ECE - New 3x3 Solving Method

[Shiv3r]

Its more of a mix of human thistlewaite and columns than belt and columns. You just construct the columns middle first, and you orient them all
Belt method Is when you solve the belt without any orientation. Belt methods are less efficient than CFOP. ECE is more efficient than roux.

 

[sqAree]

Its more of a mix of human thistlewaite and columns than belt and columns. You just construct the columns middle first, and you orient them all
Belt method Is when you solve the belt without any orientation. Belt methods are less efficient than CFOP. ECE is more efficient than roux.

That sounds quite fanboyish.
I personally think that as ECE builds a belt we can consider it a belt method. Your reason for this not being belt is basically that belt is not efficient while ECE is ; I tend to think of it like ECE is finally an efficient belt method.
The wiki page about belt suggests orienting edges before the belt for example, so I think the concept is free enough to delay the belt a bit and still call it like that.

 

[Shiv3r]

lets please not call it a belt, call it what its really called. you dont just call CFOP "variant on the layer by layer method". You call it CFOP. Please do the same with this method.

Im not trying to be mean, Im just a little defensive of this method.

 

[sqAree]

lets please not call it a belt, call it what its really called. you dont just call CFOP "variant on the layer by layer method". You call it CFOP. Please do the same with this method.

Im not trying to be mean, Im just a little defensive of this method.

You are not mean, no worries, and feel free to defend this method as much as you want; after all it's a great method!
It's just that calling it a belt method is not an insult towards the method.
Let's do a comparison: Calling ECE a belt method is similar to calling CFOP a method that builds the cross first (and likewise that doesn't say CFOP is inefficient just because beginner's also builds a cross).

 

[Shiv3r]

This method is almost like destiny to become a popular method: it has been independently invented by at least 5 people, and it is more efficient than roux. It has also gotten sub-10, and tell me how many of those "new 3x3 speedsolving method" can do that? Especially if you soup this method up as much as you can, you get about 50 max. algorithms, compared to CFOP's 78 minimum.
You've got Roux's alg count mixed with CFOP's consistency and ZZ's "everything is a lucky case". Who else thinks that this thing is gonna be big?

 

[sqAree]

This method is almost like destiny to become a popular method: it has been independently invented by at least 5 people, and it is more efficient than roux. It has also gotten sub-10, and tell me how many of those "new 3x3 speedsolving method" can do that? Especially if you soup this method up as much as you can, you get about 50 max. algorithms, compared to CFOP's 78 minimum.
You've got Roux's alg count mixed with CFOP's consistency and ZZ's "everything is a lucky case". Who else thinks that this thing is gonna be big?

That sounds quite fanboyish.

5 people, really?
But yeah, agreed about this method being good, although I rather root for SSC-O.

 

[Shiv3r]

The Eoline makes it not as good. Also, Shadowslice is collaborating with us on this method, and agrees that this one is probably better.
Even he admits that EZD is maybe the fastest way to do a method like SSC and ECE.
they are pretty much exactly the same, except for the Eoline in the first steps of SSC

 

[Shiv3r]

5 people, really?
But yeah, agreed about this method being good, although I rather root for SSC-O.

Yes, actually 4 known people have made methods around this principle:
Shadowslice E
Crafto22
10 second cube
Morwen Thistlewaite

 

[sqAree]

The Eoline makes it not as good. Also, Shadowslice is collaborating with us on this method, and agrees that this one is probably better.
Even he admits that EZD is maybe the fastest way to do a method like SSC and ECE.
they are pretty much exactly the same, except for the Eoline in the first steps of SSC

I find the best ways to do this for speed are EZD and SSC-O

But yep, those methods are similar to an extent that we can call them "essentially the same". I think as for now both SSC and ECE are just a huge set of possible variations on essentially the same method we'll soon know which one will be the most promising and develope it further.

 

[efattah]

This method is almost like destiny to become a popular method: it has been independently invented by at least 5 people, and it is more efficient than roux. It has also gotten sub-10, and tell me how many of those "new 3x3 speedsolving method" can do that? Especially if you soup this method up as much as you can, you get about 50 max. algorithms, compared to CFOP's 78 minimum.
You've got Roux's alg count mixed with CFOP's consistency and ZZ's "everything is a lucky case". Who else thinks that this thing is gonna be big?

