Will anyone ever come out with a better method?

[thiJUMBA]

Yes. And it will be me.

you've only just joined the cubing community [so did i...] :fp

i've voted yes on the poll, 'cos i think people have got potential on finding a new method but i doubt there would be another method as popular as fridrich
because the number of algorithms is reasonable and since its algorithmic it can be performed faster and i reckon the number of algorithms affect its popularity cos most people aren't that hardcore into learning over a thousand algs. to get sub 10

-thiJUMBA

 

[mazei]

Roux

But its not good for big cube reduction but who cares, its fast on 3x3.

 

[Muesli]

A technique to spot the optimal solution...somehow...

 

[Ton]

Will anyone ever come out with a better method that is widely recognised to be faster or more popular than Fridrich F2L + OLL + PLL?

Eventually there will be better and more popular method's

In the 80's a cornerfirst method was more popular , this was only changed after 2000-2003 to a F2L method. So it may take some time.

If we would believe F2L is the best, there will never be progress.

 

[Tim Major]

I think Petrus would have real potential, if people focused on it, instead of Fridrich. It has fewer moves, but keeping up a good tps is hard. If you practise hard enough, finding the pieces become second nature. If everyone practised Petrus, I think sub 10 averages would be achieved, then sub 9. Beyond that? Sub 8 a12? You need a different method than Fridrich, and probably Petrus. However, I think Petrus has real potential that hasn't been reached, I'm just not a block builder.

 

[cmhardw]

In the 80's a cornerfirst method was more popular , this was only changed after 2000-2003 to a F2L method. So it may take some time.

Considering how quickly M and U triggers can be done, I have no idea why corners first isn't still a viable method. I guess it's more of a "popularity" contest than anything else. David Allen, if he still practices corners first, was quite fast with the method. I wonder what his times are now, Gene Means too.

Chris

 

[AvGalen]

Will anyone ever come out with a better method that is widely recognised to be faster or more popular than Fridrich F2L + OLL + PLL?

Eventually there will be better and more popular method's

In the 80's a cornerfirst method was more popular , this was only changed after 2000-2003 to a F2L method. So it may take some time.

If we would believe F2L is the best, there will never be progress.

CFOP has proven to be the best method for a long time now. It was in use when there were 100 official speedcubers and it is still the best now that there are > 5000 official speedcubers. I wouldn't say there hasn't been progress though. Lots of algorithms have changed and in general the consensus is now that longer algorithms that are more fingertricky are faster than optimal or near optimal algorithms. (U/Z/T/Y are the best examples)

I also don't think the fastest people in the world really use pure CFOP. They often choose which way to insert a pair, which pair to insert first, and from which angle to perform a certain OLL/PLL combination to optimise for speed. Edge/Corner-control and (partial) ZB are also quite common. When you give the same scramble to 10 speedsolvers you are likely to see 7 or 8 different solutions depending on style and different optimisations.

Finally, the people that have chosen to focus on a "high alg, less steps" method have improved, but the people that have chosen to focus on "pure CFOP with personal tricks" have improved faster. CFOP isn't perfect, but getting fast with it is relatively easy and improving after that comes from experience. The balance between alg-count, execution-speed and learning-curve is really good.

 

[TMOY]

Considering how quickly M and U triggers can be done, I have no idea why corners first isn't still a viable method. I guess it's more of a "popularity" contest than anything else. David Allen, if he still practices corners first, was quite fast with the method. I wonder what his times are now, Gene Means too.
Chris

Considering the huge number of stupid clueless comments I have heard about CF from cubers (even good ones) that's not very surprising
In particular the "high move count" I often hear just makes me laugh. Which STM counting (which is the most relevant one) it has a lower move count than Fridrich. And even when sticking with HTM, I have won FMC events with CF (OK, not really pure CF, but still CF).
And what's more important: I relaly like this method

 

[DavidWoner]

In the 80's a cornerfirst method was more popular , this was only changed after 2000-2003 to a F2L method. So it may take some time.

Considering how quickly M and U triggers can be done, I have no idea why corners first isn't still a viable method.

I think Roux is a sort of modified corners first. Some early corners first methods would build a layer (1x3x3) or layer minus an edge as they solved corners. Roux takes this same sort of idea, but instead builds 2 1x2x3s in order to reduce the cube to <M,U> faster.

But in the literal sense of the word Roux is also an F2L method, it's just that the first two layers happen to be R and L instead of adjacent layers.

