Alternative to f2l+Oll

[JeffDelucia]

A few days ago I learned all of the zbf2l cases for when you have the corner correctly oriented and in position and the edge in position but flipped. I noticed there was only two cases for this. Recently when I get an f2l pair with the edge flipped instead of correcting it I have been just inserting it incorrectly and finishing the f2l. After that I have been using the zbf2l case and getting an Oll with all of the edges oriented. I noticed that if you came up with algorithms that would flip the edge at FR and orient the top layer you could have much more freedom during f2l than you would with standard fridrich.

For example if at any point in your f2l you came across a pair that had a flipped edge you could just insert it and finish the rest of your f2l like normal. Or you could make a pair with a flipped edge in situations when it would be easier than making the pair correctly. If it's more convenient to make the f2l like normal you could make the pair with the flipped edge at the last slot.

I really hope I was clear because I see great potential in this method. I was also wondering if anyone could help me out with calculating the number of algorithms for this and also maybe helping generate some algorithms for the cases as I have no experience creating algorithms.

 

[Sa967St]

or you could just use MGLS

 

[JeffDelucia]

or you could just use MGLS

Yes but my goal isn't to not use fridrich my goal is to have a potentially faster method...

 

[Sa967St]

or you could just use MGLS

Yes but my goal isn't to not use fridrich my goal is to have a potentially faster method...

I was kidding

 

[JeffDelucia]

or you could just use MGLS

Yes but my goal isn't to not use fridrich my goal is to have a potentially faster method...

I was kidding

Was I clear in explaining my method? I wasn't sure if anyone else could understand me haha.

 

[mr. giggums]

This is kind of like EJF2L. Only with edges instead of corners.

 

[JeffDelucia]

This is kind of like EJF2L. Only with edges instead of corners.

I realized that as I was typing it and as ejf2l yields itself toward petrus users this yields itself toward fridrich users

 

[4Chan]

A few days ago I learned all of the zbf2l cases for when you have the corner correctly oriented and in position and the edge in position but flipped. I noticed there was only two cases for this. Recently when I get an f2l pair with the edge flipped instead of correcting it I have been just inserting it incorrectly and finishing the f2l. After that I have been using the zbf2l case and getting an Oll with all of the edges oriented. I noticed that if you came up with algorithms that would flip the edge at FR and orient the top layer you could have much more freedom during f2l than you would with standard fridrich.

For example if at any point in your f2l you came across a pair that had a flipped edge you could just insert it and finish the rest of your f2l like normal. Or you could make a pair with a flipped edge in situations when it would be easier than making the pair correctly. If it's more convenient to make the f2l like normal you could make the pair with the flipped edge at the last slot.

I really hope I was clear because I see great potential in this method. I was also wondering if anyone could help me out with calculating the number of algorithms for this and also maybe helping generate some algorithms for the cases as I have no experience creating algorithms.

>A few days ago I learned all of the zbf2l cases

>My face:

But then I read
>for when you have the corner correctly oriented and in position and the edge in position but flipped. I noticed there was only two cases for this.

This image contains all of my reactions, especially the last frame.:

Spoiler

 

[JeffDelucia]

A few days ago I learned all of the zbf2l cases for when you have the corner correctly oriented and in position and the edge in position but flipped. I noticed there was only two cases for this. Recently when I get an f2l pair with the edge flipped instead of correcting it I have been just inserting it incorrectly and finishing the f2l. After that I have been using the zbf2l case and getting an Oll with all of the edges oriented. I noticed that if you came up with algorithms that would flip the edge at FR and orient the top layer you could have much more freedom during f2l than you would with standard fridrich.

For example if at any point in your f2l you came across a pair that had a flipped edge you could just insert it and finish the rest of your f2l like normal. Or you could make a pair with a flipped edge in situations when it would be easier than making the pair correctly. If it's more convenient to make the f2l like normal you could make the pair with the flipped edge at the last slot.

I really hope I was clear because I see great potential in this method. I was also wondering if anyone could help me out with calculating the number of algorithms for this and also maybe helping generate some algorithms for the cases as I have no experience creating algorithms.

>A few days ago I learned all of the zbf2l cases

>My face:

But then I read
>for when you have the corner correctly oriented and in position and the edge in position but flipped. I noticed there was only two cases for this.

