Should I switch back from ZZ to Roux?

[edward_9x]

Hello. I'm using ZZ-b (actually just ZZ-VH for 99% of my solves) (edit: actually no, I used ZZ-OP lmao) and I average around 25-35 seconds.

I have trouble planning EO-line, a lot of the time I come up with a solution on the fly after I start the timer. Last layer also gets very boring after a while, it's just algorithm-algorithm. The ergonomics are nice, but Roux is mostly <R,Rw,U,M>-gen which is nicer than ZZ's <R,U,L>-gen. The blocks are also more efficient than ZZ because you don't have to "dodge" the EOLine with U moves.

Should I switch back to Roux? It was my main method when I was averaging 45-50s. While I think ZZ is still a very awesome method that deserves more recognition (and will probably still be my secondary method), I don't know if it's what I want to use as my *main* method... should I stick with ZZ or return to using Roux?

 

[OreKehStrah]

Hello. I'm using ZZ-b (actually just ZZ-VH for 99% of my solves), and I average around 25-35 seconds.

I have trouble planning EO-line, a lot of the time I come up with a solution on the fly after I start the timer. Last layer also gets very boring after a while, it's just algorithm-algorithm. The ergonomics are nice, but Roux is mostly <R,Rw,U,M>-gen which is nicer than ZZ's <R,U,L>-gen. The blocks are also more efficient than ZZ because you don't have to "dodge" the EOLine with U moves.

Should I switch back to Roux? It was my main method when I was averaging 45-50s. While I think ZZ is still a very awesome method that deserves more recognition (and will probably still be my secondary method), I don't know if it's what I want to use as my *main* method... should I stick with ZZ or return to using Roux?

I mean, Roux is just a better method than ZZ and is the best method for OH given what the Rouxers are putting down at home.

It sounds like you like the method more as far are solve style goes so just switch back to Roux.

ZZ really is an enthusiast method at this point. You’re not using any advanced alg sets for it either, so you’re not getting anything out of doing EO.

 

[efattah]

Yeah as OreKehStrah said, without full ZBLL, you don't gain anything from ZZ.

 

[OreKehStrah]

Yeah as OreKehStrah said, without full ZBLL, you don't gain anything from ZZ.

I’d say OLS is also valid. That’s what I’m slowly learning now

 

[ProStar]

I’d say OLS is also valid. That’s what I’m slowly learning now

How many algs is OLS with EO already finished? And I assume it would be unreasonable to try to learn OLLCP-LS?

 

[OreKehStrah]

How many algs is OLS with EO already finished? And I assume it would be unreasonable to try to learn OLLCP-LS?

ZZ OLS is 502 algs. 509 if you count the 7 OCLLs. There are 18 sets of 27 and 2 sets of 8 (the cases where both are in the slot but the corner is twisted).

 

[edward_9x]

I think I know what to do now. I should use Roux as my main method in timed solves, and ZZ for casual cubing when I'm bored.

 

[edward_9x]

One last thing: because RUL blockbuilding feels so nice (tps and how it "flows"), I actually still want to do ZZ as well. Is it viable to main 2 methods at once? I know Tao Yu is method neutral

 

[ruffleduck]

Is it viable to main 2 methods at once?

Practically? Not a great idea. If you don't care too much about speed and want to have fun? Go for it.

 

[edward_9x]

If you don't care too much about speed and want to have fun? Go for it.

Alright, so double-method neutral it is. Quite a plot twist :] I was just worried I would get slower at ZZ if I kept doing Roux

 

[artless1der]

Hello. I'm using ZZ-b (actually just ZZ-VH for 99% of my solves), and I average around 25-35 seconds.

I have trouble planning EO-line, a lot of the time I come up with a solution on the fly after I start the timer. Last layer also gets very boring after a while, it's just algorithm-algorithm. The ergonomics are nice, but Roux is mostly <R,Rw,U,M>-gen which is nicer than ZZ's <R,U,L>-gen. The blocks are also more efficient than ZZ because you don't have to "dodge" the EOLine with U moves.

Should I switch back to Roux? It was my main method when I was averaging 45-50s. While I think ZZ is still a very awesome method that deserves more recognition (and will probably still be my secondary method), I don't know if it's what I want to use as my *main* method... should I stick with ZZ or return to using Roux?

