ZZ speedcubing method

Cride5 said:
Hay people! I've just finished my second all-nighter working on a ZZ cubing page. It explains EOLine and F2L in some detail, along with algs, diagrams, animations etc. Get it here:

http://cube.crider.co.uk

Let me know what you think .. if you have any comments or suggestions etc.

Now I need to get some ZZzzz's :D
Click to expand...

Awesome. We need more support for ZZ!

Also, doesn't ZZLL (ZZ-b) require 167 algorithms?
 
Lord Voldemort said:
Where does 167 come from?
40 COLLs that preserve edges
40 COLLs that switch two opposite edges
Click to expand...

Lol, if only it were that simple :rolleyes:

During phasing, the cube is setup so that two edges are permuted correctly with respect to eachother. Note that this does not guarantee correct permutation with respect to the corners.

If two opposite edges are permuted correctly, then it follows that the two other edges are permuted correctly with respect to eachother. However, this does not mean that all four edges are permuted correctly in relation to eachother since two edges may be swapped - consider the edges in a T-perm for example.

So there are two possible cases when considering edge permutation in isolation:
NO PARITY - All four edges are permuted correctly
PARITY - Two edges are swapped (like edges in T-perm)

Now when we consider edge permutation along with corner permutation, then when there is no parity, then it is impossible for the edges to be in any position other than the correct position with respect to the corners, since any other positions with the edges permuted correctly with respect to eachother break the laws of the cube - ie: 3 swaps would be required to correct it. If you don't believe me, try to imagine a PLL where all edges need to be rotated by 90°, while edges are solved - it doesn't exist!

Now when we consider edge/corner permutations where there is parity then it gets a bit more complicated. There are actually 2-ways the edge permutation with swapped edges can be permuted with respect to the corners: Z-perm and T-perm. With the T-perm only two edges need to be swapped which means two will be solved during ZZLL, thus two must already be in a solved state. With the Z-perm all four edges are in an unsolved state (with respect to the corners), and so the algorithm will need to swap two pairs of edges. It is called ROTATED PARITY, since it is the same as twisting the edges in a T-perm by 90°.

So, what it boils down to is that you need COLL with NO PARITY, COLL with PARITY (swaps two edges), COLL with ROTATED PARITY (swaps four edges).

With regards to number of algorithms required: COLL has 40 distinct cases, but can be solved with knowledge of 25 algs, thus by your reasoning ZZLL would only require 50 algs total. This however is not the case, because of what I described above. ZZLL actually has 167 cases, which are solvable with knowledge of 80 algs.

Hope that made sense!!
 
Cride5 said:
Lofty said:
Right... I know all of these tough cases and use them, that is why I said for the 4th pair alg/intuition and not just intuition like I did for the 3rd pair. The point of ZBF2L is to orient the LL edges as I'm sure you know, so I don't know how that is even remotely applicable for the ZZ method. The only thing I can think of is that when doing the EOLine you ignore the LL edges and then the number of ZBF2L would be cut in half if you count mirrors as separate algs as you wouldn't have to mirror anything since the last edge is already oriented.
Click to expand...

Actually, I think that figure of half would probably be less than a quarter in reality. Anyhoo ... the point is that ZBF2L cases where edges are already oriented are the ones which would apply. The point about the ZBF2L algs is that the cases with LL edges already oriented preserve LL edge orientation, where normal Fridrich F2L algs would destroy edge orientation. Its deffo worth a mention because a lot of work has already gone into generating these algs, which is something that the ZZ community can tap into for free :D

With regards to solving the '3rd pair' ... I just need to make sure we're talking about the same thing, and you aren't thinking in terms of EOCross. When I do ZZ its broken down into two 1x2x2 blocks and two 1x1x2 blocks. For any given side, its 1x2x2 comes before the 1x1x2. If I come across a connected case after completing the 1x2x2 block, I'll just pick a different set of pieces to build the next 1x2x2. If, however I've completed the two 1x2x2 blocks and both my remaining corner-edge pairs are connected then I'll apply my connected case algs. I do this because using the free 1x1x2 slot rarely, if ever results in an optimal move count. In addition, because you already know these algs, then you should be able to rattle them off really easily.
Click to expand...
A quarter? try going from about 300 to about 20... For most of the 41 F2L there are 8 different cases for the edges. Hence why VHF2L is "32 algs" but only one of those 8 has all the LL edges oriented. But then more than the LL edges are oriented, all of them are. That cuts out half of the 41 F2L cases leaving about 20.
For me I don't think of inserting a pair using 2-gen algs (which all or most all of these algs would be) as ZBF2L. Using very simple edge control like that is a totally different thing than ZBF2L. So yea. For those of us who have worked on any kind of edge control for our fridrich solves (mainly just went on Lar5 and saw all the ways to solve the last slot of the petrus method) this step is trivial while ZBF2L is no where near trivial. Its also nowhere near half but maybe blah will clarify that for us.
Yes, f'or the third and forth pairs we are talking about the same thing.
Edit: Learn the double sune cases! At least some of them are pretty nice.
 