This image contains all of my reactions, especially the last frame.:

Spoiler

At first when I was reading that I was like "Oh noes I forgot to write it was only for those cases" but then I read on and I'm like "Oh"

 

[mr. giggums]

There would be 54 cases for this. 27 for 1 edge flipped on the layer layer and another 27 for 3 edges flipped on the last layer.

 

[JeffDelucia]

There would be 54 cases for this. 27 for 1 edge flipped on the layer layer and another 27 for 3 edges flipped on the last layer.

Not much less than Oll but I'm pretty sure this is faster

 

[Feryll]

Sounds nice, but I think OLL may be simpler, and less constricting than your idea.

Spoiler

If I get what you're saying here, then I sort of had a similar idea to help 2 look OLL solvers. You solve cross, one edge flipped, solve f2l normally, at OLL there is either 3 edges flipped or just 1, and you can fix them by either doing M' U M U' M' U2 M (3 flipped on top) or y2 U F2 M2 U M U' F2 U' M (1 flipped on top). It will avoid dot OLLs, and what I find nice is that it preserves corner orientation so as to be able to look ahead to the 2LOLL case. But yeah, the second alg is pretty awful, so I don't put much hope in my idea.

 

[vcuber13]

There would be 54 cases for this. 27 for 1 edge flipped on the layer layer and another 27 for 3 edges flipped on the last layer.

Not much less than Oll but I'm pretty sure this is faster

I only get 52 since 6 of them have 4 and the H has 2 since there is 2 symmetries.
\( 2(6\times4+2)=52 \)

 

[mr. giggums]

There would be 54 cases for this. 27 for 1 edge flipped on the layer layer and another 27 for 3 edges flipped on the last layer.

Not much less than Oll but I'm pretty sure this is faster

I only get 52 since 6 of them have 4 and the H has 2 since there is 2 symmetries.
\( 2(6\times4+2)=52 \)

+2 more for when you have solved ocll.

 

[JeffDelucia]

Sounds nice, but I think OLL may be simpler, and less constricting than your idea.

Spoiler

If I get what you're saying here, then I sort of had a similar idea to help 2 look OLL solvers. You solve cross, one edge flipped, solve f2l normally, at OLL there is either 3 edges flipped or just 1, and you can fix them by either doing M' U M U' M' U2 M (3 flipped on top) or y2 U F2 M2 U M U' F2 U' M (1 flipped on top). It will avoid dot OLLs, and what I find nice is that it preserves corner orientation so as to be able to look ahead to the 2LOLL case. But yeah, the second alg is pretty awful, so I don't put much hope in my idea.

This is also 2 look friendly because you only need 2 different algorithms for the cross and then a Corner oll also this is less constricting than regular f2l.

 

[FatBoyXPC]

Jeff I think you'd like VHF2L better if it weren't for the fact that for it to be optimized you need to learn the algs + mirrors for both angles (front and back). I really looked into it and since all OCLLs are really fast it just made sense to me that I should learn it, but I didn't feel like learning 30 some cases just due to mirrors.

In regards to what you want though, you should probably make a video to make it a bit clearer. I think I understand what you are saying though, but I think that it only adds time to your F2L since you have to go back and fix it eventually (unless you feel like learning ZB but you know what I mean).

 

[JeffDelucia]

Jeff I think you'd like VHF2L better if it weren't for the fact that for it to be optimized you need to learn the algs + mirrors for both angles (front and back). I really looked into it and since all OCLLs are really fast it just made sense to me that I should learn it, but I didn't feel like learning 30 some cases just due to mirrors.

In regards to what you want though, you should probably make a video to make it a bit clearer. I think I understand what you are saying though, but I think that it only adds time to your F2L since you have to go back and fix it eventually (unless you feel like learning ZB but you know what I mean).

Ehh no Vhf2l is not what I'm looking for. Baisicly what I've created here is ejf2l using disoriented edges instead of corners.

 

[FatBoyXPC]

So is the only advantage to using this the fact that it might be easier to put in a flipped CE pair rather than a correct CE pair, and then you learn algs to OLL while fixing that last slot/pair?

 

[JeffDelucia]

So is the only advantage to using this the fact that it might be easier to put in a flipped CE pair rather than a correct CE pair, and then you learn algs to OLL while fixing that last slot/pair?

yes. Also it has 3 less algorithms than Oll but thats not many..

 

[FatBoyXPC]

So are you going to purposely misalign a pair so you won't have to learn OLL? Don't forget that 10+ OLLs are INCREDIBLY easy and fast (Sune, AntiSune, a few dot cases, both T shapes, two B shapes, etc)