 

[qqwref]

I think Petrus would have real potential, if people focused on it, instead of Fridrich. It has fewer moves, but keeping up a good tps is hard. If you practise hard enough, finding the pieces become second nature. If everyone practised Petrus, I think sub 10 averages would be achieved, then sub 9. Beyond that? Sub 8 a12? You need a different method than Fridrich, and probably Petrus. However, I think Petrus has real potential that hasn't been reached, I'm just not a block builder.

Sub-8 avg12? Congratulations, you made me laugh.

Personally I think Petrus does not have quite as much potential (on average) as Fridrich - the block building part does save a few moves, but recognition is harder than for F2L and the moves are a bit less fingertricky. And if you just want to use as few moves as possible, why choose Petrus and not Heise? I'm sure with a sub-40 move average a fast turner could easily break 10 seconds... that is, if 4 tps is even possible with Heise ;-)

IMO, if there's any method out there today (not counting things requiring memorization of several hundred algs) that has the potential to beat Fridrich, it's Roux, but getting fast will still require a ton of practice and some natural talent (as well as being good with M-slices, which not everyone is). But I won't be surprised if in 20 years the top solvers are still using CFOP.

 

[Tim Major]

I think Petrus would have real potential, if people focused on it, instead of Fridrich. It has fewer moves, but keeping up a good tps is hard. If you practise hard enough, finding the pieces become second nature. If everyone practised Petrus, I think sub 10 averages would be achieved, then sub 9. Beyond that? Sub 8 a12? You need a different method than Fridrich, and probably Petrus. However, I think Petrus has real potential that hasn't been reached, I'm just not a block builder.

Sub-8 avg12? Congratulations, you made me laugh.

Personally I think Petrus does not have quite as much potential (on average) as Fridrich - the block building part does save a few moves, but recognition is harder than for F2L and the moves are a bit less fingertricky. And if you just want to use as few moves as possible, why choose Petrus and not Heise? I'm sure with a sub-40 move average a fast turner could easily break 10 seconds... that is, if 4 tps is even possible with Heise ;-)

IMO, if there's any method out there today (not counting things requiring memorization of several hundred algs) that has the potential to beat Fridrich, it's Roux, but getting fast will still require a ton of practice and some natural talent (as well as being good with M-slices, which not everyone is). But I won't be surprised if in 20 years the top solvers are still using CFOP.

I'm not talking about the next year or too. Josh (or is it John) Larsen is pro at Petrus, however, if more people focused on Petrus, new variations/better ways of solving, still using the skeleton Petrus method, would probably be invented. If practised enough, I'm sure some fast cuber would get to a stage he can turn almost as fast as with Fridrich. Heise is another story. People have sub 30ed with Heise, though, and with enough practise... No. Heise was made for FMC. It would never be a suitable speedsolving method, unless inspection was increased to one hour, and you could write down your solution. Heise would never be fast, but Petrus could be. I love watching Josh's/John's videos, for slower turning. Petrus is more of an elegant method, whereas often Fridrich is just sharp and choppy.
I agree with your last comment though. Not the Roux part, the next part, though maybe Roux could work too.

Edit: And I never said sub 8 would be achieved with Petrus. I said " I think sub 10 averages would be achieved, then sub 9. Beyond that? Sub 8 a12? You need a different method than Fridrich, and probably Petrus."
See? Stop laughing

 

[mazei]

What do you mean 'maybe Roux could work too'? Its like, the method that will overpower Fridrich one day. But I find that one reason some people aren't keen on learning it is because like me, I love big cubes and so far find reduction more fun compared to other methods. And try doing last 6 edges on Roux on a 7x7, its gonna kill you. I've tried Stadler but meh, until someone gets a sub-1 on 4x4 with it, I'm not convinced.

So really, what are we trying to discuss here? A method that can be effectively used on any sized cube or just 3x3? Because if its just 3x3, I would say Roux, if the other, Fridrich.

 

[qqwref]

Yeah, the weakness of Roux is that it is only really good for 3x3 speed. On big cubes and OH, it is completely destroyed by Fridrich. This is pretty much the main reason why I'll never fully switch to another method - Fridrich is just too versatile, and I've never only been interested in 3x3 speed.

 

[AvGalen]

In the 80's a cornerfirst method was more popular , this was only changed after 2000-2003 to a F2L method. So it may take some time.

Considering how quickly M and U triggers can be done, I have no idea why corners first isn't still a viable method. I guess it's more of a "popularity" contest than anything else. David Allen, if he still practices corners first, was quite fast with the method. I wonder what his times are now, Gene Means too.

