Petrus vs. Roux

[Smartyy]

I know theres a lot of "fridrich vs. X" threads, but I want to learn either Roux or Petrus method.
Here's the pros and cons (for me personally) I find with each method:

Roux:

Pros:
-It's very unique, and, although that doesn't count for anything, I like to be different.
-You kinda build the whole cube in one step. Now I say that and I mean that you solve the cube in pieces and it all comes together towards the beginning, you just finish it towards the end. I dunno, it just seems to be solved in one step.
-There aren't many algs to memorize
-There is no step that is "excruciatingly harder" than any other step, they are all about the same difficulty.

Cons:
-It uses the M layer a lot, and that's kinda weird (weird finger-wise).. however I'll get use to it
-A lot of the few cases look very alike, and the algorithms that are associated with them REALLY confuse me and therefore almost never work. (mainly talking about the last step)
-The VERY few tutorials I found are hard to understand and confuse me a lot, so just learning it will be hard. (i did solve it once, but I got lucky on the last step..)


Petrus:

Pros:
-Mostly intuitive and almost no algs to learn (unless you go and learn all OLL and PLL algs to eliminate steps 5-7)
-The tutorial I found on his website was really easy for me to follow, so learning it obviously will be a lot easier.
-Usually a bit of the first block/the entire first block can be built in a few moves, so that makes it quicker
-Mostly all the steps are relatively easy to learn.. I just started petrus like, not even an hour ago and have all but the 4th step pretty good. (not fast though)

Cons:
-The fourth step is really hard, but with practice it will come easier
-If i wanted to be REALLY fast at it, I would still have to learn all the OLL and PLL algs that I was trying to avoid in the first place Tongue


So which should I learn? I know you all can't choose for be but you can always persuade me. Tongue
Which is more commonly faster, Petrus or Roux?


Thanks,

-Smartyy!

 

[Johannes91]

I want to learn either Roux or Petrus method.

Why not both?

-Mostly intuitive and almost no algs to learn (unless you go and learn all OLL and PLL algs to eliminate steps 5-7)

You probably know this, but you need only 7 of the 57 OLL algorithms in Petrus. I think it's better to use OCLL for those 7 algs, to make it clear that it's not the same as Fridrich OLL.

-If i wanted to be REALLY fast at it, I would still have to learn all the OLL and PLL algs

How fast is REALLY fast?

 

[fanwuq]

I don't know. They are both very nice and fast. M turns are very fast.
I prefer Roux at the moment. Petrus is more similar to Fridrich, just block building and orienting instead of using slots.
Roux is block building and corner first.
Step 4 is easy and very fast (I average sub-4 seconds on it when I try Petrus)! Orientation Identification kills my times, so I use Fridrich.
Both are easy to understand. But it was difficult when I was using beginner LBL many months ago.

 

[pjk]

Roux:
Cons:
-It uses the M layer a lot, and that's kinda weird (weird finger-wise).. however I'll get use to it
-A lot of the few cases look very alike, and the algorithms that are associated with them REALLY confuse me and therefore almost never work. (mainly talking about the last step)
-The VERY few tutorials I found are hard to understand and confuse me a lot, so just learning it will be hard. (i did solve it once, but I got lucky on the last step..)

All of those will fix as you practice. M moves will become very fluid, cases in the end will be easy to recognize, and you won't need any tutorials. In regards to the last part, what is confusing/hard to understand?
Go with Roux

 

[Smartyy]

In regards to the last part, what is confusing/hard to understand?
Go with Roux

Basically i get confused after you do the second 1x2x3 block, but the part I can NEVER get is the last step.. as in "Solving the last 6 edges and 3 centers"
The algs are like M'UM and They just never work. I'll do them like, 6 times in a row and then they will finally work on the 6th or 7th one, but i'm sure that you shouldn't have to do that.

As for which i'll learn, I could just learn both, as somebody suggested above After I get them down pretty good i'll just choose whichever one I like the best. Right now it's Petrus, because I can always solve it, but once I figure out Roux who knows I may like roux better... so I'll just learn both and choose once they are good.

 

[immortalcube]

I also had this dilemma when I first started speedcubing. I couldn't understand a lot of what was on the Roux site, so I ended up using Petrus. Don't worry about the 4th step, just go play around with it for a while and you'll get it. You just need to understand how to move pieces around without messing up other stuff. Eventually it'll become very easy and natural.

 

[CanadianPires]

You are probably having trouble with the last step because the alg for ORIENTATION is M'UM'. However this is not the alg for the last step. The last step only requires three algs:

1. M' U2 M'
2. M2 U2 M2
3. M' E M' (i think not 100% on this one, i know its something with an "E")

 

[fanwuq]

3. M E2 M E2

 

[Swordsman Kirby]

-There is no step that is "excruciatingly harder" than any other step, they are all about the same difficulty.

