The New Method / Substep / Concept Idea Thread

[Sion]

I came up with an f2b variant that is pretty unique and I find it efficient myself, and there are two steps:

Solve the DL and DR edges.

Then, slove four f2l pairs around them. Acess to the M slice makes this step much faster and efficient.

Then, you have acess to any last steps you want, but I suggest CMLL and l5eop or lse.

 

[Shiv3r]

I came up with an f2b variant that is pretty unique and I find it efficient myself, and there are two steps:

Solve the DL and DR edges.

Then, slove four f2l pairs around them. Acess to the M slice makes this step much faster and efficient.

Then, you have acess to any last steps you want, but I suggest CMLL and l5eop or lse.

FB is considerably faster, and the DR edge+ 2 F2L pairs is what most CFOP usere transitioning to Roux does.
I tried something like this, but got slower times.

Any thoughts on TFOP, the transitional-into-CFOP method I proposed earlier that no one commented on?

 

[gyroninja]

FB is considerably faster, and the DR edge+ 2 F2L pairs is what most CFOP usere transitioning to Roux does.
I tried something like this, but got slower times.

Any thoughts on TFOP, the transitional-into-CFOP method I proposed earlier that no one commented on?

I don't think a transition method to f2l is needed. F2L in cfop is already simple enough.

 

[Sion]

I use this to optimize my pcms so I don't need to worry as much about my S slice pairs.

 

[Shiv3r]

I don't think a transition method to f2l is needed. F2L in cfop is already simple enough.

Yes, but how hard is it for beginners to even understand F2L? I think that as a trainer for CFOP, TFOP may be better just because it teaches the same psuedo-F2L I learned as a beginner.(I placed E edge first, then solved it with a few algs) just those algorithms alone and I learned intuitive F2L by myself.

 

[Sion]

The method is made for an easier refference f2l f2b hybrid as it has no rotations, o e of the other reason why I chose Colums over cfop. But you are right, and the method can easily be used for transition.

 

[mark49152]

i mean, cmon, ECE has gotten sub-10, and Ive seen no other new up and comig method get that low.

Which sub-10 solver uses ECE?

 

[wir3sandfir3s]

Which sub-10 solver uses ECE?

The only one I know of is crafto22, the creator.

 

[sqAree]

Which sub-10 solver uses ECE?

crafto basically.

 

[Shiv3r]

Which sub-10 solver uses ECE?

crafto22 has gotten sub-10 solves, the mark of a good method.

 

[mark49152]

crafto22 has gotten sub-10 solves, the mark of a good method.

I'm not going to go back and read the whole ECE thread again, but I don't recall seeing any worthwhile evidence to validate this as a "good" method. That's not to say it doesn't have potential, but one or two lucky solves doesn't prove it.

The steps to develop a new method should be as follows: come up with an idea, develop it to prove it works, then show the evidence to others; and if you do a good enough job, other people will pick it up too.

In the case of ECE and SSC, there are no videos of proper averages as evidence, and no users of the methods. In the method poll two weeks ago crafto22 voted for CFOP. So the inventor doesn't even rate the idea enough to switch to ECE himself. The methods are just unproven theories.

I don't mean to disparage the methods or the desire to innovate, but I do feel that the SSC and ECE threads are rambling and directionless. They keep popping up to the top with new posts, but the content is just conjecture and more random ideas, and the method development doesn't go anywhere. What they really need is someone to actually commit to their idea, spend a month or two practising, and then make an honest ao12 video (or more) and explain how it works for real, after a serious attempt to develop and use it properly.

I hope the method developers read this as constructive advice, not criticism.

 

[Shiv3r]

I'm not going to go back and read the whole ECE thread again, but I don't recall seeing any worthwhile evidence to validate this as a "good" method. That's not to say it doesn't have potential, but one or two lucky solves doesn't prove it.

The steps to develop a new method should be as follows: come up with an idea, develop it to prove it works, then show the evidence to others; and if you do a good enough job, other people will pick it up too.

In the case of ECE and SSC, there are no videos of proper averages as evidence, and no users of the methods. In the method poll two weeks ago crafto22 voted for CFOP. So the inventor doesn't even rate the idea enough to switch to ECE himself. The methods are just unproven theories.

I don't mean to disparage the methods or the desire to innovate, but I do feel that the SSC and ECE threads are rambling and directionless. They keep popping up to the top with new posts, but the content is just conjecture and more random ideas, and the method development doesn't go anywhere. What they really need is someone to actually commit to their idea, spend a month or two practising, and then make an honest ao12 video (or more) and explain how it works for real, after a serious attempt to develop and use it properly.

I hope the method developers read this as constructive advice, not criticism.

Mark, would you like to try it out? also, according to effatah: the mark of a decent method is when it has gotten a sub-10 solve, even a lucky one. He himself has gotten sub-10 seconds with his own method. the only problem I have with it is that it requires you to know full EG and a lot of other algorithms.

 

[sqAree]

To be honest there exists a pretty decent version of SSC. True, people still think they can do better and create a better post-belt system, but there are at least two fine-enough systems for it already.

I see your point about the missing evidence. As someone who partially helped developing a current version of SSC (all credit to shadowlice who developed like 99% of it!) I have a good overview about the whole method and am convinced it can get as fast as CFOP or Roux. I have a busy semester and no time to commit myself to get fast at another method right now but I actually plan to do that in the summer.

 

[Shiv3r]

To be honest there exists a pretty decent version of SSC. True, people still think they can do better and create a better post-belt system, but there are at least two fine-enough systems for it already.

I see your point about the missing evidence. As someone who partially helped developing a current version of SSC (all credit to shadowlice who developed like 99% of it!) I have a good overview about the whole method and am convinced it can get as fast as CFOP or Roux. I have a busy semester and no time to commit myself to get fast at another method right now but I actually plan to do that in the summer.

