• Welcome to the Speedsolving.com, home of the web's largest puzzle community!
    You are currently viewing our forum as a guest which gives you limited access to join discussions and access our other features.

    Registration is fast, simple and absolutely free so please, join our community of 35,000+ people from around the world today!

    If you are already a member, simply login to hide this message and begin participating in the community!

New ZZ Variant: ZZ - Profane Koala (ZZ-D with 4 algs)

Joined
Jul 21, 2008
Messages
1,627
WCA
2010YUPH01
YouTube
Visit Channel
Hey Everyone,

I am writing to present a new ZZ variant I've been working on with Statue. We thought of this idea more than a year ago and we recently reconvened to discuss actually going forward with formalizing the method. It's based on ZZ-d.

Method Breakdown
(This description presumes white EO Line)

1. EO Line
2. Left block
3. Push your two remaining white corners to the two bottom-right corner spots (orientation and permutation do not matter) (this usually takes 1-2 moves)
4. Corner permutation (sometimes the corners will come permuted) (otherwise, the algs are 5-6 moves)
5. Right block
6. 2GLL

*You can also approach solving the block + EO with Petrus
**You can mirror this stuff if you prefer making a right block first and doing lefty 2GLL

Algs for Basic ZZ - Profane Koala
Good + Good: 2GEN
Good + Adj: [R] D' R U R' D
Good + Opp: L R U' R U L'
Bad + Good: R' D' R U' R' D
Bad + Adj: [F] U D R' U R D'
Bad + Opp: 2GEN

*The first term refers to corner permutation
**The second term refers to what swap is required to reduce the cube to a 2gen state
***The letters in brackets denote what face the swap will take place

Hypothetical Extensions
Instead of pushing the two corner pieces to the bottom layer, you can extend to pushing them to other, more convenient spots. This would reduce the move count of step 3, which should allow easier transition from step 3 into step 4. Eventually, we expect solvers to know so many "pushing arrangements" that they can make at most 1 move and then permute the corners straight after.

Help!
We're looking for ideas on how to recognize corner permutation for step 4. Any help would be greatly appreciated.

Video
For those of you that want a video:


Many thanks,

Statue of Kitten

 
Last edited by a moderator:
Joined
Nov 8, 2010
Messages
5,472
Location
near Ottawa, Canada
WCA
2010CANT02
YouTube
Visit Channel
I have toyed around with something similar (finishes with putting the two right corners on D and does CP, then 2gll). Here is what I have found:
- Putting in the two corners is a bit annoying and takes a lot (relatively) of time
- CP recog is annoying
- It's annoying when you did all the stuff and you get CP skip anyway

edit: The way I would recog CP is I would do a U move until there were two solved pieces. If they were diagonally swapped, I would do the diag alg, and two adj. were solved, I would put then on B and do the alg. Similarly, if the two corners that I had inserted were not in the right slots, I would do U until there were two solved corners on U, and do the opposite, so if there were two diagonally swapped corners on U CP was good, and if they were solved I'd have to do the diag swap alg, and if there were two adj. ones solved I'd put them on F then do the adj. swap alg in order to get diag. swap on the U layer, and solved CP on the whole cube.
 
Last edited:

TheNextFeliks

Member
Joined
Oct 27, 2012
Messages
2,414
Location
KCKS
WCA
2013POPE01
YouTube
Visit Channel
Isn't this more moves then CFOP now?
No. The average 2gll case is approx 13 moves. The avg for this is 4 (6 if you don't include the solved). 17 vs 24 (iirc). And zzf2l is usually more efficient.

OP: I'll try this. If I like it, I might stick with zz. I kinda switch back and forth. I just like being sub-20 with CFOP. I avg barely sub-25 with zz.
 

aceofspades98

Member
Joined
Mar 17, 2013
Messages
228
Location
Lakewood, Colorado
WCA
2013REYN01
No. The average 2gll case is approx 13 moves. The avg for this is 4 (6 if you don't include the solved). 17 vs 24 (iirc). And zzf2l is usually more efficient.

OP: I'll try this. If I like it, I might stick with zz. I kinda switch back and forth. I just like being sub-20 with CFOP. I avg barely sub-25 with zz.
To solve the corners there are about 6 moves though.
 
Joined
Apr 18, 2009
Messages
454
Location
San Diego, California
WCA
2007ESPI01
YouTube
Visit Channel
This is brilliant! I feel like this would be a sub-set of zz-d and all the cases you mentioned of different positions for the last 2 white corners would be full zz-d. I think recog would be similar to recog for cpls, or rather, if you could recog cpls or eocpll then you should be able to transfer over reocg methods. Great job!

