[Help Thread] CPLS and 2GLL discussion

[StachuK1992]

Hello, all.
I just thought I'd introduce these two together, mainly because of their close relation to each other.

I recommend looking at the CPLS page first, and then the 2GLL page.

I did not invent/create either method/sub-step; I just think that they haven't been published very well (especially CPLS) and thought it would be nice to give them a bit of a boost, so I went ahead, found/documented a bunch of algs, made some print-out pages, and made the wikis.

Right now, pretty much all of my stuff for both of these is geared for OH, but 2H-specific algs shall come eventually if requested.

Neither of these pages are perfect, or even complete for that matter - far from it! However, I think that this should help get some people interested.

Some stuff not noted in the wikis that probably won't interest you:

Spoiler

What I plan to do after learning CPLS/2GLL:

• For C cases, just do COLL for non-sune/anti-sune cases, and just (anti)sune then PLL for those.
• For I and Im cases, I plan to just do CLS, then PLL. I and Im cases are very easy and simple to memorize. One can only learn 8 of them and be able to successfully sub2 any of these cases, with recognition, very very easily.
• For -/+/O cases, if I get a case that is an adjacent switch in the back (UBL, UBR), I know that a 2-gen CLS will reduce the case to EPLL.
• If the aforementioned adjacent switch is on the left (UFL, UBL), then I plan to do a y' rotation, then a CLS alg from that angle, also resulting in an EPLL.
• For I, Im, and C cases, if you come across them and want to do an 'intuitive' CPLS, then you can simply Niklas your way out of an adjacent swap on the right or F triple-sexy F' for a diagonal, then proceed to either CLS or simply finish the rest with other 2-gen awesomeness.
• CPLS can be a pain to recognize at first, but once you get used to searching fast for the three stickers and comparing them to your color scheme, it gets okay - I've been on this for only a few night (I haven't memorized anything yet, though) and it's working pretty nicely.
• This seems to have much more potential for OH than for 2H. Actually, I think it has LOTS of OH potential, if the CPLS recognition is relatively fluent.
• I'd like to see how color-neutral users deal with this. Only opposite-neutral myself, I'd say that simply knowing which side I'm ending up on allows me to recognize either case just as fast as the other.
• 2GLL algs feel awesome both OH and 2H.
• 2GLL algs average ~13 moves from the look of it, and decent CPLS algs average ~10 moves. If you were to look at the OCLL+PLL move-count average, or even the OLL+PLL move-count average, then I think you'll be surprised by the closeness. And this even does some F2L!
• 2-Gen algs are a pain to learn; they take me ~150% of the time it takes me to learn a "regular" alg. I figured this out when learning CLS - it's just hard to differentiate between algs.
• 2GLL cases may take a bit to recognize. It's essentially looking for OCLL then EPLL.
• TOAD!
  [*]YOU CAN 2-GEN THE LAST F2L CORNER+LL 1/3 OF THE TIME!
• I'd love to hear responses.

CPLS
2GLL

My unfinished CPLS and 2GLL alg print-ables: .doc .pdf
Right now, I have at least one alg for every 2GLL case (although I've found a few cases that seem to be wrong ) and I have both RH and LH OH algs for -, +, and O cases. I will get to adding I and Im soon, and update this post and the wiki with those.

If you feel anything needs to be or should be added, please say so.

Have fun!
Statue

 

[riffz]

Still waiting for CPLS recognition...

I really hope more people will learn this and practice with it. Leaving as much as possible to the end for 2-gen seems like a really good idea.

As for me, I'm terrible at OH so I don't intend to learn this.

 

[StachuK1992]

Still waiting for CPLS recognition...

I really hope more people will learn this and practice with it. Leaving as much as possible to the end for 2-gen seems like a really good idea.

As for me, I'm terrible at OH so I don't intend to learn this.

I explained CPLS recog in the wiki a little bit...

Recognition

Recognition is definitely the most prominent downfall to this method, at least in terms of personally finding an intuitive way to do so. However, a system has been found that works very well.

Before you start, it must be emphasized that you *must* know your color scheme fairly well.


Let's set up a case. Do x R' U R U' R' U R U' R' U R U' x' onto a scrambled cube with yellow on top and red on front.

