zz vs. zbf2l

[nickvu2]

Over the last couple days I've been reading up on what ZB is all about. Granted I'm new to these methods, but it looks like ZZ and ZBF2L both solve first two layers and orient LL edges. Am I missing something; why learn some 125 algs when the same thing can be done intuitively?

 

[miniGOINGS]

ZZ starts with an EOline, ZBF2L starts with a cross.

ZZ:
-EOline
-2 1x2x3 blocks
-EO'd LL

ZBF2L:
-Cross
-3 1x1x2 pairs
-ZBF2L algorithm
-EO'd LL

I would say ZZ is closer to Petrus, with the blockbuilding, and ZBF2L is closer to Fridrich, only differing with the last F2L pair.

 

[eastamazonantidote]

Though at first the difference may seem clear-cut, the advantages and disadvantages create some debate as to which one is "better". miniGOINGS has it all correct, but I'll expand on it with some opinions.

ZZ is done first and intuitively, meaning it takes up your inspection time, making the cross slightly more difficult. ZZ RESTRICTS YOUR SOLVES TO <R,L,U,D,F2,B2>. While it might seem nice, many F2L shortcuts that you learn require some F and B moves. Some cases turn into really strange ones and others are the same. I find, as a casual ZZ user, that there are many more U2s in ZZF2L. Thus, F2L takes slightly longer in ZZ than in pure Fridrich until you adjust. I cannot say what it is like after the adjustment because I haven't gotten there yet. But remember that you have a step before your cross in ZZ.

ZBF2L is quite different. I use an extended version of VHF2L (the 32 + 4 cases that suck otherwise), so I'll use that for my comparison. Your F2L stays the same right up until the insertion. I prepare the edge-corner pair for insertion, then perform my algorithm. For ZBF2L you just take the last pair before this preparation. While it may seem like a lot of algs, many are intuitive. I think I learned 8 algs for my extended VHF2L and then 60 or so tricks to generate those algs. However, while this is a fast step (usually faster than Last Slot + OLL), it is certainly slower than pure Fridrich F2L.

The only reason to use ZZ or ZBF2L over pure Fridrich is if you want to learn a more advanced LL technique. COLL + EPLL is great and, for me at least, is slightly faster than vanilla Fridrich. ZBLL is great but time consuming (I learn a COLL set, forget it, and repeat ever 2 weeks). I like ZZ for learning about the mechanics of the cube and for FMC (mostly just good to know when your edges are oriented). ZBF2L just sounds like it isn't worth it. Intuitively preparing my last F2L pair is just so fast I don't see a reason to learn 240 more algs. I even argue that the VHF2L I know means I know ZBF2L, but I'll save that argument for another day. Hope I helped!

 

[qqwref]

The difference is that ZBF2L waits until F2L is almost done to orient edges, while ZZ orients them at the very beginning of the solve. Both methods have their advantages and disadvantages. ZZ lets you do the whole F2L intuitively, but makes you spend extra moves at the beginning orienting stuff, whereas ZB lets you use Fridrich and then orient edges in only a few extra moves, but requires learning tons of algs.

 

[Cride5]

But remember that you have a step before your cross in ZZ.

There is no cross in ZZ.

Doing the cross after EO means you will not be able to take advantage of many efficient blockbuilding tricks and use many more moves during F2L. Doing only the line means EO and line are planned in inspection, allowing lookahead to the first block during EOLine execution. In short, EO+Cross is useless.

ZZ lets you do the whole F2L intuitively, but makes you spend extra moves at the beginning orienting stuff, whereas ZB lets you use Fridrich and then orient edges in only a few extra moves...

Average optimal cross is ~5.81 moves, and average optimal EOLine is ~6.13, giving a difference of 0.32 moves.
For colour neutral solvers average optimal cross is ~4.81 and EOLine is ~5.01, giving a difference of 0.2 moves.
Figures from here and here.

The difference in F2L move count between pure Fridrich and ZZ isn't as easy to determine. My opinion is that experienced solvers of either method will use roughly the same number of moves for F2L.

ZBF2L is clearly more moves than pure Fridrich F2L, and almost certainly more than ZZ F2L. I can't comment on case recognition/execution of ZBF2L from a personal perspective because I don't use it, but generally more algorithms means recognition, recall and execution are usually slower.

For me the clear choice would be ZZ over ZBF2L, but then not being a Fridrich user my opinion is probably biased. ZZ F2L should be faster in theory. However, ZZ is so different from Fridrich that if you're already a quick Fridrich user it may well be easier to switch to ZB F2L or VH F2L than to learn a completely new method. To the OP I'd recommend having a go at ZZ

 

[qqwref]

ZZ lets you do the whole F2L intuitively, but makes you spend extra moves at the beginning orienting stuff, whereas ZB lets you use Fridrich and then orient edges in only a few extra moves...

Average optimal cross is ~5.81 moves, and average optimal EOLine is ~6.13, giving a difference of 0.32 moves.
For colour neutral solvers average optimal cross is ~4.81 and EOLine is ~5.01, giving a difference of 0.2 moves.
Figures from here and here.

These statistics are pretty much meaningless because after EOline you still have to put in the DL/DR edges to get something equivalent to the cross. If you want to know how many extra moves it takes to orient the edges, you'd have to compare EOcross to cross, or EOline to two opposite cross edges.

