I need help for method I thought about (2x2 PBL on a 3x3?)

[Perroh]

So I was sorta playing around with my 3x3 and this random idea of an edges first method came up to my mind because i haven't really heard of anyone using something like it. then I sort of tried to invent the method. But it requires algs that I don't know, they probably don't exist. Here's the method (note that this is just a method to play around with, it has no intention to be used for speedsolving but if anyone knows how to make it faster please take your time to tell me)

1. EO
2.Cross
3. Solve 3 E layer edges
4. Use keyhole to insert 3 bottom corners (don't have to match up with edges, could be any corner for example a white-blue-red corner sandwiched by white-green-orange and white-blue-orange edges.) (to reduce alg count drastically)
5. Pair up which ever corner and edge correspond in the first two layers. Then insert the "pair"
6.Oll
7.Edge Pll
Here's the part that I need help with, this part should be exactly like pbl on a 2x2, same alg count and cases, but I don't know any algs that won't mess up the edges I currently built

---------------------------------------------------------------------------------------------------

Edit: I have to change up some steps. Here's the new method. I'll generate the algs.

- EO
- Cross
- Solve 3 E layers edges
- Use keyhole to insert 3 bottom corners (don't have to match up with edges, could be any corner for example a white-blue-red corner sandwiched by white-green-orange and white-blue-orange edges.) (to reduce alg count drastically)
- Pair up whichever corner and edge correspond in the first two layers. Insert
- Oll
- 3x3 PBL
- Some alg that will finish up the cube (It should just permute Top and bottom crosses)

 

[FJT97]

Try to use cube explorer and genertate some algs for it:
http://kociemba.org/cube.htm

 

[applezfall]

I generated the algs
2bars: U R' U R U2 L U' L' U L' B2 L B2(13htm)
1bar (ON U layer) D2 F R' U' R F' D2 F R' U R F'(12htm)
No Bars:U' B F L R U2 L' R' U2 B' F' U'(12htm
I hope it helps ask me if you need more algs

 

[Perroh]

I generated the algs
2bars: U R' U R U2 L U' L' U L' B2 L B2(13htm)
1bar (ON U layer) D2 F R' U' R F' D2 F R' U R F'(12htm)
No Bars:U' B F L R U2 L' R' U2 B' F' U'(12htm
I hope it helps ask me if you need more algs

Thanks for the algs, they helped out a lot. I also realized that permuting top layer edges goes after 3x3 pbl because if you have to adjust the top cross, when you do the alg you'll get an edge pll. I think I'm going to change up some steps and see what works.

 

[DGCubes]

Here are a couple alternative adjacent-adjacent algs (some don't preserve edge permutation, and some require AUFs at the beginning or end):
R2 U2 R2 B2 U' R2 U' R2 U2 B2
R2 U B2 U' R2 F2 D L2 D' F2
R2 F2 U2 F2 R2 U' R2 U2 F2 U F2 R2
R2 U2 R2 B2 U' R2 U' R2 U2 B2

 

[applezfall]

Here are a couple alternative adjacent-adjacent algs (some don't preserve edge permutation, and some require AUFs at the beginning or end):
R2 U2 R2 B2 U' R2 U' R2 U2 B2
R2 U B2 U' R2 F2 D L2 D' F2
R2 F2 U2 F2 R2 U' R2 U2 F2 U F2 R2
R2 U2 R2 B2 U' R2 U' R2 U2 B2

He said that he wants to preserve edge permutation.And the algs I generated are pretty good and done mess the edges

 

[applezfall]

Thanks for the algs, they helped out a lot. I also realized that permuting top layer edges goes after 3x3 pbl because if you have to adjust the top cross, when you do the alg you'll get an edge pll. I think I'm going to change up some steps and see what works.

Ok if you need more idk why the I will generate some also do you have a YouTube channel ? If so maybe upload some tutorials

 

[Perroh]

Ok if you need more idk why the I will generate some also do you have a YouTube channel ? If so maybe upload some tutorials

Don't worry, I'll generate the algs myself, i dont have a YouTube channel but I'd be interested in uploading some sort of tutorial (assuming it's about my weird method) I could do example solves but that would be a bad idea since I'm not that experienced.

 

[Perroh]

Here are a couple alternative adjacent-adjacent algs (some don't preserve edge permutation, and some require AUFs at the beginning or end):
R2 U2 R2 B2 U' R2 U' R2 U2 B2
R2 U B2 U' R2 F2 D L2 D' F2
R2 F2 U2 F2 R2 U' R2 U2 F2 U F2 R2
R2 U2 R2 B2 U' R2 U' R2 U2 B2

Thanks a lot for the algs (Love your channel btw) It's ok that some of them don't preserve edge permutation since I added an extra step which is edge permutation since I realized the pbl would mess up the edges.

