[Help Thread] ZZ and ZB Discussion

[Petro Leum]

I like to build the "left" and "right" block entirely at once (so I cancel moves, etc). This requires that I have some idea of where all the pieces required for each block are. However, when I can't see the orientations of my F2L corners, or I can't even find the edges (lol), I end up solving the cube by 1) placing the cross edges, 2) solving the F2L like CFOP, which is terrible for my efficiency.

haha ok

do you know full WV btw?

 

[mDiPalma]

haha ok

do you know full WV btw?

I do for FMC (and pray for PLL skips LOL), but I only use the good cases for speedsolving.

 

[ChickenWrap]

My last Ao100 for CFOP was 19.80, and after about a week of ZZ practice, my solves are between 17-29 (I am not consistent). So my two questions are:

1. Should I stick with ZZ? I like the method alot but it is discouraging to still be so slow.

2. I love solving big cubes, so how could I solve the 3x3 stage with ZZ (or what should I do that isn't ZZ and why)?

 

[cubingawsumness]

My last Ao100 for CFOP was 19.80, and after about a week of ZZ practice, my solves are between 17-29 (I am not consistent). So my two questions are:

1. Should I stick with ZZ? I like the method alot but it is discouraging to still be so slow.

2. I love solving big cubes, so how could I solve the 3x3 stage with ZZ (or what should I do that isn't ZZ and why)?

1. I think you should keep trying since you enjoy it. You've only been using it for a week, after all, and seem relatively close to your CFOP times.

2. For 4x4, I use hoya because methods like hoya and yau are just so much faster than reduction. In addition, ZZ is a very inspection-dependent method. I don't really practice any bigger cubes, but iirc Eli (uvafan) uses redux -> ZZ for 5x5 and up.

Good luck!

 

[ChickenWrap]

2. For 4x4, I use hoya because methods like hoya and yau are just so much faster than reduction. In addition, ZZ is a very inspection-dependent method. I don't really practice any bigger cubes, but iirc Eli (uvafan) uses redux -> ZZ for 5x5 and up.

Good luck!

Thanks for your reply. Are you saying that people actually use ZZ for the 3x3 stage during reduction on a big cube?

 

[cubingawsumness]

Thanks for your reply. Are you saying that people actually use ZZ for the 3x3 stage during reduction on a big cube?

Yup! Just not for 4x4. I don't do it though so I'm not sure how/why it's done.

 

[mDiPalma]

Thanks for your reply. Are you saying that people actually use ZZ for the 3x3 stage during reduction on a big cube?

For a good Yau-ZZ hybrid, instead of solving the 3 cross edges (for Yau), you can solve the line edges as well as a random edge pair. This is less moves than traditional Yau for obvious reasons.

Then, after completing the centers, you solve another random edge. Make sure both your random edges are in the D-layer, oriented.

Then, you pair up edges and remove them from the E-slice in such a way that most of the edges stay oriented. This will cost some moves here and there, but it is mostly a wash, because you need to remove the edges from the E-slice anyway.

You may have parity, so it may not be completely possible to orient edges at this time. But because you can identify parity earlier, you can apply a different (more efficient) parity alg that does not need to preserve the F3L, CP, etc.

Then you solve like ZZ, and apply a PLL parity alg, if necessary.

Surely it's not as good as traditional Yau, but it's still worth a try.

 

[uvafan]

Thanks for your reply. Are you saying that people actually use ZZ for the 3x3 stage during reduction on a big cube?

Once you get enough practice with ZZ inspection can be done in ~2 seconds, barely more than CFOP, or at least enough inspection to start solving the EOLine. The transition isn't that big of a deal.

 

[Petro Leum]

For a good Yau-ZZ hybrid, instead of solving the 3 cross edges (for Yau), you can solve the line edges as well as a random edge pair. This is less moves than traditional Yau for obvious reasons.

Then, after completing the centers, you solve another random edge. Make sure both your random edges are in the D-layer, oriented.

Then, you pair up edges and remove them from the E-slice in such a way that most of the edges stay oriented. This will cost some moves here and there, but it is mostly a wash, because you need to remove the edges from the E-slice anyway.

You may have parity, so it may not be completely possible to orient edges at this time. But because you can identify parity earlier, you can apply a different (more efficient) parity alg that does not need to preserve the F3L, CP, etc.

Then you solve like ZZ, and apply a PLL parity alg, if necessary.

Surely it's not as good as traditional Yau, but it's still worth a try.

thats what i do. im exactly as fast as this with standard yau + cfop. i find standard yau just has faster edgepairing, even if you dont have thinking time about the eo in the ZZ variant.

