[Unofficial] 15.827 Ao12 using ZZ-CT: The LL-Skip Method.

[4Chan]

Oh, I actually also don't know how many cases there are.
I said 108, because I thought someone mentioned it previously.

I use a few tricks and stuff, so I don't actually know how many cases I know, it's more of the same "qualitative" information that I have. But the other probabilities are calculated!

But for those cases, I don't actually use multiple triggers.
I actually use a bunch of RUD algs, I'll make an addendum in my original post when I get back!

 

[AlphaSheep]

U R U' R' U R U' R' U2 R U R'
U2 R U R' (Bet this makes you feel like an idiot)
R U R' U2 R U R' U R U' R'
R U2 R' U R U R' U R U2 R' U' R U' R'

Sorry, I should have made it clear, those are the setups to the cases, not the algs that solve the cases themselves.

 

[4Chan]

For the second one, I use R2 D R' U' R D' R2.
I think I have decent solutions for the rest, but I haven't checked them yet!

I'll post my solutions for those later, but I have a comp in 4 hours LOL

 

[gyroninja]

Sorry, I should have made it clear, those are the setups to the cases, not the algs that solve the cases themselves.

In that case
R U' R' U R U2 R' U' R U R' U R U R'
R2 D R' U' R D' R2
U2 F' R U R' U' R' F R
R U2 R' U R U R' U R U2 R' U' R U' R'

Those are the algs I have on my collection.

 

[sub20oneday]

this looks so much cooler than coll epll so before i start ive got a few questions
1.where can i generate algs
2.hows the recognition
3.before i learn all the algs are there any ways i can use the method like 4lll with cfop before 2lll

 

[gyroninja]

this looks so much cooler than coll epll so before i start ive got a few questions
1.where can i generate algs
2.hows the recognition
3.before i learn all the algs are there any ways i can use the methods like 4lll with cfop before 2lll

1. On your computer with CE or acube.
2. Pretty good, tsle recognition is pretty easy and lots of ttll cases have blocks in them.
3. The equivalent of 4lll is:
-sexy spam to get a case with 2 or 3 (3 or 4 for the Cw subset) (You can also influence this with intuitive corner control with the previous slot)
-Do 1 of 30ish tsle
-Do 1 of 6 ttll
-epll

Obviously that is a lot worse than normal zz-ct, but you can slowly add cases in to reduce those steps to just tsle, ttll.

 

[wir3sandfir3s]

I think I'm gonna wait for the algs, I would probably screw them up pretty badly, and I got some other new method developments going on...
Also, what's a good tutorial on how to use the newest aCube? Haven't really found a decent one.

 

[sub20oneday]

can someone post a link to acube or ce i cant find it

 

[wir3sandfir3s]

can someone post a link to acube or ce i cant find it

If you go on the wiki, they can both be found in the "Computer Programs" thread.

 

[4Chan]

So I wanted to drop some knowledge on you guys.

As you would've guessed, ZZ-CT is an incomplete method.
It's a subset of a much larger concept, which is to abuse rotational symmetry and other cool attributes of a fully oriented cube to drastically reduce algs.


Here's something that I've been working on, but have given up on because I don't have any time.
I mentioned it to some cubists at Northeastern today, so I might as well post it here for everyone to hear.

So far, it's absolutely absurd, like, ZZ-CT has one F2L corner in the U layer, and solves everything in 72 cases, right?

Well, what about two F2L corners?

What if we didn't even care about building a block, and just put the edges in place and like, just COMPLETELY IGNORE CORNERS.

How many cases would it be to solve 3rd slot, 4th slot, and LL in ONE ALG after that?

With my calculations, it's just 288 algs. (216 + 72 TTLL)

So this expansion pack is still less algs than ZBLL, but you don't even have to worry about building a second block!

The scheme goes like this:
1. EOLine
2. 1 Block (doesn't matter if on left or right)
3. Completely ignore corners, put 3 F2L edges in place and use TSLE (Statistically, 1 is already there LOL)
4. Solve Third Slot, Last Slot, and Last Layer in ONE ALGO




(I'll stop now, I don't want to spam like the new method thread LOL)

Also, totally unrelated, I was doing FMC at Northeastern Spring today, and I used insertions, CTLS, ZBLL, and all that fancy stuff, and I couldn't get below a 39 because my blockbuilding is bad, and I'm garbage.

But with ZZ-CT, I got a linear 38 with it. It really surprised me how few moves this method is!
(I also know that CFOP can get really low move counts, but that's usually with skips)

 

[gyroninja]

So I wanted to drop some knowledge on you guys.