I'll disagree and say that step 1 is by far the hardest of the four steps.

Cons:
-The fourth step is really hard, but with practice it will come easier

S4 is trivial compared to S1 or S2.

1. M' U2 M'
2. M2 U2 M2
3. M' E M' (i think not 100% on this one, i know its something with an "E")

1. M' U2 M' U2
2. M2 U2 M2 U2
3. M' E2 M' E2

Have to be precise, in my opinion.

 

[Lofty]

Do you ever want to become fast at OH?
Roux probably isn't the best method to pick for OH because it has so many M turns. But other than that they are both great methods learn them both.
If you learn COLL as opposed to CMLL you can use it on both Petrus and Roux for a nice LL.

 

[pjk]

I actually think Roux could be very quick 1 handed. Doing an M move with 1 hand wouldn't be too hard with practice. And while doing one hand, I think you could get a move count down to 40-45 avg.

 

[SD14]

i have been playing around with roux here latly and seems that it could be faster than petrus with some work cause well i at least seem to skip lots of steps and the last part the cases are really easy to recognize

 

[Smartyy]

Okay so this is what I am doing.. since I am in no rush to become a really good speedcuber, I might as well experiment for a bit.
For one week I will only solve the cube with Petrus, even in 4x4 and 5x5 solves I will only use petrus. By the end of the week (This sunday, I started last sunday) I will time myself on the 3x3, 4x4, and 5x5. Obviously they will probably not be record times or pb times since I have been doin LBL for a good three months off and on.
Started next monday I will only be learning Roux. All my 3x3 4x4 and 5x5 solves will be using only Roux method, and the following monday I will time myself with the 3x3 4x4 and 5x5. I will compare my two sets of times and also give some additional thought into how the methods felt (strictly) in my opinion.

I will not choose whichever method I choose (unless, of coarse, I learn both!) based solely on my times, because (for example) Roux method might be slow at first, then I could get immensely fast at it and start progressing really far with it while Petrus stays slow and steady with it's progress.
However, I am considering learning both of them because come time to compete, I might find in inspection that the first two steps could be done in lets say 12 moves with Roux and it would take 15 for Petrus (not saying I could even look ahead that much, but you get the picture) or vice versa. This way I could adapt to the situation instead of taking the only way I know. In my humble opinion, adapting to a situation is extremely beneficial and a lot better than not being able to adapt.
Besides, learning Petrus I would eventually have t learn 57 PLL and the 2-something OLL (or was that backwards?) so I could also learn Fridrich down the path of time a bit faster since I already know the PLL and OLL.

Thanks for all your help, I will make a new thread for all the times etc. on Sunday.


-Smartyy!

 

[Lofty]

Well doing Petrus there isn't really a set LL system to do. If you were to do OLL+PLL you would only have 7 OLL and 21 PLL so not near as many as you thought.
You could also learn one of the many LL+Last Slot variants like Winter/F2LL or MGLS. There is COLL which is 5+6 of 2GLL which is 6+7 or ZBLL which is 5+6+7. And I'm sure there are even more things that can be done with Petrus.
Anyway I think its a great idea you are pursuing! You are probably the first beginner who has said he is not in a rush to get faster and actually wants to find what is best for him instead of just learning something to get fast now tho it may not be best for him.
Good Luck!

 

[Smartyy]

Lofty:
Okay wow lol the first half of your message made like.. no sense to me. The only acronyms I know are OLL and PLL.. you confused me with your F2LL MGLS COLL ZBLL etc.. lol I was like "wtf.. how do you know all this!?" ah well.. i'll learn them over time.
But the second paragraph I have to thank you. My grandpa came over yesterday and he and I had never really talked.. but yesterday he an dI had like a 4 hour conversation mainly about patients... and why it's so important. That is why I suddenly went from "Which is fastest?" to "I am in no hurry."
Thanks for your post.. somehow it inspired me to actually pursue my new goal knowing I am the only one you know of that is doing/saying what I did.

-Smartyy!

 

[Lofty]

Haha sorry, I have been cubing for over a year now so its just different things you pick up lol. Mainly I was just making the point that there are so many options with Petrus.
F2LL is when you have your edges oriented and when you insert the last pair you do it in such a way to orient corners too.
MGLS- very similar only first you insert just the edge then you insert the corner while also orienting the corners.
COLL- 5+6 at once. So basically your corners go from unsolved to completely solved and only edges are left to deal with. You both orient and permute the corners in one alg.
ZBLL- Solve everything in the LL (your edges are already oriented) in one alg.