Chris

The biggest reason corners first isn't as popular is because of look-ahead. For CFOP you use inspection to determine the edges on the bottom and after that you will never have to look for pieces there. For corners first the edges can be everywhere on the cube, including that horrible DB.
Also, MU is only fast for me when I have the cube in a really stirdy grip (for H-Perm or blindfolded edge-orienting). If I have to rotate the cube and follow it by one or more slice turns it will be much slower.
I think these are the main reasons why most corners-first users have stopped using it, or have switched to Roux (some corners, but also some edges, less cube rotations which allows for fast MU)

I don't think a new method is needed for sub 10 or sub 9 averages. Sub 10 averages have already been done by some people and sub 9 averages aren't out of reach for them after more practice/experience.

Pure turning speed isn't the problem. Recognition and unplanned cube-rotations are. Remember that Eriks WR was 42 turns and 7 seconds giving 6 moves per second and he isn't the fastest turner in the world.

 

[Escher]

I can't believe that nobody has mentioned ZZ yet...

I practiced the method for about 2 weeks and got 12 second averages, and my EOLine sucked.

Pros
- 1LLL w/ Phasing (ZZLL) is quite achievable
- <R, U, L> f2l
- Good f2l movecount
- Blockbuilding f2l is fun!
- Very 'modular' - easy to practice substeps
- Expandable - many different options for the LL

Cons
- EOLine is really difficult. I'd estimate it'd take 6 months to a year of solid practice to get decent at.
- At least as a beginner/intermediate, you need to stick to using one cube position for EOLine inspection, potentially missing out on easy EOLines...
- Doing EOLine with a block (like X-Cross) is nigh-on impossible
- You really do need to be/get good with your non-dominant hand. Although it's possible to finish the R side block and then do a z and finish the L block on the U layer, you will still use L moves in the first part and you will rotate twice before the LL...
- Getting used to <R, U, L> blockbuilding and doing it fast is difficult. It is harder to 'automate' blockbuilding rather than 'automate' solving cfop pairs.

 

[4Chan]

I agree with Rowan, if ZZ was refined and practiced to the degree of Fridrich, with 1LLL, it would be the best method for sure.

Darn EOline. d:

 

[TMOY]

For CFOP you use inspection to determine the edges on the bottom and after that you will never have to look for pieces there. For corners first the edges can be everywhere on the cube, including that horrible DB.

The DB edge is not a real problem. At the beginning of the edge phase, simply ignore it, you have plenty of other edges to solve first. and if you need it near the end of the solve, either you already know that edge iif you have paid enough attention or you can deduce it from the position of the other ones.In Roux too, you solve it only at the end of eht solve and this is obviously not a huge drawback.

 

[DavidWoner]

until someone gets a sub-1 on 4x4 with it, I'm not convinced.

Waffle has done this. I don't know if he's actually gotten a sub-1 avg12 yet, but he has been painfully close a few times. I think his redux is usually around 35ish when he is getting good times. M slices on 4x4 are not too hard, but yes, when you get to cubes like 6x6 and 7x7 it really becomes a problem, because STM is no longer viable for analyzing movecount.

For CFOP you use inspection to determine the edges on the bottom/left and after that you will never have to look for pieces there. For corners first the edges can be everywhere on the cube, including that horrible DB.

errr..

 

[miniGOINGS]

I can't believe that nobody has mentioned ZZ yet...

I believe I did...

Cons
- EOLine is really difficult. I'd estimate it'd take 6 months to a year of solid practice to get decent at.

After a couple of weeks, maybe 3 days of actually practice, I'm getting a pretty good handle on it.

- At least as a beginner/intermediate, you need to stick to using one cube position for EOLine inspection, potentially missing out on easy EOLines...

This is the exact reason that I chose ZZ.

- Doing EOLine with a block (like X-Cross) is nigh-on impossible

doesn't really matter...

- You really do need to be/get good with your non-dominant hand. Although it's possible to finish the R side block and then do a z and finish the L block on the U layer, you will still use L moves in the first part and you will rotate twice before the LL...

I'm working on it.

- Getting used to <R, U, L> blockbuilding and doing it fast is difficult. It is harder to 'automate' blockbuilding rather than 'automate' solving cfop pairs.

I would have to dissagree. It may be hard for someone to switch from Fridrich to ZZ, I'm having no problem from Roux to ZZ.

 

[jms_gears1]

More popular? No, Fridrich will forever be known.
Faster? I believe there are already methods in existence that are faster than Fridrich. They (the methods) just haven't been perfected by an individual down to the level Fridrich has.

This.

id say that BigGreen has perfected roux, beyond the level of fridrich.