As you would've guessed, ZZ-CT is an incomplete method.
It's a subset of a much larger concept, which is to abuse rotational symmetry and other cool attributes of a fully oriented cube to drastically reduce algs.


Here's something that I've been working on, but have given up on because I don't have any time.
I mentioned it to some cubists at Northeastern today, so I might as well post it here for everyone to hear.

So far, it's absolutely absurd, like, ZZ-CT has one F2L corner in the U layer, and solves everything in 72 cases, right?

Well, what about two F2L corners?

What if we didn't even care about building a block, and just put the edges in place and like, just COMPLETELY IGNORE CORNERS.

How many cases would it be to solve 3rd slot, 4th slot, and LL in ONE ALG after that?

With my calculations, it's just 288 algs. (216 + 72 TTLL)

So this expansion pack is still less algs than ZBLL, but you don't even have to worry about building a second block!

The scheme goes like this:
1. EOLine
2. 1 Block (doesn't matter if on left or right)
3. Completely ignore corners, put 3 F2L edges in place and use TSLE (Statistically, 1 is already there LOL)
4. Solve Third Slot, Last Slot, and Last Layer in ONE ALGO




(I'll stop now, I don't want to spam like the new method thread LOL)

Also, totally unrelated, I was doing FMC at Northeastern Spring today, and I used insertions, CTLS, ZBLL, and all that fancy stuff, and I couldn't get below a 39 because my blockbuilding is bad, and I'm garbage.

But with ZZ-CT, I got a linear 38 with it. It really surprised me how few moves this method is!
(I also know that CFOP can get really low move counts, but that's usually with skips)

By ignore the corners you mean solving the second and third corners with a random one but solve it's co?

 

[4Chan]

Yesss, so you'd still use TSLE, but the probability that something with just one R U R' trigger is practically 100%, since you have so many options and ways of putting the edge in. But now that I'm doing it, I think it'd be hard to practice and see, but it'd be useful if things were already there. I think I might experiment more with algs and easy cases.

Example solve:

Scramble: F' U2 B' U2 L2 F L2 F2 D U' B U R2 B2 L2 B2 L D' B2

EOLine: x' y R' D' R' r U' F' x' y' (6/47)
Block: R U' R2 U2 L U L' U R2 U2 R' L U2 L' (14/47)
2 Edges: U2 R' U' R' U' (5/47)
TSLE: R2 U' R' U2 R U' R' U' (7/47)
Advanced TTLL: R2 U' R2 U' R2 U' R2 D R2 U' R2 U R2 D' R2 (15/47)


I now think that, this approach isn't so great, because making the second 2x2x1 block and getting to last slot is pretty efficient, and that it'd only save 2-3 moves at most on average.

But then it'd come at the cost of losing those sweet skip percentages, so I think the ONLY case in which using algs in which there are two corners in the top is when it happens randomly, which is extremely rare. I think this idea is bad now.

 

[pinser]

^Sounds likes ZZ EF
Yeah, it won't work

 

[Y2k1]

Hello, I was wondering if or when ttll algs will be comming out. (Already have tsle)
Sorry if there there and I missed it

Thanks

 

[Loiloiloi]

WHA-What? A last layer skip method? No way! I gotta watch this thread.

 

[Y2k1]

Nvm, I found the algs. Thanks so much to 4chan for making my new ll method.

http://gyroninja.net/zzct/zzct.html#about

 

[4Chan]

Just one change I forgot to add:

U y' R2 D' R2 U2 R2 D R2 D' U' R2 U' R2

Is actually better as

D R2 D' R2 U2 R2 D R2 D' U' R2 U' R2

 

[Petro Leum]

Thanks to gyroninja, that page's layout is awesome and the algs are also really nice!

 

[PhillipEspinoza]

Mr. Tran it makes me happy to see you still so involved with cubing and developing new methods, this truly is great stuff.

In reading the Disadvantages section on the wiki page by Andrew Ricci "Zzct seems like it would be a cool method. The only disadvantage would be that you have to use zz" (paraphrasing) and also noting someone else's comment comparing zzct to mgls, it got me thinking.

What would an advanced set of ELS algs look like that also oriented the corners. How not feasible would that be? ECOLS would be the idea. Then from there continue with TTLL? Too high move count? Not speed friendly? Too high alg count?

BTW I'm imagining these methods from an OH perspective, which could be a game changer if recog could be mastered.

Again, love this and love you Chris Tran.

 

[4Chan]

Hello Mr Espinoza, it's been like, 6 years!
Hope life is treating you well

Thanks a lot!
Hmmm, thinking your idea, it'd be around 800 algorithms.
Although it'd be totally humanly possible, I think it may be too high for most people.
The move counts and speed ergonomics would be sick though!