I say *YES* ... Roux is more fun and overall provides silutions with less moves

I myself started with CFOP and right about when I started to get solves in 1:30 or less I switched to Roux and very soon began getting faster solves, getting under 1min in a week or so

I find the intuitive nature of the steps of the method to be more fun and feel like this provides more opportunities for influencing solves as they evolve on the fly (which I also find to be more fun))

 

[abunickabhi]

I say Yes, Roux is more varying and fun.

 

[BradleyGallella]

Hello. I'm using ZZ-b (actually just ZZ-VH for 99% of my solves), and I average around 25-35 seconds.

I have trouble planning EO-line, a lot of the time I come up with a solution on the fly after I start the timer. Last layer also gets very boring after a while, it's just algorithm-algorithm. The ergonomics are nice, but Roux is mostly <R,Rw,U,M>-gen which is nicer than ZZ's <R,U,L>-gen. The blocks are also more efficient than ZZ because you don't have to "dodge" the EOLine with U moves.

Should I switch back to Roux? It was my main method when I was averaging 45-50s. While I think ZZ is still a very awesome method that deserves more recognition (and will probably still be my secondary method), I don't know if it's what I want to use as my *main* method... should I stick with ZZ or return to using Roux?

Use whatever method you find the most fun. If you have more fun, not only will you get better faster, because you will feel more motivated to practice, you will obviously just enjoy the hobby more!


However, if you really want an answer for which method is better then Roux is better, it’s not really even a contest. (Brace yourself, I’m about to walk of text)

ZZ’s widely accepted benefits are mainly that it has a relatively low move count and is very ergonomic.

when it comes to the move count argument, ZZ only comes close to matching Roux’s move count if you know full ZBLL, and ZBLL has a swathe of problems itself. Namely, bad ergonomics for a lot of the cases, long recognition pauses, and just bad recognition in general. In many cases it’s honestly just faster to use OLL and PLL, but then why would you use ZZ in the first place?

in terms of ergonomics I actually think the argument for ZZ is also rather weak. It’s rotation less but that’s about it. The amount of re-grips, and switching from Left to Right hand turning you have to do, due to pieces for one side block existing on the other side, honestly counter acts any other ergonomic advantages IMO. Not to mention ZBLL doesn’t have the best ergonomics (which again has to be used to make the method viable in terms of movecount)

edit: I’ll include this because people keep mentioning it, most ZZ users started using EOcross over EOline because it has better ergonomics. I think the argument there is even worse, because it takes quite a sizable efficiency hit, and it just makes your inspection even worse compared to F2B. At which point you have a method that has SLIGHTLY better efficiency then CFOP (not any better if you consider you could probably just plan an Xcross with the same difficulty as an EOcros), and SLIGHTLY better ergonomics compared to CFOP
(EOcross has worse ergonomics than Regular cross), for astronomically worse inspection, and worse recognition. So compared to Roux it’s not even a question.

On the other hand Roux doesn’t really have either of these problems. Roux has a comparable move count, but CMLL and LSE don’t have the problems ZBLL does. CMLL is a small, very well refined set of algs that have great recognition, move count, and ergonomics, and LSE has easy look ahead and is 2 gen.

Comparing F2B and ZZF2L again there’s no contest. The ergonomics are ACTUALLY good, requiring little to no L R switching as you just plan one block in inspection, then the rest is R r U gen.

and as much as we say F and B moves are bad when talking about these methods, there is a time and a place where they’re just simply faster to use. If using them is the difference between a 2 move pair and a 5 move pair then the hit to the ergonomics might just be worth it, or not, the point is is that with Roux you have the option to choose which one is optimal for you at the time. You could never use them, or use them all the time, or only when it’s optimal. With CFOP you’re forced to use them or a Rotation, in ZZ you are forced NOT to used them, with Roux you have the choice.

but then finally, there is one thing that I haven’t mentioned yet, but I believe to be the main reason why Roux is better then ZZ, and that’s the inspection. During inspection you want to look ahead as far as possible to try to maintain smooth turning and high tps, with few pauses throughout the solve while staying efficient. Top roux solvers can plan most of not all of F2B in inspection, but with ZZ that’s much harder. EOLine is very difficult and is a much less significant portion of a solve in terms of movecount and time, it will take up far more of your inspection time that you could instead be using to look further into your solve.

