Is this 2x2 method feasible?

[benjediman]

Hello!

I'm trying to develop a new method (or at least, generate new, good algs) for the 2x2 based off Ortega. It's similar to OFOTA, in a way, but this doesn't involve opposite color faces. Actually, this is kinda the CLL version of OLL, this time it's completing the first same-color face and OLL at the same time. I thought of this when I always noticed two things. One, 90% of the time there is already same-color pairs we can use to build the first step for Ortega. Two, if I plan out far enough, I notice that some of the moves I used to build my first face would actually cancel out with the moves I might use for OLL (like R U' R for the first face but then I notice the OLL is a sune that exactly eats every move I made). And also, there's also the possibility of having more optimal algorithms that may break the available same-color pair but have very short move count (so short, you won't mind it's not fingertrick friendly). I may use ACube to do this, but two things: I do not know how to use the program, and I don't know if I'm just gonna waste my time with this. So I'm asking for your opinions.

So some might think it's so redundant since the same moves can actually be planned out using the current knowledge in Ortega and some good look ahead, but I just kinda want to formalize the whole thing so that it will look like I have a 2-step method for 2x2

 

[AvGalen]

You should first calculate 3 things:

1) What are the chances of having a "same-color-pair" at the start (without this start a mental 1-move start might be needed and wouldn't always be enough)
2) How many possibilities are there for the other 2 pieces of that "same-color-pair"
3) How many positions are there for the orientation of the other 4 pieces (I would guess 8 for orientation just like the regular 8 COLL positions on 2x2x2)

Someone else (you) should do the math, but it seems excessive to me. The 3*40 2-step method is probably easier and as far as I know that already was to much work for everyone to learn

 

[masterofthebass]

If you are talking about doing the last corner and OLL at the same time, then it's already been done. Lucas uses it for 3x3 as "CLS" in his method, but for 2x2, Timothy Sun does that same thing. esk2.awardspace.com is his website. BTW... it's crazy fast.

 

[AvGalen]

He is actually talking about doing all orientation and separtion in 1 step + finish with the 5 Ortega-algs. But he seems to expect that there will be 2 corners already positioned most of the time so in reality he is more like talking about doing the last 2 corners (orient+seperate) and OLL at the same time.

This is much harder (humanly impossible?) compared to the "last corner + OLL" method

 

[benjediman]

Actually, the method you linked is what I was hoping to do If you think about it, the last corner+OLL method is just one or two moves awayfrom the actual algs I was looking for. So this is basically it, I didn't know it was already explored but gut feeling did tell me it had been already. Thanks for the link though!

 

[philkt731]

I've been considering making algs for a somewhat new method that would cover all of OFOTA and SS and it would pretty much cover all of the first two steps of Guimond into one step.

Step 0. Create a face that contains at least three opposite colors (put on D)
Step 1. Orient and separate the colors
Step 2. Permute both layers

Now for step 1, there are three cases with all four opposite colors oriented (OFOTA). For the cases with three oriented opposite colors, lets just say for now, we put the unoriented corner at DFR. Then there are two cases where the three oriented colors are all the same, one with the unoriented corner at FDR and one at RDF (SS). There are also two cases for having color 1 at DFL and DBL and color 2 at DBR. There are two cases for color 1 at DBL and DBR and color 2 at DFL. There are two cases for color 1 at DFL and DBR and color 2 at DBL. So in total there are 11 cases (5 of which are already in known methods)

For Step 1, we need to see how many algs there are for each case. For the three OFOTA cases, there are ~80 algs. For the two SS cases (not icluding OFOTA, counting reflections) there are ~80 algs. For each of the other 6 cases, there are 10 different color variations and 8 different orientations, so each of those cases has ~80 algs. So that is ~640 algs total. This is an overestimate though because for the orientation case of all turned CW or all turned CCW, there are less color cases, because you can AUF, so there are actually a total of just under 600 algs.

I think its a bit of a stretch to make this method. Though I do know around a hundred of these "algs" already, and about half of the rest I can figure by using lookahead on simple 3 move Guimond orientation algs.

If no one has ever seriously thought of a name for this, I came up with one that I like

 

[AvGalen]

I think its a bit of a stretch to make this method

Just a bit of a stretch? You can race Jason for this method compared to ZBF2L+ZBLL (and probably loose)

Edouard Chambon actually just seems to use LBL/OLL/PLL so maybe look-ahead and fast moves are more important than few moves?

I can also come up with a couple of names for this method. Maybe you will like some of them:
OMFG (oh my ..... Frank Morris)
ONLYPHIL (Only Phil)
731-132 (Phil method, just under 600 algs)
ZB++Pocket (More algs then ZB, on a pocket cube)
SWOT

 

[Swordsman Kirby]

I've been considering making algs for a somewhat new method that would cover all of OFOTA and SS and it would pretty much cover all of the first two steps of Guimond into one step.

Step 0. Create a face that contains at least three opposite colors (put on D)
Step 1. Orient and separate the colors
Step 2. Permute both layers

Now for step 1, there are three cases with all four opposite colors oriented (OFOTA). For the cases with three oriented opposite colors, lets just say for now, we put the unoriented corner at DFR. Then there are two cases where the three oriented colors are all the same, one with the unoriented corner at FDR and one at RDF (SS). There are also two cases for having color 1 at DFL and DBL and color 2 at DBR. There are two cases for color 1 at DBL and DBR and color 2 at DFL. There are two cases for color 1 at DFL and DBR and color 2 at DBL. So in total there are 11 cases (5 of which are already in known methods)

For Step 1, we need to see how many algs there are for each case. For the three OFOTA cases, there are ~80 algs. For the two SS cases (not icluding OFOTA, counting reflections) there are ~80 algs. For each of the other 6 cases, there are 10 different color variations and 8 different orientations, so each of those cases has ~80 algs. So that is ~640 algs total. This is an overestimate though because for the orientation case of all turned CW or all turned CCW, there are less color cases, because you can AUF, so there are actually a total of just under 600 algs.

I think its a bit of a stretch to make this method. Though I do know around a hundred of these "algs" already, and about half of the rest I can figure by using lookahead on simple 3 move Guimond orientation algs.

If no one has ever seriously thought of a name for this, I came up with one that I like

Symmetries, please! I don't like the idea of learning so many algorithms just to save up on maybe two moves.

BTW... it's crazy fast.

3.15 average with no lucky solves? Yes.

 

[philkt731]

Haha, I like counting symmetries. I'm really bad at looking at algs for a certain case and then imagining its reflection and doing the reflection of the alg. Then I end up doing it almost all left handed.

My idea for a name was G-FASSST (Guimond - First And Second Steps Squished by Thomas)

Wow Tim, now down to 3.15!!? Boy nationals will be fun

 

[AvGalen]

nice name. I am guessing it should be pronounced as either
a) Go fast
b) Je(zus) fast

 

[Swordsman Kirby]

Haha, I like counting symmetries. I'm really bad at looking at algs for a certain case and then imagining its reflection and doing the reflection of the alg. Then I end up doing it almost all left handed.

My idea for a name was G-FASSST (Guimond - First And Second Steps Squished by Thomas)

Wow Tim, now down to 3.15!!? Boy nationals will be fun

Don't worry, you'll win. I'm terrible in competition.