Method Comparison Thread

[Anonymous]

So recently there was a thread where the OP asked whether Fridrich, ZZ, or Roux was the best method (he seems to have forgotten Petrus). He recieved a lot of criticism by people who were firm believers in the idea that these methods are all equal. This is probably true.

What some people tried to do in that thread before it was closed was discuss the pros and cons of each method. Personally I don't think that there's anything wrong with that- discussing the viability of EOline as opposed to the cross, first block, or 2x2x2 block is perfectly reasonable. The point here isn't to debate which method is best. What I think would be useful is to use comparisons like this to find the potential flaws in methods, and ways we can improve them. Even though it's not likely that we'll revolutionize the way people build crosses or something, it's still interesting discussion, IMO.

I guess I can start with a question, mostly to Roux users- why do you wait until after the Roux blocks are done to perform EO? Using the freedom of ZZ-style EO, couldn't you reduce your movecount? I don't really know much about Roux, of course, so sorry if it's a dumb question.

 

[Kirjava]

FRIDRICH IS THE BEST METHOD BECAUSE THE WORLD RECORD WAS SET WITH IT

 

[bluedasher]

Even though I use Fridrich I believe Petrus has some potential. I believe Fridich has the fastest LL method, but using a block building technique for the F2L may be faster for some people than pairing and slotting corner - edge pairs. Some people think "BLOCKS" better than "CROSSES" or corner edge pairs. F2L could, I believe get faster, but I think Fridrich still has the optimal way for the LL.

 

[a small kitten]

The pros and cons of each of the mentioned methods are pretty well outlined in the Speedsolving Wiki already. ZZ is hard to approach and practice because after the the EO line, there are 8 possible blocks you can build. Picking the one with the shortest moves can be quite annoying. Additionally, if you are doing OH, you have to also consider which of the 8 is easiest to execute.

I think Fridrich still has the optimal way for the LL.

That's as of now.

 

[Daniel Wu]

I guess I can start with a question, mostly to Roux users- why do you wait until after the Roux blocks are done to perform EO? Using the freedom of ZZ-style EO, couldn't you reduce your movecount? I don't really know much about Roux, of course, so sorry if it's a dumb question.

I practice Roux a bit and have a decent average (21ish). I'd say that I wouldn't do EO before the blocks because that blocks are easier to inspect and doing EO with less edges makes recogniton much faster in addition to it being <M,U>. Plus, there's more freedom to build blocks without the restriction of EO.

 

[Andrew Ricci]

FRIDRICH IS THE BEST METHOD BECAUSE THE WORLD RECORD WAS SET WITH IT

Then why do you use roux?

edit: LOL i fail

 

[cincyaviation]

FRIDRICH IS THE BEST METHOD BECAUSE THE WORLD RECORD WAS SET WITH IT

Then why do you use roux?

This made me laugh very, very hard.

 

[waffle=ijm]

roux is so slow. I don't know why people use it.

Also we wait until blocks are done for EO since there are less EO possibilities to be recognized. Also we have more freedom when doing blocks which is important.

 

[JeffDelucia]

FRIDRICH IS THE BEST METHOD BECAUSE THE WORLD RECORD WAS SET WITH IT

Then why do you use roux?

sigh...

 

[EVH]

roux is so slow. I don't know why people use it.

Very True

On the other hand to me it seems like orienting the edges may not be the best idea because the CMLLs may mess up the edges, however do not take my word for it, as I am no expert.

 

[waffle=ijm]

CMLLs are optimized to preserve only blocks. This makes is shorter than most COLLs and some are faster.

 

[EVH]

CMLLs are optimized to to preserve only blocks. This makes is shorter than most COLLs and some are faster.

Yay I was right!

 

[flan]

Ok I'm a roux noob but I love to toy around with it and work it out because I know the general stages. Anyway I was wondering is there an intuitive way of doing the last bit with the m slices. Right now I insert the last 2 pairs with m slices then do an edge only OLL then an EPLL. I'm not so serious about roux to learn all the cases and such, but working out the last all in 1 step would be fun and I was wandering if it would be feasable? sorry if that was a bit off topic

 

[waffle=ijm]

the last six edges is all intuitive...there really isn't any algs. Just know what you're doing.

 

[EVH]

Ok I'm a roux noob but I love to toy around with it and work it out because I know the general stages. Anyway I was wondering is there an intuitive way of doing the last bit with the m slices. Right now I insert the last 2 pairs with m slices then do an edge only OLL then an EPLL. I'm not so serious about roux to learn all the cases and such, but working out the last all in 1 step would be fun and I was wandering if it would be feasable? sorry if that was a bit off topic

You mean you are a Rouxbik's Noob?

I am having trouble understanding what you are asking, could explain?

 

[_D2_]

No love for heise? I don't use it but it should be in the discussion. I think heise is an advanced sexy method. Neways, i think fridrich has indeed a boring F2l, i have yet to learn petrus, Roux is very interesting though im still slow at block building, ZZ i have yet to try though i dont like itmad:EOline), and Heise... im so slow at block building and Heise blocks are so small i almost cant do it, cant keep track of them.

 

[y3k9]

FRIDRICH IS THE BEST METHOD BECAUSE THE WORLD RECORD WAS SET WITH IT

Corners first is the best method ever because world record was set with it.

 

[jiggy]

...but I think Fridrich still has the optimal way for the LL.

Well, what about Roux? =p

Neways, i think fridrich has indeed a boring F2l...

How is F2L boring? It's hugely varied; there are many different cases, any number of ways to approach each case, and so many little tricks to keep things interesting!


I know Big Green has sub-10'd with Roux in comp, can anyone tell me the fastest Petrus time recorded in comp, and/or the current fastest Petrus solver?

 

[riffz]

I know Big Green has sub-10'd with Roux in comp, can anyone tell me the fastest Petrus time recorded in comp, and/or the current fastest Petrus solver?

Erik Johnson, I believe.

 

[somerandomkidmike]

FRIDRICH IS THE BEST METHOD BECAUSE THE WORLD RECORD WAS SET WITH IT

Corners first is the best method ever because world record was set with it.

Corners first has more potential than people give it credit for.

the last six edges is all intuitive...there really isn't any algs. Just know what you're doing.

If you don't want to use intuition, then you can always use the 2.5 look last 6 edges on this site here: http://rubikscube.info/lastsix2look.php .