The New Method / Substep / Concept Idea Thread

[shadowslice e]

Bad recognition

No worse than CP. Plus that's why you predict. I find it comparable to ZBLL recog. Probably easier.

 

[1001010101001]

@shadowslice e
How do you solve CP on inspection?

 

[shadowslice e]

@shadowslice e
How do you solve CP on inspection?

I still use commutators. It' probably better to use teoidus' method though

 

[1001010101001]

I still use commutators. It' probably better to use teoidus' method though

Link teodus’ method pls

 

[shadowslice e]

Link teodus’ method pls

Search for 2gr

 

[SomeRandomZZUser]

Can someone make an example of a full in depth 2gr solve on youtube. Some of the concept are hard to understand but maybe an example solve would help.
-Regards, some random noob on the internet

 

[Neuro]

So I think I've found a way to make EODFDB pretty viable but I'd like opinions on it.

It's very similar to EOLR but has a 2 key differences: you need to pay attention to U/D Centers and the DFDB edged end up on the U layer. Basically the first step is orienting all edges while pairing DF/DB with the D layer center. This line forms on the U layer. After that, you do an AUF and an M2; EODFDB is done!

EODFDB could be done similar to traditional LSE but I think it'd be much slower than this. I think this would be about 2x the "algs" of EOLR but it should be quite learn-able and comparable to the most efficient solutions for all cases. Hopefully the tracking is simple enough that you can predict your edge cycle in ZBLL.

Let me know your opinions. If there's enough demand I'll try and make algs for it.

 

[Arc]

What if you solved CP while making a V, then just do an alg to solve everything else?

That's just 2x2GR and it's probably better. It's more algs though. 22GR was just a fun idea. But I realized it's actually decent for having so few algs.

Or you could just make a v then use an algs to solve everything else

Bad recognition

Definitely better but who has time for so many algs. Recog I think is quite interesting, it's CPLS + CLS. It should just be possible to learn the 6 CP cases per CLS case (minus symmetry) and use CxLL-style recog. Though it is a lot of cases.

@shadowslice e
How do you solve CP on inspection?

http://teoidus.github.io/dump/2GR/CPLine/CPLine.html

Can someone make an example of a full in depth 2gr solve on youtube. Some of the concept are hard to understand but maybe an example solve would help.
-Regards, some random noob on the internet

I think that Teoidus is probably the only one that actually knows 2GR so far. I don't have the quadruples memorized. That said I definitely know enough to be able to fake it for one solve lol, but I'm waaay too lazy to make an example solve. Also if I did an example solve with CP it would definitely be Briggs which is still useful for CP recog/tracking.

 

[1001010101001]

That's just 2x2GR and it's probably better. It's more algs though. 22GR was just a fun idea. But I realized it's actually decent for having so few algs.



Definitely better but who has time for so many algs. Recog I think is quite interesting, it's CPLS + CLS. It should just be possible to learn the 6 CP cases per CLS case (minus symmetry) and use CxLL-style recog. Though it is a lot of cases.


http://teoidus.github.io/dump/2GR/CPLine/CPLine.html


I think that Teoidus is probably the only one that actually knows 2GR so far. I don't have the quadruples memorized. That said I definitely know enough to be able to fake it for one solve lol, but I'm waaay too lazy to make an example solve. Also if I did an example solve with CP it would definitely be Briggs which is still useful for CP recog/tracking.

Cool

 

[shadowslice e]

Definitely better but who has time for so many algs. Recog I think is quite interesting, it's CPLS + CLS. It should just be possible to learn the 6 CP cases per CLS case (minus symmetry) and use CxLL-style recog. Though it is a lot of cases..

Depending on how much you control there would be either 42 or 128 algs

 

[cuber314159]

I have been thinking about a rotationless 4x4 method and came up with this idea:

1. Solve L and R centres
2. Solve RF, RD and RB edges
3. RFD and RBD corners
4. LF, LD and LB edges
5. LFD and LBD corners
6. Finish centres like Yau
7. Solve two sets of three edges, this can sometimes be done by doing r, MU u perm, r', otherwise new algs would need to be developed and I have very little experience with that.
8. CMLL
9. LSE

 

[shadowslice e]

I have been thinking about a rotationless 4x4 method and came up with this idea:

1. Solve L and R centres
2. Solve RF, RD and RB edges
3. RFD and RBD corners
4. LF, LD and LB edges
5. LFD and LBD corners
6. Finish centres like Yau
7. Solve two sets of three edges, this can sometimes be done by doing r, MU u perm, r', otherwise new algs would need to be developed and I have very little experience with that.
8. CMLL
9. LSE

You may like to look up stadler.

 

[1001010101001]

You may like to look up stadler.

Mirrored stadler

 

[YLK RUBIKS]

I am sorry if this has already been asked before but does anti eg-1 exist because theoretically it could I mean if you do the alg lets say for example an easy case like Pi-2 (R' F R2 U' R2 F R ) if you did y2 and did the same alg on Pi-1 + R2 F2 R2 it would solve the cube, would this work and be a viable technique?

 

[1001010101001]

Or you could just do a U2.




My Orient First method(kinda sucks):
1.EO
2.Bottom face opposite colours except for 1 slot
3. Use slot to insert 3 E-slice edges
4. Assemble a corner edge pair and do a Winter Variation to orient corners.
5.Orient everything else using R2/L2/F2/B2 and U/D
6. Half turn everything else

 

[Duncan Bannon]

I am sorry if this has already been asked before but does anti eg-1 exist because theoretically it could I mean if you do the alg lets say for example an easy case like Pi-2 (R' F R2 U' R2 F R ) if you did y2 and did the same alg on Pi-1 + R2 F2 R2 it would solve the cube, would this work and be a viable technique?

I hope I am getting what you are saying. Sure it would work, but I don't think it would be useful. But any insertions that would leave bar in front requiring a y2 could just be inserted in the back. If a U R U' R' insertion left bar in front. Do a y2 then U L U' L'

 

[macncheese]

Just a very rough idea of a method
1. Four 1x2s around the cube
2 Permutate M slice
3 Permutate S slice
4 Last four corners

Hey guys.
I got a draft for my new method! Here are the steps
1. 4 1x2s
2. Edge insertion
3. LL
The first step gives you the freedom to use the M slice. Thus for efficiency use roux pair making styles to do this step.
The second step is very simple. We will solve the D edges. First align a D edge so an F2 can solve it. Then do M' U2 M to solve it. Repeat to solve all edges. Is M slice to orient the edges.
The third step is LL. Use any LL variation to solve it

The first step is intuititve and the last step can use COLL

 

[macncheese]

Hey guys.
I got a draft for my new method! Here are the steps
1. 4 1x2s on the D face standing upright ( basically the size a CE pair)
2. Edge insertion
3. LL
The first step gives you the freedom to use the M slice. Thus for efficiency use roux pair making styles to do this step.
The second step is very simple. We will solve the D edges. First align a D edge so an F2 can solve it. Then do M' U2 M to solve it. Repeat to solve all edges. Use the M slice to orient the edges.
The third step is LL. Use any LL variation to solve it

 

[shadowslice e]

The first step is intuititve and the last step can use COLL

You can't use COLL for the last step. You' have to use L4C. This also looks basically like PCMS but in the wrong order (which handicaps it)

 

[macncheese]

You can't use COLL for the last step. You' have to use L4C. This also looks basically like PCMS but in the wrong order (which handicaps it)

Nvm I uploaded on a separate thread with great improvement Edit- apparently they merged it