# OSPA(New 2x2 Method)

#### emolover

##### Member
This thread is for opinions, improvements, and criticism of a new method for 2x2 I have created.

OSPA stands for Orient Separate+Permute AUF. Essentially what this method is, is you use the SOAP algs to orient opposite sides that contain opposite colors in each. This is a lot like step one of Guimond where you make a side in the bottom with two opposite colours and step two of Guimond where you use an OLL to orient the top side. But this is done in one step instead of two. The next step is to Separate and Permute the opposite color sides in one algorithm. This combines step 3 of Guimond which is separation of opposite colours into two different sides, then the PBL that is the last step of Guimond and Ortega. This combines them into one algorithm. The only thing after that is the AUF which is able to be predicted.

Seeing that I suck at explaining, these 5 example solves should help you understand how this method works.

Scramble: R2 F2 R F2 U' F U2 F2 R'
Presolve: x2 y'
Orient: U' y' R' U' R U R' U' R
Separate and Permute: z2 y (U R2 U R2' U' R2 U' R2') U2
__________________________________
Scramble: R' F2 U2 F' R2 U' R U2
Presolve: x2 y
Orient: R U' R' U R U' R'
Separate and Permute: y' U' R U' R' U' R2 F2 U' R U R
__________________________________
Scramble: F R' F2 U' F2 U R' U R'

Pre-solve: z' y2
Orient: U' (R' U R' U2 R U' R2)
Separate and Permute: U' R U F' R2 F R' F
__________________________________
Scramble: U' R2 U F' U R U' F R'
Pre-solve: y2
Orient: U (R U R' F R F')
Separate and Permute: y U2 (R2 U' R U F' R2 F R' F) U2
____________________________
Scramble: R' F' R F R2 U' F2 R2 F'
Pre-solve: x' y2
Orient: U' (R' U R U' R' U R) y'
Separate and Permute: U (F2 U R2 U' F2 U R2)

The pros things about this method is that it is only two steps, has between 80-90 algorithms which is less then EG, easy to one look as long as you know what your SOAP algs do, heavy use of the double move(R2, U2...) if you like that, and can have absolutely no intuition or as much of it as you want.

The only two cons with this method that I see is recognizing the Separate+Permute case and heavy use of double moves if you don't like that.

Solutions to those could be establishing a recognition system where you realize that opposite colours are the same and the same colours are opposite if you are looking at two different opposite color top(or bottom) stickers. And algorithms that have almost only single move moves(F R U R"...).

As I said before I would like your opinions and suggestions, but please realize I am not done with this method. Also feel free to ask me questions and I will most likely answer them.

Last edited:

#### HelpCube

##### Member
Hmm... Pretty cool. Could be really effective. And just so you know, on the second example solve there is no scramble.

#### Kirjava

##### Colourful
similar to a bunch of other random 2x2x2 methods you don't really hear about because second step recog is unfeasable

you may aswell document it though so we have algs floating about

#### AustinReed

##### Member
Seems kinda cool. However, I think the first step takes more moves on average than the first step of EG-1. Because of that, I'll stick to EG.

#### Zarxrax

##### Member
Is the separate+permute done intuitively or do you actually have algs for that?
I had considered combining seperation and permutation steps a couple of years ago, and someone shot it down telling me that it has a HUGE number of algs. Or is this somehow simplified so that the alg count is reduced?

#### emolover

##### Member
Seems kinda cool. However, I think the first step takes more moves on average than the first step of EG-1. Because of that, I'll stick to EG.

The algs I used in the example solves were SOAP which is meant to make two sides of one color each. I am sure if better algs would be generated, it would be more efficient.

Is the separate+permute done intuitively or do you actually have algs for that?
I had considered combining seperation and permutation steps a couple of years ago, and someone shot it down telling me that it has a HUGE number of algs. Or is this somehow simplified so that the alg count is reduced?

Both. It would take about 30 algs.

#### PandaCuber

##### Member
Ill make this my main method cause I hate CLL. Just so boring.

Can you get a PDF or something set with algs so we(community) can start learning?
Lol

#### emolover

##### Member
Ill make this my main method cause I hate CLL. Just so boring.

Can you get a PDF or something set with algs so we(community) can start learning?
Lol

Thank you.

That is what I do not have done and I might need some help generating algs for. I have used this which is not a good generator but I do not know of any good 2x2 generators.

#### Zarxrax

##### Member
I also used ksolve when I was generating algs for soap, but I dont remember if it would work for detecting opposite colors. I think it can though.

#### PandaCuber

##### Member
Thank you.

That is what I do not have done and I might need some help generating algs for. I have used this which is not a good generator but I do not know of any good 2x2 generators.

Something you need is a Video Tutorial.
I know im saying a lot, but these are just suggestions. And cause I like videos...

#### Jaycee

##### Member

Like I said in the PM, I'm just waiting for algs and I might try to use this alongside CLL or just make it my main

#### emolover

##### Member

That is very nice of you but you don't even know the cases for SP(Separate and Permute) so that wouldn't work out for you. But if you could help me with some opposite color SOAP algs(and find a name for what I am talking about) that would be really nice and helpful.

Something you need is a Video Tutorial.
I know im saying a lot, but these are just suggestions. And cause I like videos...

Remember this thread is just for suggestiong, help and stuff. I am not done, but I will when everything is finished.

I will do some major work on OSPA tomorrow.

#### Tim Major

##### Platinum Member
Difference from Guimond?

#### aronpm

##### Member
Very likely faster and is only two steps instead of four.

Except not. Go back and read Kirjava's post.

#### Tim Major

##### Platinum Member
Very likely faster and is only two steps instead of four.

Oh my bad, I basically looked at the initials' description and read it as orient, separate, permute.
I thought the + was "and then" rather than just "and".
Anyway, I've forgotten a lot of CLLs, I'll look into this when you post algs.

#### emolover

##### Member
Except not. Go back and read Kirjava's post.

I see Kirjava's post and it says nothing about speed, only recognition. Even so I will still finish this, document it, learn it my self, then prove that it can be fast.

#### Ickenicke

##### Member
I will take a better look at this method later today, but it seems like a method that I would be intersested in using.

I have just learned som CLL cases, and like Pandacube said, it is very boring. If I can help with something, please let me know.

#### emolover

##### Member
If I can help with something, please let me know.

Same thing I told PandaCuber, see if you can make up some good OOC algs(orient opposite color) that are efficient and easy to execute. Does not have to be 2gen but can be 3gen using LUR or RUF.

Also if you find a good 2x2 solving program, that would be wonderful.