Mah YOUTUBE vids. :U (Just search ZZ, I think they're still the only video source there.)
 
Good choice! Deffo have a good stab at it, especially EOLine. Its hard at first but gets easier later on, trust me ;)
Start by doing EO first, followed by the line (EO+Line). Integrating EO and Line is easy enough once you've mastered EO.

Good luck :)
 
Yes. I know this is a rather large bump, but I think that I may have something to add to help some with (the solving, not recog. of EOline, mainly with EO).

M' U M'.
While, yes, one would mainly think that this could only be used in cases such as [UR, UL, UF, DF] disoriented, it could be used in any situation where there are 3+ disoriented edges on a given face.

Given the following scramble, (white top, green front), one can see the immediate advantage of (M' U M').

U' R2 F R' B U' F2 R L U2 B2 D2 F D2 B L B L B L U L2 R2 U2 R'

One should see 6 disoriented edges (All on U face + DB + FL).

First, do a y2 during inspection, making it so after (M' U M'), there are only 2 disoriented edges.
Next, do (M' U M')
Then fix the other two however.
Well, while that was an alright example, I have confidence that there are many GREAT examples. Using this EO "method", I got a new PB avg of 5, so needless to say that I'm liking it.
 
I guess conceptually, what you're doing is temporarily changing the edge orientation face to U. The concern I'd have with M' U M' is that it essentially flips the cube over, so I'd probably prefer something like M' U M, or M U M' to preserve the cube colour scheme. NOTE: This is exactly equivalent to doing L' R B L R', or L R' F L' R respectively. It could be useful in FM situations where slice-turn metric was used, or if you're faster at M-slice moves.

My normal EOLine for this case would be: L' B' D2 L R' F' B2. The EOLine using M-slice turns could be: M' U M L' F' L F'. In this case its no more efficient, but but deffo something worth including in the EO hacking repertoire, even if its just making EO more finger friendly ;)
 
Last edited:
ZZ method EOline help

In the ZZ speedcubing method, the first step is the EOline. You have to look and make sure all the edges are flipped the right way. How do you go about this? I can do the Roux method, when you have to make sure the last 6 edges are flipped right, but how do you make sure all 12 are flipped right? The website suggested looking at Macky's tutorial, but all I could find on his website was the BLD edge tutorial. Are you suppose to use this?

Here is the ZZ speedcubing system, if any of you want to see my question in context. It is very possible i am just misunderstanding the EOline step.

http://www.emsee.110mb.com/Speedcubing/ZZ%20speedcubing%20system.html

Thanks,
Zach
 
Yes look at macky's BLD edge tutorial. Don't read about how to orient the edges with this method, just what the definitions of oriented and misoriented edges are. I think the site you linked quotes the part you'd need from Macky's tutorial.

A short summary:
-U/D colors on U/D = good
-R/L colors on U/D = bad

For the middle slice edges:
U/D colors on F/B = good
U/D colors on R/L = bad
R/L colors on R/L = good
R/L colors on F/B = bad
 
If you can bring the little lost edge home without using F or B moves, it's good. That's how I see it.
 
Ellis covered it,. Just a word of advice from experience though, at first it’s really hard, but after a few weeks, the bad edges just come out.
 
It's just like Petrus EO, except with 4 colors instead of 2. If you look at opposite colors as being the same, it's just normal Petrus.
 
Ellis said:
Yes look at macky's BLD edge tutorial. Don't read about how to orient the edges with this method, just what the definitions of oriented and misoriented edges are. I think the site you linked quotes the part you'd need from Macky's tutorial.

A short summary:
-U/D colors on U/D = good
-R/L colors on U/D = bad

For the middle slice edges:
U/D colors on F/B = good
U/D colors on R/L = bad
R/L colors on R/L = good
R/L colors on F/B = bad
Click to expand...

This is pritty much the way I do it, apart from there's one thing missing. For F/B colours on U/D, check the side, U/D is bad, L/R is good. I created a whole heap of stuff on how to detect and solve EOLine here. Let me know if there's anything you're having trouble with ;)
 
You must log in or register to reply here.

Similar threads

Caleb Valenzuela
How to make a “Super Duper Unique” Speedcubing Method (for all events)
Replies
1
Views
483
Jared Burkhart
Jared Burkhart
JackTriton
New 3x3 Method: (Named) JLA 3x3 method, based on Lau Method
Replies
19
Views
2,123
JackTriton
JackTriton
Ralph Laurence
Thoughts on Lau Method
Replies
11
Views
3,525
Ralph Laurence
Ralph Laurence
Michael Wimberly
Making A New Speedcubing Method
2
Replies
32
Views
3,877
Michael Wimberly
Michael Wimberly
Epinore
Gear Cube X Method: Basic Speedcubing Tutorial
Replies
3
Views
1,901
Epinore
Epinore
Back
Top