 

[TDM]

1. Should I stick with ZZ? I like the method alot but it is discouraging to still be so slow.

If you don't stick with it, you'll never be fast. Yes, it's discouraging (and why I've given up on using ZZ for 2H twice), but if you stick with it you'll eventually get just as fast. Don't do another CFOP solve until you're down to your CFOP speed.

 

[aznanimedude]

My last Ao100 for CFOP was 19.80, and after about a week of ZZ practice, my solves are between 17-29 (I am not consistent). So my two questions are:

1. Should I stick with ZZ? I like the method alot but it is discouraging to still be so slow.

2. I love solving big cubes, so how could I solve the 3x3 stage with ZZ (or what should I do that isn't ZZ and why)?

I'd be impressed if you immediately were able to get back to your CFOP times after learning a new method for only a week, much less a good EOLine in a week.

 

[whiteleys]

Hey all,

I've been solving with just zz for a little over a week now and I'm really enjoying it! I got my first sub 30 average of 12 with zz yesterday (it was a 27.5s, I think the slow solving I did earlier really helped) and there's so much extremely obvious stuff in each solve I notice I can improve on which I think is encouraging. My EO line is still pretty bad but I can plan the whole thing out in one go ~80% of the time, its just making it finger friendly that I have trouble with, but that'll come with practice.

The thing I want to ask about though is zzf2l. I haven't really found it that much different to cfop f2l in terms of pairing etc, except for the pairing with the D layer edge first, leaving slots open and preserving pairs more often (oh, and I never put the cross edge in straight away unless there's nothing nice to be done), all of which seemed quite natural to me. When I do slow solving (like 1-1.5 tps) my move count for F2l excluding eoline is 23-24 moves on average, does this seem like the right sort of efficiency for doing zz 'correctly'?

Also my EOline to f2l transition is terrible, despite my cross to f2l transition being pretty good in cfop, does anyone have any tips on this? I've heard slowing down on the last couple moves of EOline can help.

 

[Petro Leum]

Hey all,

I've been solving with just zz for a little over a week now and I'm really enjoying it! I got my first sub 30 average of 12 with zz yesterday (it was a 27.5s, I think the slow solving I did earlier really helped) and there's so much extremely obvious stuff in each solve I notice I can improve on which I think is encouraging. My EO line is still pretty bad but I can plan the whole thing out in one go ~80% of the time, its just making it finger friendly that I have trouble with, but that'll come with practice.

The thing I want to ask about though is zzf2l. I haven't really found it that much different to cfop f2l in terms of pairing etc, except for the pairing with the D layer edge first, leaving slots open and preserving pairs more often (oh, and I never put the cross edge in straight away unless there's nothing nice to be done), all of which seemed quite natural to me. When I do slow solving (like 1-1.5 tps) my move count for F2l excluding eoline is 23-24 moves on average, does this seem like the right sort of efficiency for doing zz 'correctly'?

Also my EOline to f2l transition is terrible, despite my cross to f2l transition being pretty good in cfop, does anyone have any tips on this? I've heard slowing down on the last couple moves of EOline can help.

sounds like you got the concept of EOLine and zzf2l quite well. your movecount is good average.
transition from EOLine to F2l was my biggest problem too (and still sometimes is)

what helped me:
1. track/predict the LD/RD edge during EOLine, so you either know exactly which block you're starting with, or
2. predict the LD edge and track a pair of the left block if you always start with left block (bad for efficiency, great for ergonomics)

 

[whiteleys]

Oh okay awesome, thanks!

Cheers, the tracking the LD and RD really helps, I'm not too good at it yet but I've noticed it's already helped me out a few times.

 

[Petro Leum]

out of interest, i just did an average of 12 timing EOLine/F2L/Last Layer separately. this is what i got:

EOLine:

Average: 2.034 (σ = 0.21)
Mean: 2.046

Spoiler

Time List:
1. 1.880 R2 U2 B2 D' R2 D' L2 F2 L2 F R' B' L2 U B' F D' B L2 R
2. 2.219 R2 L D' R L U F' B2 L F B2 D2 B' R2 L2 U2 B U2 B' U2
3. 1.778 D L2 D' F2 L2 U2 L2 D' B2 F2 U' B' D2 L2 R' D' R U2 L2 F R2
4. 1.955 R2 B2 D R2 F2 R2 D2 R2 U' R2 F' L' U' R' D R D' L2 R' D2
5. 1.904 U' L2 U2 B2 L2 B2 U L2 U F2 U R' F2 L2 R2 U F' R U' B' D2
6. 1.476 D F2 U' L2 D F2 R2 D R2 U R2 F' L' U R2 B F2 D2 R2 B
7. 2.357 F D2 B2 D2 B2 L2 U2 F2 R2 F' U2 L' U' L2 B' R B2 L2 D B2 L'
8. 2.231 D R D L B' D R F U' B' D B2 D L2 U' B2 U F2 R2 F2
9. 2.283 L' D2 R B2 D2 R' D2 B2 D2 L2 D2 U L2 R B' F2 U' B F R
10. 2.738 F2 L2 D2 L2 U' R2 U2 R2 B2 U2 F L2 B' L F' D' L B' F' R' U'
11. 1.904 U B2 D2 U R2 B2 U2 F2 U' R2 B2 R' D' F' L' B D' L' U' F L
12. 1.829 L' B R' F2 U' F2 D F2 B' R U2 R2 D2 L2 F2 L2 B' U2 D2 F' D2

F2L:

Average: 5.307 (σ = 0.38)
Mean: 5.347

Spoiler

Time List:
1. 5.399 B2 U2 R2 U2 B' R' U F D R' L2 F2 R' D2 R U2 B2 D2 R'
2. 4.881 U' D F U B2 L' B2 R2 L D F2 D2 B R2 F U2 R2 B2 D2 R2
3. 6.655 L2 U R2 B2 U B2 L2 R2 U L2 B R2 U' L2 U' F U B R' U
4. 5.852 L2 D2 U2 B2 L B2 U2 R' F2 D2 L F' U L B' L U R2 D2 L D'
5. 5.195 F U2 F' D2 B U2 F2 D2 F U B F2 L' U2 R' B2 R2 B' F'
6. 4.837 R U2 R2 B2 F2 L' D2 R' B2 L2 B' D B F L B' R2 F' U R2
7. 5.716 R2 D2 U2 F' D2 F' U2 B' L2 D2 F' U B D2 U L2 U' R' D' R' F'
8. 5.551 F' D2 U2 L2 F' L2 F R2 B' D2 U B U L D L2 F D2 R2 B2
9. 5.328 R2 B2 D2 U2 L2 U2 F2 U R D F' L2 F L' B' U L' U F'
10. 4.748 U R U2 R B R U2 L2 U' B2 R U2 R2 B2 L U2 B2 R' B2
11. 5.562 D' U R2 D' F2 R2 F2 U L2 F' U' R' B' D2 B L' U L' R'
12. 4.439 F2 U' L2 U R2 D' B2 F2 L2 U' F' D2 U F2 D' L2 R F' D'

Last Layer:

Average: 3.157 (σ = 0.54)
Mean: 3.174

Time List:

Spoiler

1. 3.697 F2 L F2 D2 B2 L D2 L B2 L' F R D' F2 D B' U' L2 F'
2. 3.076 L' R2 D2 L' R2 U2 F2 R2 D B2 U B' R2 D2 F' R D' L'
3. 2.765 F' L2 F' B2 R' D' F' B' L B' L2 B2 L2 U D F2 U B2 U' L2
4. 3.437 F' D2 F' U B' U F2 D L D2 F2 D' L2 D' B2 R2 F2 R2 D L2
5. 2.851 F' U L D R' D L B L' D2 F' L2 F2 R2 B' R2 F2 U2 R2 L2
6. 2.382 D2 F2 L' U2 D R' B L2 U F L2 D L2 B2 U' R2 D L2 D L2
7. 2.751 L2 D F2 D' F2 U F2 D2 F2 R2 D2 R B R' F R2 U L R' D2
8. 3.694 R2 F2 D2 L2 B2 U' L2 U' F2 L2 B2 F' D U2 L2 F' R2 D F2 L' D2
9. 4.136 R B' U R' U2 L' D2 R2 U' B2 U2 F' L2 F L2 B U2 R2 F B2
10. 2.559 B2 U2 L D2 F2 R' F2 L D2 F' D' L2 B2 L2 B D2 F2 U'
11. 2.629 F' L2 D2 U2 B2 F U2 B D2 L2 D2 U R' D2 U2 F L2 U' R' B F2
12. 4.111 U2 F2 L' F2 D2 L2 U2 F2 L2 F2 U' F2 L R F R B' F2 D2 U B'

adding uo the averages i get
10,498

i average 12 seconds (today i was mostly at 11.5)
although i tried, this does not include COLL/ZBLL recognition and, far more concerning, doesnt include proper transition from EOLine to F2L.

i also didnt influence LL during F2L which is rather common for me.
i got 2 or 3 rather long ZBLLs in the average.

are these normals splits? what are yours? what should i improve?