Besides, color neutral EOline is completely ridiculous and far, far, far harder than color neutral cross. CN cross is doable because you're looking for four individual pieces so you can quite easily check when one or two are solved already; it doesn't take much time to completely check even 2-3 crosses in inspection, because you can just find the 4 pieces, whereas for EOline you have to basically look at every piece so doing it in a true color-neutral fashion is a lot less doable. Seriously, try doing it in 15 seconds if you don't believe me. Try just figuring out two completely different EOlines (say, red top with blue front vs green top with yellow front) in 15 seconds.

 

[4Chan]

Lol, obligatory zb post.

 

[Cride5]

Besides, color neutral EOline is completely ridiculous and far, far, far harder than color neutral cross.

OK, full CN EOLine is probably pushing it a bit, but certainly opposite colour neutral is possible. I'm not sure if stats have been generated for opposite CN EOLine, but I'd doubt if its any more than 1 move more than full CN cross.

These statistics are pretty much meaningless because after EOline you still have to put in the DL/DR edges to get something equivalent to the cross. If you want to know how many extra moves it takes to orient the edges, you'd have to compare EOcross to cross, or EOline to two opposite cross edges.

The goal here was to compare the whole of Fridrich F2L and the whole of ZZ F2L to allow comparison with ZBF2L. The extra moves involved in EO are compensated for by the strategy used to solve the remainder of F2L.

Placing DL/DR edges is part of the second F2L step and is done along with placement of other pieces as blocks. Using EOCross to analyse ZZ move count would be much like analysing Fridrich move count with separate placement of corners and edges. It's not part of the method and doesn't make any sense.

The reason I compared the move count of EOLine and Cross is that they both form the first step of their respective methods and are very similar in move-count (0.32 moves for non-CN and no more than 1-move difference in the case of CN solvers). If the second step of ZZ F2L has a similar move-count to that of the second step of Fridrich F2L then it means that the whole of F2L for the two methods have a similar move count. If this is indeed the case then ZZ has a clear move-count advantage over ZBF2L since extra moves are used to orient edges on the last slot.

Finally, the move-count issue is always a difficult one where ZZ F2L is concerned because blockbuilding strategy can vary quite a bit between different solvers. A good strategy can significantly lower move count of ZZ F2L. More discussion on it here.

 

[qqwref]

These statistics are pretty much meaningless because after EOline you still have to put in the DL/DR edges to get something equivalent to the cross. If you want to know how many extra moves it takes to orient the edges, you'd have to compare EOcross to cross, or EOline to two opposite cross edges.

The goal here was to compare the whole of Fridrich F2L and the whole of ZZ F2L to allow comparison with ZBF2L. The extra moves involved in EO are compensated for by the strategy used to solve the remainder of F2L.

Placing DL/DR edges is part of the second F2L step and is done along with placement of other pieces as blocks. Using EOCross to analyse ZZ move count would be much like analysing Fridrich move count with separate placement of corners and edges. It's not part of the method and doesn't make any sense.

If you wanted to compare the methods in stupid ways, you shouldn't have quoted my post where I was explicitly talking about where in the method the edges are oriented. I will quote: "ZZ ... makes you spend extra moves at the beginning orienting stuff". Comparing two wildly different first steps and saying "hurr guess it looks like EO doesn't waste any moves at all" is absolutely useless and nothing more than propaganda.

How do you not understand this? Are you actively trying to spread misconceptions just so some method you didn't even invent can be used by more noobs? If you want to talk about how many extra moves EO takes, you have to compare equivalent steps in each method, one with oriented edges and one without. That's all there is to it. If you want to compare the entirety of F2L, go ahead; if you want to compare placing two edges, go ahead; but the first steps of Fridrich and ZZ were both designed to fit in inspection, not to be the same. There is NOTHING to be gained from comparing them. I don't agree with your biased assertion that ZZF2L (not counting the few moves for EO) is more efficient, either - sure, you can do blockbuilding tricks in ZZ, but there are also plenty of F2L tricks for Fridrich, and nobody ever takes those into account. And I'm not even going to bother bringing up how movecount isn't everything (EXCEPT, of course, in the slow beginning stage...).

 

[miniGOINGS]

...I think nickvu2 gets the idea...

Recap: ZZ orientes edges at the beginning of the solve, ZBF2L orientes edges right before the last layer.

 

[Cride5]

qq, if you think ZZ F2L is more moves than ZB F2L could you please provide a reasonable argument as to why this is the case. I've looked at optimal move-counts for different blockbuilding strategies and I don't think my opinion is particularly misinformed.

I also agree that move-count isn't everything, and is another reason why ZZ makes a better choice than Fridrich (and/or ZBF2L).

EDIT: I've generated some statistics here to help clarify the move-count discussion.

 

[Innocence]

...I think nickvu2 gets the idea...

"I'd like to have an argument, please."

"Certainly sir, have you been here before?"

 

[4Chan]

Having switched between both methods for a while now (And memorizing most of ZBF2L, too.) I think I know a little bit about both.
ZZF2L does have a lower move count, but takes a whole different approach and takes lots of practice if you're used to fridrich F2L.

Also, eastamazon, you can totally use F turns and B turns in ZZF2L.
You just need to know how, and a few tricks and such.

 

[dunpeal2064]

...I think nickvu2 gets the idea...

"I'd like to have an argument, please."

"Certainly sir, have you been here before?"

lol @ Flying Circus quote, if that was on purpose xD