The advantage of ECE is the no EOline.

also, Crafto branched off in a different direction with it, with EZD(considered the fastest post-belt strategy) and a lot of other things. Also, Crafto has committed to playing with it
so I talked to shadowslice, he's giving the credit for the ECE method to crafto22 because of the extent he's experimented with it.

what variant is the one youre talking about?

 

[sqAree]

I am purely talking about SSC. The EOline is actually something I prefer over the ECE-variants.

I'm quite skeptical EZD is the fastest post-belt strategy, even for the ECE variants.

I know that SSC and ECE are closely related and at this point I stop caring who of them gets the credit for what. We know that shadowslice created the original SSC method and crafto developed it with the idea of not solving EO directly. I myself only gave an idea for a nice CO system.

As for post-belt methods the original approach of SSC is good enough in my opinion with the advantage of good accessibility for Roux users and a total alg count of 5 (or 8) algs. The other system one could use is the one with Roux blocks and PLL finish, which is even more efficient.

My point was only that we have working systems (for both ECE and SSC variants) and that I agree someone should write a complete tutorial (which I plan to do once I have time).

 

[Shiv3r]

I'm quite skeptical EZD is the fastest post-belt strategy, even for the ECE variants.

crafto22's fastest solve with ECE(8.something seconds) was done with EZD. Its a nice finish, and also the algorithm count is max 16(really 10 algs)

 

[mark49152]

Mark, would you like to try it out? also, according to effatah: the mark of a decent method is when it has gotten a sub-10 solve, even a lucky one.

Yes, I would play around with it just as I have with other methods because I enjoy learning more about how the cube works, but I wouldn't devote time to learning algs or proper practice without more experienced users to tell me I'm not wasting my time. I am not a good enough solver to test it or help with development though. I don't think anyone would be convinced by seeing me solve with ECE .

We are overdue a new method to challenge the big four. It will come eventually, and I'm hopeful to see someone stand up and promote a good idea and develop it into a real contender - but that will take much more than posting an idea and naming it.

I'm not sure I agree with the comment you attributed to @efattah. While it would be difficult to get a sub-10 single with a bad method, the occasional fast single is not enough to qualify a method as good.

 

[AlphaSheep]

I agree with Mark.

Sub 10 singles are meaningless when we don't know the average. What's his average of 100 like? Or even better, average of 1000?

I'd like to propose a solid and measurable sign of a good method: if a person can set their PB average of 100 with it, and get a sub-15 official average with it, then it's a good method.

I do think that someone who devotes effort to SSC/ECE/EZD/ABC/whatever could do this easily, but until that happens, it's all just speculation.

 

[shadowslice e]

I would love for SSC to be much more popular than it currently is but I do see why people have reservations about using it. This is probably why I am trying to get sub-10 with a proven method (roux) first (hopefully this will be sooner rather than later or maybe over the summer). Honestly, at the moment I average around 18-20 with it using a much more basic form as I simply haven't had the time to get around to learning the algs which is about the same as my ZZ times, slightly faster than petrus, slightly slower than CFOP etc. I think you get the point.

When I first proposed the method, it was definitely met with mixed reviews (this was not helped by how I just couldn't do notation) though overall more positive than I was expecting and I did not expect that some point down the line that anyone would become interested in it. It was almost purely hypothetical. I had no knowledge of HTA, kociemba, Orient first etc as I was still much of a noob then (I hope I am better now but sometimes...) and was simply trying to improve on the first i proposed; something like RC-A i may or may not have left the page on the wiki if anyone wants to look. I was intitailly, very unsure of it because i had been speedcubing for a very short time, had joined the forums only a little later and had only run the idea past a couple of people with a bit more experience than me: Berd and Dan Cochrane. Basically, i was a noob then. However, looking back on the method and how it has developed i do think that it has potential. I personally have never said it was an outright "brilliant/fantastic/revolutionary gamechanging method; only that it has potential and it could be very good and have the highest speed potential of any method i have made. The main problem i have to getting faster times is that it is just so different from any other method I've tried (except maybe HTA) because it uses no blockbuilding (actually the while point behind the method) so i have trouble as i have tried to develop my blockbuilding over everyone else.

With all that said, i do believe that SSC cab stand on par with at least Petrus, maybe some of the other methods, potentially even CFOP and Roux (though i think roux is already approaching the best method you can get for 3x3, i sort of agreed with Antoine in this respect in that it may be better than CFOP) though if someone is willing to try without having the backup of another method being sub-10, they are a braver person than i.

Sorry if this was meandering and repetitive, it was a train of thought kind of thing but I hope you can see how I feel about SSC.

 

[efattah]

People often don't realize what a huge uphill battle it is to develop a new method that is competitive; consider that although CFOP was generally in place at the 1982 Worlds, even after 21 years of working on it mostly in isolation, the 2003 worlds saw a winning average of 20.0 seconds. Most methods improve when they are worked on by hundreds of people at once, and refined. One or two people working on a new method face a multi-year struggle to create something fast, not to mention a huge and risky time commitment into a project which will most likely fail. I believe official results are really what matters, so even though I have been working on a new method for 6 months, there's no point in even talking about it until me (or someone else) registers a fast official average, and the same goes for other new methods.

One thing I will mention is that while teenagers have the advantage in terms of hand speed and reaction time, the truth is that all of the major methods were developed by people who were significantly older and 'past their prime' in terms of age related advantages. This is why if the developer of a method can even register a sub-15 official average (at age 30, 40, 50 or whatever), there is promise that a teenager who invested in it could do much better.

Eric Fattah
Vancouver, BC