 

[xsolver]

After trying out Roux, i am really interested in trying ZZ too. I pretty much love the thing that you have to use more mind into this method, than into CFOP.

Would you recommend for me to try this method even tho i am quite a new solver and avg around 25-6s with CFOP? I only know 2look OLL, so i wont be loosing much i think. Also PLL is the same in this method as in CFOP, right?

Thanks.

 

[Bindedsa]

After trying out Roux, i am really interested in trying ZZ too. I pretty much love the thing that you have to use more mind into this method, than into CFOP.

Would you recommend for me to try this method even tho i am quite a new solver and avg around 25-6s with CFOP? I only know 2look OLL, so i wont be loosing much i think. Also PLL is the same in this method as in CFOP, right?

Thanks.

If you are interested, start now. There is no advantage to spending time improving CFOP then switching. As for LL, yes PLL is the same, but instead of FULL OLL being he next step after PLL it's COLL. There are a 40 alg, but many people don't learn sunes making it only 28.

 

[AKOM]

After trying out Roux, i am really interested in trying ZZ too. I pretty much love the thing that you have to use more mind into this method, than into CFOP.

Would you recommend for me to try this method even tho i am quite a new solver and avg around 25-6s with CFOP? I only know 2look OLL, so i wont be loosing much i think. Also PLL is the same in this method as in CFOP, right?

Thanks.

What is your problem with roux? maybe your issues with roux will be the same with zz.

 

[xsolver]

What is your problem with roux? maybe your issues with roux will be the same with zz.

I didnt really have any problems with Roux, but i decided that a transition from CFOP to Roux at that point would not be worth it. I find more similarities in ZZ than in Roux.

If you are interested, start now. There is no advantage to spending time improving CFOP then switching. As for LL, yes PLL is the same, but instead of FULL OLL being he next step after PLL it's COLL. There are a 40 alg, but many people don't learn sunes making it only 28.

Ill take a look into this method as soon as i have more free time. I think that ill start with Phill's videos.

 

[whiteleys]

out of interest, i just did an average of 12 timing EOLine/F2L/Last Layer separately. this is what i got:
...

are these normals splits? what are yours? what should i improve?

I know I'm not the best person to help but sometimes other people's viewpoints no matter how inexperienced can be useful. A method for improving that could be good especially with your splits would be to individually train a split till it is below a goal. The good thing about your splits with respect to this method is that yours are all fairly close to the second below (EOL is 0.05 away from 2, f2l is 0.3 from 5 and last layer is 0.16 away from 3). You could target f2l practice solidly say till you got to a 5s average on it. This gives you more readily attainable goals which are reasonably significant in themselves (i.e. give you a sense of satisfaction when you reach them) than say having a goal to get a sub 11 average. When you add these small improvements you'd have a 10s execution with 1s for recognition in between steps and theres your 11s average sorted. I suppose that just boils down to breaking it up into manageable chunks though, and setting goals you want to work towards and can reach.

They do seem like normal splits from the faster (sub 15) ZZ solves I've seen so there's nothing to worry about there.

I haven't actually timed mine but I'd guess about 6/15/5 for me including recognition. The weird thing with me is that I did an average of 20 on last 1x2x3 block and last layer combined (RUL scramble so I don't get free CP skips with about 0.5s inspection to simulate lookahead) and I got a sub 10 average, which suggests my first block/pair etc and eoline- first block transition needs work. What do you think?

If I were you I would work on f2l because it takes up over half your solve and so little improvements in technique etc can make a big difference. The other thing I would look at is your last layer. This isn't because it's slow but you have a standard deviation of 0.5 on it, which is a large in comparison to your other S.D.s despite your last layer being quick. However you did say you had a few long ZBLLs in it so I'm not sure if that significantly affected this. If it didn't I'd just look at the algs you tend to lock up on or find the execution a little clunky till they are at the same level as some of your preferred algorithms to give you a more consistent last layer which should help your averages out. Either way those are some really impressive splits, especially to someone of my level, so great job!

On a side note, you said that the EOLine to F2L transition sometimes catches you out, so an interesting split to see could be EOLine + 1 1*2*2 block.

Hope this helps, sorry I'm not really in the position to give technique based advice,

whiteleys