EDIT: What if you could create algs that would fix cp during the construction of the last slot (or at least the 1x2x2 block of the last block?) If you could master recognition, do you think looking into last block corner permutation would be feasible?
 
Last edited:

AHornbaker

Member
Joined
Oct 9, 2012
Messages
64
Hypothetical Extensions
Instead of pushing the two corner pieces to the bottom layer, you can extend to pushing them to other, more convenient spots. This would reduce the move count of step 3, which should allow easier transition from step 3 into step 4. Eventually, we expect solvers to know so many "pushing arrangements" that they can make at most 1 move and then permute the corners straight after.

Help!
We're looking for ideas on how to recognize corner permutation for step 4. Any help would be greatly appreciated.
Here's an idea on the topic of recognition:
Another way to place the corners is one on ULB and one on DRB. You can see all the yellow corners clearly, and it only takes two setup moves. If I'm right, there are only 12 recognition cases: four that are correct, eight that are swapped. However, you would have to take note on which of the white corners you put in which place; the recognition would still be 12 cases, but which ones are correct and not would change. This takes a lot of guesswork from looking at the top layer and deciphering which corners are which.

This stemmed from another idea I had to learn the patterns for all cases, ie. adjacent corners, opposite corners, and one corner on each layer. This would reduce to a case for adjacent corners on U and it's mirror on R, a case for opposite corners on U and its mirror on R, and a case for one corner on each layer at ULB/DRB. The last case I realized could be achieved in two moves, which is where I came up with the above idea. Hopefully this sparks some discussion.


Also, I don't understand the point of doing CP after EO. You might as well solve a 2x2x3 block like in Petrus, then do both CP and EO in one alg set. A more efficient method would be like porkynator's CP during the first block of ZZ here:
http://www.speedsolving.com/forum/showthread.php?43236-New-Approach-to-ZZ-d
On a related note, it doesn't make sense to do CP if you're not going to do 2GLL. You might as well just stick with PLL or CLL.
 
Last edited:
Joined
Mar 17, 2013
Messages
447
Location
England
This takes a lot of guesswork from looking at the top layer and deciphering which corners are which.
A lot of the guesswork should be able to be eliminated by tracking at least one, preferably two of these corners while placing the white corners in ULB and DRB, and taking note of what the sticker that will end up on the L/B/D face is.
 

AHornbaker

Member
Joined
Oct 9, 2012
Messages
64
A lot of the guesswork should be able to be eliminated by tracking at least one, preferably two of these corners while placing the white corners in ULB and DRB, and taking note of what the sticker that will end up on the L/B/D face is.
True, I was just thinking about this myself. Would definitely reduce the # of algs to recognize. I haven't looked extensively at ZZ-Orbit but I wonder how it compares to this and other methods of recognition.

**EDIT: I'm thinking that since you can see so much on this variation, it might not be so much of identifying all the yellow corners and their permutation scheme, but maybe just looking to see if the blue corners are adjacent and the green corners are adjacent/opposite to determine the CP. I'll begin research on this tomorrow.
 
Last edited:

RubiXer

Member
Joined
May 13, 2012
Messages
182
Location
So Cal
WCA
2012MORA05
Correct me if I'm wrong but isn't this just ZZ Rainbow w/o the isolation? If so I do do this in my OH solves every once in while if I immediately recognize the CP case.
 

AHornbaker

Member
Joined
Oct 9, 2012
Messages
64
Correct me if I'm wrong but isn't this just ZZ Rainbow w/o the isolation? If so I do do this in my OH solves every once in while if I immediately recognize the CP case.
Please elaborate, I couldn't find any documentation of it online.
 

AHornbaker

Member
Joined
Oct 9, 2012
Messages
64
**EDIT: I'm thinking that since you can see so much on this variation, it might not be so much of identifying all the yellow corners and their permutation scheme, but maybe just looking to see if the blue corners are adjacent and the green corners are adjacent/opposite to determine the CP. I'll begin research on this tomorrow.
Just by doing some preliminary testing, this looks like it could be THE cp recognition scheme. It looks as though it could be as simple as placing the two corners on ULB and DRB with two moves, then based on which corner is where, looking at the other yellow corners and seeing if there is a common color between two certain corners. I'll have to look into it more, but I wanted to get this out there so I have dibs on the name :D if it happens to work out.
 
Top