Roux is just better. Not to mention it’s got a MUCH better moveset for OH if you care about that.

 

[edward_9x]

Somehow, after 4 days of going "method neutral" with Roux my ZZ blocks got faster. I only now realized how much I relied on EO already being solved when looking ahead...

 

[ruffleduck]

Use whatever method you find the most fun. If you have more fun, not only will you get better faster, because you will feel more motivated to practice, you will obviously just enjoy the hobby more!


However, if you really want an answer for which method is better then Roux is better, it’s not really even a contest. (Brace yourself, I’m about to walk of text)

ZZ’s widely accepted benefits are mainly that it has a relatively low move count and is very ergonomic.

when it comes to the move count argument, ZZ only comes close to matching Roux’s move count if you know full ZBLL, and ZBLL has a swathe of problems itself. Namely, bad ergonomics for a lot of the cases, long recognition pauses, and just bad recognition in general. In many cases it’s honestly just faster to use OLL and PLL, but then why would you use ZZ in the first place?

in terms of ergonomics I actually think the argument for ZZ is also rather weak. It’s rotation less but that’s about it. The amount of re-grips, and switching from Left to Right hand turning you have to do, due to pieces for one side block existing on the other side, honestly counter acts any other ergonomic advantages IMO. Not to mention ZBLL doesn’t have the best ergonomics (which again has to be used to make the method viable in terms of movecount)

On the other hand Roux doesn’t really have either of these problems. Roux has a comparable move count, but CMLL and LSE don’t have the problems ZBLL does. CMLL is a small, very well refined set of algs that have great recognition, move count, and ergonomics, and LSE has easy look ahead and is 2 gen.

Comparing F2B and ZZF2L again there’s no contest. The ergonomics are ACTUALLY good, requiring little to no L R switching as you just plan one block in inspection, then the rest is R r U gen.

and as much as we say F and B moves are bad when talking about these methods, there is a time and a place where they’re just simply faster to use. If using them is the difference between a 2 move pair and a 5 move pair then the hit to the ergonomics might just be worth it, or not, the point is is that with Roux you have the option to choose which one is optimal for you at the time. You could never use them, or use them all the time, or only when it’s optimal. With CFOP you’re forced to use them or a Rotation, in ZZ you are forced NOT to used them, with Roux you have the choice.

but then finally, there is one thing that I haven’t mentioned yet, but I believe to be the main reason why Roux is better then ZZ, and that’s the inspection. During inspection you want to look ahead as far as possible to try to maintain smooth turning and high tps, with few pauses throughout the solve while staying efficient. Top roux solvers can plan most of not all of F2B in inspection, but with ZZ that’s much harder. EOLine is very difficult and is a much less significant portion of a solve in terms of movecount and time, it will take up far more of your inspection time that you could instead be using to look further into your solve.

Roux is just better. Not to mention it’s got a MUCH better moveset for OH if you care about that.

I disagree with your assessment of ZZ. ZBLL is without a shred of doubt the future of not only ZZ but of cubing at large. The alg quality is high and improving constantly with further developments, and recog systems are still to this day being refined and put into practice (see Tymon with his unfathomably fast recog). You also judge ZZ by its obsolete EOline version, which has long been dismissed as poor for 2H solving. All serious ZZ solvers use EOcross nowadays. Regarding inspection, top ZZ solvers can plan EOcross and even first pair. Overall, I would judge ZZ at least theoretically equal to CFOP. (See my paper in my signature.) That being said, I can't argue with your point that Roux is the dominant method. Roux is simply the best method, and it hardly is close.

 

[edward_9x]

You also judge ZZ by its obsolete EOline version, which has long been dismissed as poor for 2H solving. All serious ZZ solvers use EOcross nowadays. Overall, I would judge ZZ at least theoretically equal to CFOP.