 

[Perroh]

He said that he wants to preserve edge permutation.And the algs I generated are pretty good and done mess the edges

I don't need to preserve edges, just the ones in the e layer. I'll just create an alg set to permute all 8 edges at once. My goal for this method is to make something I can just play around with but it isn't too inefficient and bad. It also seems like a pretty low alg count, but keep in mind I haven't done the edge permutation set yet. It is a total of around 16 algs right now.

 

[Perroh]

I'm making the last alg set right now and I'm having a bit of trouble finding out how many cases there will be and the cases themselves. Do you guys know how to find number of cases? I can figure out the cases if I know the amount that there are. (What the alg set does is permute top and bottom crosses) So far I know that there are at least 24 cases since there are 4 epll algs, then you combine them all which would be 4! Which would equal 24. Probably most of them are mirrors and there are a few extra cases, the set shouldn't reach more than 30 algs.

---------------------------------------------------------------------------------------------------

Edit: Forget what I said about the cases, all the epll cases combined (not including mirrors, it's easier to flip the cube over) are 10. Did the math on a board. Then there are special cases like M2 U2 M2 U2. Then there is one where one cross has two edges adjacent and the other cross has them opposite. Those are three cases. However, you have to apply those cases to every single one of the combined epll cases. I'm probably wrong but the alg count is 30.

 

[Spencer131]

I have 3 topics I want to talk about
1. In your original post when you say to solve the cross, I'm assuming you mean that the cross doesn't have to be permuted?
2. I'm pretty sure someone already generated algs for cpbl and epbl for a different belt method.
3. Maybe it would be more efficient to do f2l instead of keyhole.

 

[Perroh]

I have 3 topics I want to talk about
1. In your original post when you say to solve the cross, I'm assuming you mean that the cross doesn't have to be permuted?
2. I'm pretty sure someone already generated algs for cpbl and epbl for a different belt method.
3. Maybe it would be more efficient to do f2l instead of keyhole.

1. I realized the cross doesn't have to be permuted, the edges have to be oriented and in the face they belong. So if you are doing cross on white then you put the white edges in the bottom but they don't have to be in the right color scheme.
2. I already generated my algs. No need. However if I see better algs I'll need to see the page, and I couldn't find anything about the method, do you know the name?
3. If i were to do f2l, then pbl wouldn't matter, and if I just paired up whichever corner and edge, it would be hard to recognize so keyhole flows more. The whole point of the method is to make the belt, orient both faces (not layers) then permute them in two algs. So im going to give you a quick overview of the method.
EO
"Cross" (edges don't have to be in order)
Put three edges in e layer
Insert corners through keyhole
Oll
2x2 Pbl with algs that don't mess up belt
Permute top and bottom edges.

 

[Mastermind2368]

Well I gave it a try, here are my thoughts

The EO cross is pretty easy but I find myself making the EO-line, I'm sure most CFOP users make the cross fine. I think that the 3 edges into the Middle layer and corners can kida be put together, but not for all cases, I think it would be better to use COLL/ZBLL for the LL so maybe you use 1 alg to solve the bottom layer and a EPLL, That would be 84 algs not including PLL or cases where you have two edges swapped on the U layer and the D layer. If you did Permute top and bottom layer edges, well your in luck. About a year ago, I made a few algs for it, Check them out.

 

[Perroh]

Well I gave it a try, here are my thoughts

The EO cross is pretty easy but I find myself making the EO-line, I'm sure most CFOP users make the cross fine. I think that the 3 edges into the Middle layer and corners can kida be put together, but not for all cases, I think it would be better to use COLL/ZBLL for the LL so maybe you use 1 alg to solve the bottom layer and a EPLL, That would be 84 algs not including PLL or cases where you have two edges swapped on the U layer and the D layer. If you did Permute top and bottom layer edges, well your in luck. About a year ago, I made a few algs for it, Check them out. View attachment 8241

I agree with everything except for the part of COLL/ZBLL because an alg set in the method permutes all corners at once like PBL in a 2x2. Then the last step permutes all top and bottom edges at once. So it wouldn't be necessary to do COLL/ZBLL.

 

[shadowslice e]

The second method is essentially Beginner's SSC. I think the algs you will need for the last step are somewhere in the ECE thread and are listed under EZD
Relevant links:
SSC wiki page
ECE thread
SSC thread

Incidentally, this method still requires some improvement before being truly a speed method. It may be possible to achieve very good averages but the current consensus and feeling in the community is that the final step (after all edges and corners are oriented) need much improvement before it reaches the level of the first 2 so I would be very grateful if you had a look over it as it may well require newer eyes to find the best way to solve it.