) and LSE <MU> is worse than <RU> of ZZ and CFOP F2L pairs due to purely flicking compared to wrist then flick). Efficiency is the best though and still has a pretty high tps as LSE isn't exactly slow at all.
LEOR (aka RouxZZ) is first block, EODFDB, right block, ZBLL. I think it’s equal to Roux and ZZ. It has the low movecount of Roux and the ergonomics of right block into ZBLL. All the steps apart from EODFDB are definitely really good. Just someone has to put in the effort to develop the resources for that one step and see if it is good or not.
If you want to see more of me defending ZZ, look at the thread shared by cubinwitdapizza.
 

GuRoux

Member
Joined
May 6, 2013
Messages
1,714
Location
San Diego, California
WCA
2014TANG03
YouTube
franktangtartharakul
I think if eodfdb is one look, then it will be slow because you would need a considerable pause to recognize it. it would need to be something like lse which is split into 3 steps (4a, 4b, 4c).

i wonder if fb -> sb -> eodfdb -> zbll would be a better method. since eodfdb would be [M,U] here, you would recognize the corners of the zbll while doing it. So it should speed up zbll recognition since you only have to recognize edges.
 
Top