I think EOline is actually quite good for 2H. The moves definitely flow nicely when you use certain techniques and tricks for it, even with the ergo and lookahead problems. It's better than FreeFOP at least

 

[OreKehStrah]

I think EOline is actually quite good for 2H. The moves definitely flow nicely when you use certain techniques and tricks for it, even with the ergo and lookahead problems. It's better than FreeFOP at least

Dawg, nahhhhhh. If you ask any of us ZZ users, I think basically everyone will very quickly tell you it is not good. The efficiency gain of 4-6 moves is not worth the major drop in TPS and lookahead.

 

[Yoruba]

Should I switch back to Roux? It was my main method when I was averaging 45-50s. While I think ZZ is still a very awesome method that deserves more recognition (and will probably still be my secondary method), I don't know if it's what I want to use as my *main* method... should I stick with ZZ or return to using Roux?

Yeah, do it. Switching methods can give you a better understanding of the cube and make cubing more interesting, it's all about fun. That being said if you ever switch back go with eocross.

 

[BradleyGallella]

I disagree with your assessment of ZZ. ZBLL is without a shred of doubt the future of not only ZZ but of cubing at large. The alg quality is high and improving constantly with further developments, and recog systems are still to this day being refined and put into practice (see Tymon with his unfathomably fast recog). You also judge ZZ by its obsolete EOline version, which has long been dismissed as poor for 2H solving. All serious ZZ solvers use EOcross nowadays. Regarding inspection, top ZZ solvers can plan EOcross and even first pair. Overall, I would judge ZZ at least theoretically equal to CFOP. (See my paper in my signature.) That being said, I can't argue with your point that Roux is the dominant method. Roux is simply the best method, and it hardly is close.

My assessment of ZZ was based on the comparison with Roux not CFOP, as that’s what the original post was about. And Roux can basically do everything that ZZ does but just kind of better without really any of the same glaring issues. From what you said at the end of your comment you probably agree. But whenever I said something was bad about ZZ it was by comparison with Roux.

But I will respond to your points cause I think they’re interesting.(I’m gonna wall of text again)

I do actually agree that CFOP and ZZ may have very similar maximum potentials, I just think that ZZ is much harder to get good with, meaning that if you’re purely looking for practice time = better results, CFOP is better.



I think my assessment of ZBLL was fair, I didn’t say it was always worse, just in many cases. To give an extreme example, when it comes to any Sune or anti Sunr cases for ZBLL just doing a sune or an anti sune and then PLL is always faster, that’s the general consensus even among prominent ZBLL users, and that’s the case for a lot of ZBLL cases. And yes the recognition is getting better but it is still worse, so couple that with all the extra trouble it takes to get it to the ZBLL cube state and it stops making too much sense. It’s really only worth it to learn specific sets of ZBLL that are probably worth it, but again, if you aren’t using full ZBLL, why are you using ZZ, then you just have bad movecount, bad lookahead, bad inspection (especially if you’re a using EOcross)? What makes a lot more sense is using CFOP, learning partial ZBLL like mentioned before, and then just using edge control methods whenever it makes sense, and use OLL and PLL or some other last layer method whenever else. When used like this it becomes incredibly good. This is what Tymon does, since you cited him as an example of how ZBLL is good. It’s just not good within the context of ZZ. My criticism was of ZBLL was within the context of ZZ, not CFOP as that was what the original post was about.



When it comes to EOline vs EOcross, the reason I judged it based off of EOline is because I honestly don’t see a point in doing EOcross. At that point you’ve got a method that trades a SLIGHTLY better movecount, and SLIGHTLY better ergonomics compared to CFOP, for an astronomically worse inspection, and worse recognition, but when it comes to that argument it’s a little fuzzier. It’s kinda hard to tell, so this is just my opinion.



And believe me, as a Roux main and ZZ enjoyer, saying this isn’t doing me any favors. I really do wish ZZ was better.

 

[BradleyGallella]

Dawg, nahhhhhh. If you ask any of us ZZ users, I think basically everyone will very quickly tell you it is not good. The efficiency gain of 4-6 moves is not worth the major drop in TPS and lookahead.

I actually think it’s comparable, I used to practice ZZ and Roux a lot up until recently. The efficiency loss is more than 4 - 6 moves, And look ahead is honestly just as good with EOline IMO (that may be because I came from roux). The ergonomics are better but it trades that for worse inspection.

I don’t think it makes much sense honestly. I explain more in another reply.