Seeing this, I know that this is an O case, and that it's already positioned correctly, in URF, to be recognized.

Next, I need to look at the remaining three U-layer pieces.

Let's keep a system to *always* start with UFL.

Once reaching any of the three corners, in this order, the first goal is to locate the U-layer sticker, in my case yellow. This should be on the top for this case.

Next, look at the sticker *clockwise* from this sticker. In other words, the sticker on the F face, or the FLU sticker. This should be red. I note this, and that my red FACE is currently on B. I now pretend that the entire F face is orange, because when rotated, that sticker lies on that face.

Following that, do the same for the LUB and BRU corners.

For the L face, you should come out with green, since the sticker at LUB is green.

For the B face, you should come out with orange, since the sticker at BRU is orange.

One could write this case down as an [O BLF], being as though B is on F, L is on L, and F is on B. Depending on initial AUF, however, this may change

So, from the above, we can visualize that F = red, L = green, and B = orange. Knowing that yellow is our U-layer sticker, we know that the above can't be right - those colors aren't allowed to be like that! So what we must do is switch those "face." To do this, we must switch the corners that these pieces are represented by. For example, the UFL piece represents the Front face. In this case, we must switch the UFL and UBR corners. To understand what to do now, we go over to https://sites.google.com/site/devastatingspeed/3x3x3/cpls and try to find the right case. Knowing that we have a diagonal switch (switching UFL and UBR has essentially the same effect as switching UFR and UBL), we look for the O case with the diagonal swap, which is in the 3rd row, second column, yielding us the algorithm x (U R' U' R)*3 x'

After applying this, the F2L should be finished, and the LL should be reduced to a 2GLL case, to be done in however many steps. More than two steps makes CPLS very unnecesarry and arguably a waste of time, though.

I plan to add a bit more, but I thought that would be enough to start.
I'll add a bit more of an explanation later, and a video as well to this thread.

I'm very underwhelmed with the response to this thread. :/

 

[Chapuunka]

Like I said before, CPLS looks nice. But after looking at recognition, I'm even more undecided: would you (or anyone else with more knowledge than me on the subject) recommend CLS or CPLS (with very little hope of learning 2GLL)? I would use it for 2H ZZ.

 

[StachuK1992]

If you've little hope for learning 2GLL, then don't bother with CPLS.
CLS is nice and recommended for ZZ.

CPLS would be nicer for OH rather than 2H, because 2-gen REALLY helps you more in OH rather than 2H now.

If you plan on learning 2GLL, then CPLS could be really nice. I'll let you know in like a month how it turns out for me.

Also, CLS and CPLS work very nicely together, which I partially mentioned in the spoiler. I plan on using both.

 

[oprah62]

Statue

 

[Chapuunka]

Ok, thanks for all the help. Maybe someday I'll be really motivated to learn insane amounts of algs like you, but for now this should be more than enough for me.

 

[riffz]

Still waiting for CPLS recognition...

I really hope more people will learn this and practice with it. Leaving as much as possible to the end for 2-gen seems like a really good idea.

As for me, I'm terrible at OH so I don't intend to learn this.

I explained CPLS recog in the wiki a little bit...

Oh, sorry. I glanced over it and forgot to read the wiki article.

It looks as though using a fixed colour scheme when using CPLS would be a better idea, now that I've read the recog. (at least choosing a colour pair to always be on L/R)

And yeah, this thread deserves more replies.

 

[StachuK1992]

Ok, thanks for all the help. Maybe someday I'll be really motivated to learn insane amounts of algs like you, but for now this should be more than enough for me.

What are you talking about?
I don't even know full OLL!

I explained CPLS recog in the wiki a little bit...

Oh, sorry. I glanced over it and forgot to read the wiki article.

It looks as though using a fixed colour scheme when using CPLS would be a better idea, now that I've read the recog. (at least choosing a colour pair to always be on L/R)

And yeah, this thread deserves more replies.

Yes, choosing a consistent color scheme would probably benefit a user a lot.
However, there should be no reason to have the same face on F or B or L or R or whatnot - you should know how everything relates.

 

[PhillipEspinoza]

So this is essentially the "missing link" part of ZZ-d. But instead of doing it when building the first block, you permute the corners during the last slot.

Hm.

THIS IS EPIC!

