[ New method]

[Shane Fang]

Hi everyone
I’d like to share a new method of 3x3 Rubik’s cube

The core idea is to skip the down front edge during the Cross and f2l(first two layers),leaving it as a “working slot” to handle EO (edge orientation)more efficiently.

Step 1: f2l without df

Solve the cross without df edge,then,do 4 f2l slots.
Because of the df edge isn’t finished ,so the f2l are more efficiently.

Step 2: EODF

This step is to finish the edge orientation and down front edge.

How to recognize the case?

Step1: Check eo pattern(which edges need to orient?)

Step2: Check the down front edge position (Is it at UF,UB,UR,UL,or DF?)

I developed the eodf system by adapting and optimizing several Roux-style algorithms.Sometimes,the df edge can be finished after finishing f2l,so it would look like Cross,f2l,last layer cross,then zbll.

Total Algorithms:29cases
Average Move Count:8.0 moves(stm)

Step 3: zbll

Because the edge orientation was finished by using eodf,so it will appear a cross on the last layer .To finish the last layer more faster,you can use zbll(if the cross is finished,you can use one algorithm to finish last layer).If you don’t know the zbll,you can use coll +epll,or oll+pll .


P.S.English is not my first language, so please grammatical errors. I will try my best to answer your questions!
I’ll upload some algorithms of eodf as soon as I finish the pdf
2026 3/7 Develop by Shane Fang

 

[Jteddy]

I think this method is very promising and top solvers may start using it.

 

[FenTheGoat]

This lowkey just sounds like freefop or cross -1 with more attention to EO

 

[Caleb Valenzuela]

i think i see some pros and cons
pros:
not solving the DF edge makes the solve have more flexible solutions overall
movecount should be lower as well
cons
switching between outer moves and inner moves then back to outer moves is not very ergonomic, which can mess up your TPS and flow since you have to regrip multiple times for the MU moves and RU moves
you also might as well just do Cross F2L ZBLS ZBLL since your method and ZB essentially do the same thing. The second step of your method solves EO and the DF edge while ZBLS solves EO and the FR pair. This means that ZB is just an overall better version of your method. ZBLS can also be done intuitively if you have a good understanding of how EO works (I use intuitive/2-look ZBLS when i do untimed solves)

TDLR; your method has some potential, but the existence of ZB makes your method look inferior.

PS. Sorry if some of this seemed too harsh/mean, i just wanted to be honest here and it was not my intention to offend anybody.

 

[Shane Fang]

i think i see some pros and cons
pros:
not solving the DF edge makes the solve have more flexible solutions overall
movecount should be lower as well
cons
switching between outer moves and inner moves then back to outer moves is not very ergonomic, which can mess up your TPS and flow since you have to regrip multiple times for the MU moves and RU moves
you also might as well just do Cross F2L ZBLS ZBLL since your method and ZB essentially do the same thing. The second step of your method solves EO and the DF edge while ZBLS solves EO and the FR pair. This means that ZB is just an overall better version of your method. ZBLS can also be done intuitively if you have a good understanding of how EO works (I use intuitive/2-look ZBLS when i do untimed solves)

TDLR; your method has some potential, but the existence of ZB makes your method look inferior.

PS. Sorry if some of this seemed too harsh/mean, i just wanted to be honest here and it was not my intention to offend anybody.

That’s ok.I know my method isn’t too perfect.

 

[Michael Wimberly]

I like this method, I'm averaging 25-27 after a bit of practice (my global is 19)

Here's a few things I'm noticing:

Because the DF edge isn't solved, I find myself often needing to take it out with M moves to make F2L pairs. This can be fixed by inserting a yellow or white edge (ignoring orientation of course) to make sure all the F2L edges are easier to get to.

You have to switch between M moves and outer moves a lot, which makes ergonomics a bit slow.

This method has a lower move count than CFOP, on CFOP solves I average 70-80 moves and with this method I am getting 60-70 mover per solve.

 

[ruffleduck]

I hate to be blunt, but it is very easy to show that this is a worse version of ZZ. The reasoning is very simple: EOcross is not anywhere near 8 moves worse than three-fourths cross. (As an additional point, the 8 moves in question from this method is MU turning which is quite a poor transition from RU(L)F turning and into ZBLL.)

 

[Shane Fang]

I like this method, I'm averaging 25-27 after a bit of practice (my global is 19)

Here's a few things I'm noticing:

Because the DF edge isn't solved, I find myself often needing to take it out with M moves to make F2L pairs. This can be fixed by inserting a yellow or white edge (ignoring orientation of course) to make sure all the F2L edges are easier to get to.

You have to switch between M moves and outer moves a lot, which makes ergonomics a bit slow.

This method has a lower move count than CFOP, on CFOP solves I average 70-80 moves and with this method I am getting 60-70 mover per solve.

Thanks. I think I will fix the problem,so some algorithms may be more faster.

 

[abunickabhi]

Its hard to discover a novel and efficient 3x3 method nowadays yo

 

[Ralph Laurence]

I’ve played around it a little, it’s easier to do DFEO recognition once you observe what the F (sledge) F’ does prior to inserting using M and U moves

To me this feel like a post-petrus, post-cfop method, I think despite that, the DFEO is really an unnecessary step tho when you can just do normal CFOP, I feel like there is more advantages in just settling for normal Advance LBL then doing Cross -1 in DF then EO.

The only justifiable way for me to open that slot, is if it doesn’t just do EO, but rather makes it so there’s an OLL skip, or OLL skip that ensures EPLL, or a LL Skip otherwise, using ZB is just a standard accessory after DFEO.

 

[Hazel]

The bad news:
This isn't a new method; it has been proposed by myself and many others multiple times in the past. Its main issue, as ruffleduck said, is that there are simply better ways of solving EO that have fewer downsides. It also makes some F2L cases quite bad.

The good news:
Method creation is awesome, and the more you mess around with creating new stuff, the better you'll be at it. We all have to start somewhere. My advice from here would be that next time you come up with something, try and ask yourself the question of why that method isn't in use by solvers already (especially if it's a rather simple change to existing methods). It might help poke holes in the method, and you can use that knowledge to make better ones in the future!

Also, new threads aren't necessarily for new methods ideas. You can post them here: https://www.speedsolving.com/threads/the-new-method-substep-concept-idea-thread.40975/

 

[Shane Fang]

The bad news:
This isn't a new method; it has been proposed by myself and many others multiple times in the past. Its main issue, as ruffleduck said, is that there are simply better ways of solving EO that have fewer downsides. It also makes some F2L cases quite bad.

The good news:
Method creation is awesome, and the more you mess around with creating new stuff, the better you'll be at it. We all have to start somewhere. My advice from here would be that next time you come up with something, try and ask yourself the question of why that method isn't in use by solvers already (especially if it's a rather simple change to existing methods). It might help poke holes in the method, and you can use that knowledge to make better ones in the future!

Also, new threads aren't necessarily for new methods ideas. You can post them here: https://www.speedsolving.com/threads/the-new-method-substep-concept-idea-thread.40975/

I apologize about the word “new”. If I can get another chance,I would try to search if someone has already posted or proposed before.

 

[efattah]

The only way I think it could be useful is if you solve DF, while orienting the edges and also either (permuting corners) OR (permuting edges). Then, the ZBLL step has way fewer algorithms, making it more accessible to mid level solvers. It still wouldn't be better than the ZB method though.

 

[efattah]

In that case it starts to look like ZBRoux, where you leave out DF/DB, then solve DF/DB+EO, then ZBLL, which I personally believe is the second best method after ZB.