# SOAP 2x2 Method

#### JustinJ

A while I had an idea for a method where you’d separate colours first, then orient them. I’ve found all the algs for it and I think it has some potential. I don't think this idea has been thought of before.

1. Separate two opposite colours into their own layers, with at least two adjacent pieces oriented. This sounds a bit weird, but basically, you need a bar, and two pieces beside that bar that are the same colour as it.
2. Orient all the pieces
3. PBL

I kind of used an OFOTA style name, so SOAP stands for Separate, Orient All, Permute.

One of the advantages of this method is it has a fairly low alg count. In total, there are 7*8 - 3 = 53 cases, but 7 of those are Ortega, 16 (8 without reflections) of them are SS (easy ones you should learn anyway), and the remaining 30 you need to learn, but 8 of those 30 are reflections, so if you can reflect algs, there’s only 22 cases you need to learn.

This method can be used fairly well as a supplement to Ortega, since it allows for two corners to be twisted in the FL.

One issue I see with this method is that most of the cases are one move away from an SS case, but this does have fewer algs, and is more of an intermediate method than advanced method.

Algs can be found here. I used the MGLS numbering for the odd OLLs, since I didn't know of any other standard naming system. I'm sure there are some better algs for some cases, let me know if you have any good ones and I'll add them.

I'm planning to learn the algs for when I get these cases, since a first step skip is somewhat common, and it'd be nice to be able to force an OLL skip.

Some examples:

I’ll just do the separation, since the rest is obvious.
Scramble: U R2 F' U R' F2 U' F' R2
Solution: U’ R2 or x y2 U2 R’

Scramble: U' R2 U' R U' R U R' F U’
Solution: x2 z U2 R2 or z2 R U2 R

Scramble: F2 U' F2 U' R2 U' F' U F2
Solution: U’ R2 or z2 y U R2

Last edited:

#### Anthony

##### Professional Speedcuber
It took you a while to "announce it."

Anyway, it doesn't look like that many algorithms so I'll consider learning some of it. If you start getting some insane times with it you'll probably be able to convince me to learn it all though. But, I have to wait for my 2x2's to come in first. :/

Oh, like Justin said, if anyone finds a few better algs, post them!

#### Edmund

##### Member
Justin are you going to switch to this? (or just use it as a back-up for easy cases). I think I'll stick with CLL and learn some ss (thats not uber basic). But I'd like to see where it goes.

#### Anthony

##### Professional Speedcuber
Justin are you going to switch to this? (or just use it as a back-up for easy cases). I think I'll stick with CLL and learn some ss (thats not uber basic). But I'd like to see where it goes.
CLL ftw.
Although, this method could be really fast in some cases.

#### JustinJ

Anyone know the average move count?

Also, could you write up some example SOAP solves?
The first step's around 2 moves on average, I think, and the second step is 6.425, Then PBL is 7.8.

So in total it's about a 16.235 move solve on average (I think, correct me if I made any mistakes). Plus AUF comes to 18.485 (for all three AUFs)

I'll write up some example solves in a bit. Edit: I added them to the first post.

Justin are you going to switch to this? (or just use it as a back-up for easy cases). I think I'll stick with CLL and learn some ss (thats not uber basic). But I'd like to see where it goes.
Yeah, it's just for certain cases. CLL is faster than this, but this can be very useful in some cases.

Last edited:

#### Edmund

##### Member
Anyone know the average move count?

Also, could you write up some example SOAP solves?
The first step's around 2 moves on average, I think, and the second step is 6.425, Then PBL is 7.8.

So in total it's about a 16.235 move solve on average (I think, correct me if I made any mistakes). Plus AUF comes to 18.485 (for all three AUFs)

I'll write up some example solves in a bit. Edit: I added them to the first post.

Justin are you going to switch to this? (or just use it as a back-up for easy cases). I think I'll stick with CLL and learn some ss (thats not uber basic). But I'd like to see where it goes.
Yeah, it's just for certain cases. CLL is faster than this, but this can be very useful in some cases.
Thanks for clearing that up for me. Are you learning (or done learning) ss?

#### dbax0999

##### Member
Thanks for the examples. If my 2x2 ever stops breaking and I can finish CLL I'll start looking into this Ugh stupid mini ds mod

#### Cride5

Interesting method, and good job creating the algs! I've put up a page for it on the Wiki here. Its just a stub really, so could do with some extra padding out..

#### JustinJ

Thanks for clearing that up for me. Are you learning (or done learning) ss?
Nah, I think I'll learn this first.

Interesting method, and good job creating the algs! I've put up a page for it on the Wiki here. Its just a stub really, so could do with some extra padding out..
Thanks a lot, I really appreciate it

#### Zarxrax

##### Member
Big bump here, but I feel that the SOAP method has really been overlooked.
I think that this is really an excellent (and fun) method for solving 2x2x2 which more people should take seriously.

I have spent the past couple of weeks researching this method further, and generating better algorithms.
You can read my thoughts about SOAP, including a detailed description of the steps, and a comparison with the SS and OFOTA methods, here.
You can get my algs here.

Some of the algs are the same ones that Justin found, but I didn't really like lots of his algs, so I think some of mine are a lot better. They average 6.2 moves. I think EVERY SINGLE alg is quite fingertricky, and most of them should be possible to sub-1.

I was originally planning to learn all of the algs before I published this, but knowing me, that could take a year. I do intend to learn this though.
As far as I know, no one has learned this full method yet. It would be great to know if anyone else commits to learning this along with me.

#### DavidWoner

##### The Punchmaster
IMO no advanced PreBL method is worth learning in its entirety. For example, this alg is kind of sucks RU2R'U2RU2R'U2R (it's just SS anyway). Separation + 9 moves is too many for PreBL, and there would likely be something better. But a lot of these are nice, I'll learn some of the easier subsets probably.

Though I will say that SOAP is nice in that it's insanely easy to get one-look solves if it's a 2gen case, since predicting PBL is LOLeasy when stuff is separated.

#### Zarxrax

##### Member
Yes, that particular case is kind of bad, but its the longest alg out of the entire thing. And on the positive side, its one of the easiest algs I have ever learned. I seriously did it once, and I will know it for life <3

#### Zarxrax

##### Member
Ok Woner, you've got me stumped here. What does being 2-gen have to do with predicting PBL? I can't figure it out. I see that the 2 pieces on the bottom left wont change... but everything else can still change.
By the way, a LOT of the non-2-gen SOAP algs can be made 2-gen just by increasing their move-count by 1.

#### Cyrus C.

##### Member
2 gen algs don't affect CP. I already know some of these cases from CLS, I think I'll learn more of the easy ones, just for the lucky solves and such. Thanks for bumping Zarxrax.

#### Zarxrax

##### Member
2 gen algs don't affect CP.
Of course they can. Look at this one for example: R'URUR'
It swaps the 2 pieces on the bottom right.