Why aren't people noticing this?

Ah it's okay. Their loss, when we'll be the ones with sub-15 OH avg's. Bwahaha.

 

[StachuK1992]

So this is essentially the "missing link" part of ZZ-d. But instead of doing it when building the first block, you permute the corners during the last slot.

Hm.

THIS IS EPIC!

Why aren't people noticing this?

Ah it's okay. Their loss, when we'll be the ones with sub-15 OH avg's. Bwahaha.

Essentially.
Also, something awesome to take note of that I don't think people have.
In 1/3 of your solves, You can do the last F2L corner+LL all with 2-gen, and can recognize if you have one of those cases as you recog CPLS. How is that not awesomely awesome.

The movecount is really nice, too, and I haven't even found good algs for half of these!

I'm not sure why people aren't noticing this - I know some people think that the method is stupid / rawr) and others are just not bothering with it because they haven't seen fast times with it yet.

I want some fast OH people to learn this - that would be so nice! (kitty is waiting until after pton)

Philip - are you planning on learning this?
I'm just quick fixing up my OH PLLs for cases where I think it's better just to use 'regular' solving. Then off to CPLS.

 

[Matt S]

I'm not sure why people aren't noticing this - I know some people think that the method is stupid / rawr) and others are just not bothering with it because they haven't seen fast times with it yet.

I think people are noticing it, but the recognition is still so daunting. I'd be very surprised if this system was more attractive than CLS + PLL, especially for people who already have done a lot of work at improving PLL.

Side-note: As someone who doesn't particularly care about one-handed solving, I've become a little disenchanted with pure 2-gen. Don't get me wrong, 2-gen algs are quite nice on average. However, there are a lot of nice 3-gen algs. My T-Perm has the highest tps of all my PLLs. And the <RUD> and <LUD> algs for the -0 and +0 CLS cases are pure awesome.

 

[StachuK1992]

I'm not sure why people aren't noticing this - I know some people think that the method is stupid / rawr) and others are just not bothering with it because they haven't seen fast times with it yet.

I think people are noticing it, but the recognition is still so daunting. I'd be very surprised if this system was more attractive than CLS + PLL, especially for people who already have done a lot of work at improving PLL.

Side-note: As someone who doesn't particularly care about one-handed solving, I've become a little disenchanted with pure 2-gen. Don't get me wrong, 2-gen algs are quite nice on average. However, there are a lot of nice 3-gen algs. My T-Perm has the highest tps of all my PLLs. And the <RUD> and <LUD> algs for the -0 and +0 CLS cases are pure awesome.

Yes, 3-gen can be great, but for OH, I know a lot of people prefer 2gen algs that they can spam, myself included.

I think a video of recog may be needed. I need to get to that soon.

Yes, many people have already done so much work on PLL, and it's hard to get people to switch from what they know. :/
Oh well, we'll see how it goes.

Thanks,
statue

 

[a small kitten]

Yes, many people have already done so much work on PLL

Hey! I haven't! =D

I'm probably going to learn this after pton fall if I can understand the recog.

 

[StachuK1992]

Yes, many people have already done so much work on PLL

Hey! I haven't! =D

I'm probably going to learn this after pton fall if I can understand the recog.

I must make video.

 

[a small kitten]

Make sure you use a normal color scheme too.

 

[StachuK1992]

Of course.
My scheme would confuse people.

 

[PhillipEspinoza]

I want some fast OH people to learn this - that would be so nice! (kitty is waiting until after pton)

Phillip - are you planning on learning this?

Ya I really wish Weston would take a look at this considering how fast is tps is for 2-gen algs and whatnot.

The way I see it is, (without the EOLine) I avg around 4 seconds, if not sub-4 for the first 2x2x3 block (this is not even with the option of putting in a pair the wrong way. I imagine it would be low 3 if this were the case and with much more practice. I just learned ZZ yesterday.). Then, I avg around 8 seconds for 2-gen scrambles. That's already 12 seconds! Add in some recog time and a good EOLine for 3 seconds, that's still a 15 second average! I imagine people like Weston could probably get sub-15 avgs if they tried this.

 

[joey]

I get the feeling Phillip doesn't understand..

 

[a small kitten]

Don't worry. Phillip gets this.