Well yesterday I got a slightly lucky 9.90 with my LMCF method (typical move count is 34-57 with an average of about 41) but it requires at least 220 algorithms and has a steep learning curve that would deter most people. I think ECE will have a far easier learning curve, far less algorithms, and get people fast much quicker, making it (in the future), really popular and effective; however, as it does start with the E-slice, and follows the general pattern of belt methods, I still consider it an advanced belt method. Another reason I consider it that is because when I came across it independently (even before I even heard of 'belt' methods), I 're-discovered' the original 'belt' method(s) and found them inefficient and realized we needed to orient corners of at least 1 layer while we solved the E-slice, so at least when I (partially) developed it, it came about as an advanced variant to the older belt method. However, I don't like the term 'belt' method anyway, as the belt has little to do with the method. The idea of any belt method is not to solve a belt but rather to separate the top and bottom layers into pieces of opposing colors, 'getting rid' of the E-slice pieces so they don't interfere. By means of this separation, we essentially execute a reduction or a reduction algorithm, reducing the possible combinations of the cube dramatically and allowing layer based operations such as orientation to be performed. I would call them reduction or layer separation methods (i.. LSM or something like that).

Eric Fattah
Vancouver, BC

 

[shadowslice e]

Even he admits that EZD is maybe the fastest way to do a method like SSC and ECE.

Some clarification: I think that the last phase of EZD is the nicest way to do the last stage of either ECE or SSC though I still prefer the EoLine first as it is more efficient.

Also, for the last phase, I think for another SSC (ie Eo first) we can separate the edges while solving the corners which would need 48 algs if i calculated correctly though many of these would be inverses of each other so we would need maybe 30 algs for the whole set. It could also lead straight to the easy D algs and from what I've genned, the algs look only very slightly longer than normal.

 

[mark49152]

Are the any videos of anyone solving with this method or SSC? There are a few people talking it up and claiming to have got fast solves, but it would be good to actually see some solves with it. I especially want to know what the sq1 step looks like fast .

 

[Shiv3r]

Are the any videos of anyone solving with this method or SSC? There are a few people talking it up and claiming to have got fast solves, but it would be good to actually see some solves with it. I especially want to know what the sq1 step looks like fast .

Ive been asking for one of these for a while. Amen!

 

[crafto22]

Hey guys, I'm currently working on speed-optimal algs for EZD and another secret alg set that will make ECE EVEN faster and move-optimal. I've been getting 37 or 39 moves with this new alg set, and I'm super excited to share it once it is done. All it requires is the memo of 49 algs and it makes the first step easier and eliminates an entire step! I'm almost finished, I'll share the google sheets doc soon!

 

[crafto22]

Ive been asking for one of these for a while. Amen!

I'll record some solves soon, I'll try to get a sub-13 ao5.

 

[Rouxkie]

Another stupid question... wouldn't it be easier to separate the corners already in step 1 instead? This would save the move-intensive step 3a. At least one could make sure not to have more than 2 wrong corners in D-layer to keep 3a short.

 

[crafto22]

Another stupid question... wouldn't it be easier to separate the corners already in step 1 instead? This would save the move-intensive step 3a. At least one could make sure not to have more than 2 wrong corners in D-layer to keep 3a short.

I'm currently developping algs that will eliminte this issue. I'm going to modify this thread soon, the ECE method is going to be changing pretty drastically.

 

[Shiv3r]

I'm currently developping algs that will eliminte this issue. I'm going to modify this thread soon, the ECE method is going to be changing pretty drastically.

woah woah woah, wasn't the best part about ECE the low alg count?
also the corners are pretty easy to track once you get used to it.
we dont want this to be like the triangular francisco method, where its efficient but nobody uses it because it uses 100+ algs

 

[shadowslice e]

woah woah woah, wasn't the best part about ECE the low alg count?
also the corners are pretty easy to track once you get used to it.
we dont want this to be like the triangular francisco method, where its efficient but nobody uses it because it uses 100+ algs

I think it's fine to add a few algs but keep it less than 50.

If we can work out a nice pattern, step 2 could also be intuitive. What we need is a tool which does a certian set of tasks in a low movecount way.

 

[Shiv3r]

I think it's fine to add a few algs but keep it less than 50.

If we can work out a nice pattern, step 2 could also be intuitive. What we need is a tool which does a certian set of tasks in a low movecount way.

Hey guys, I'm currently working on speed-optimal algs for EZD and another secret alg set that will make ECE EVEN faster and move-optimal. I've been getting 37 or 39 moves with this new alg set, and I'm super excited to share it once it is done. All it requires is the memo of 49 algs and it makes the first step easier and eliminates an entire step! I'm almost finished, I'll share the google sheets doc soon!

Please, don't add all those algorithms. It may just be like 1-look LSE: no one uses it because it can be solved just as fast intuitively and the algorithm count is fairly high.
One of the main lures to this method is the total alg count even when you max the algorithms you need to learn is still much less than CFOP, enough to make a large difference in the time spent on algorithm memorization vs time spent actually lowering your time.
Please don't make it essential to the solution, or people will be turned off by the algorithm count.
Im not saying don't make it, I'm saying keep the original ECE method, its very useful and maybe the best new method ive learned, due to the low alg count yet I can still get within 10 seconds of my main method times.