ZZ-A vs ZZ-CT Discussion

[LexCubing]

Humphrey said ZZ-A is better in the highest level of competitive solving but why? ZBLL moves ~8 pieces while TTLL moves ~9 pieces.

Is it because having an oriented cube restricts the cube alot more as a trade-off for a lower alg count? Is TSLE/TTLL Recog harder than LS/ZBLL? Like what he said ZBLL recog also gives info on the top?

I think PLL style recog and ZBLL recog is the same since (Assuming white is the top color) you "can't look" at the white sticker to determine your permutation.


You can look at the top of the cube to get info but the pieces are just flipped. The only difference is PLL style only makes you look down while ZBLL makes you look both top and down but the # sticker use to recognize is the same for both.

So they're the same thing but ZBLL is a little whackier because of the orientation but with PLL style block are more apparent.

Also what makes CT easier in terms of recog is Orientation and Permutation are recognized separately unlike ZBLL where you're forced to address "at the same time". Making blocks less apparent opposed to CT.

These are my opinions.
CT has lower alg count and sort of easier recog. I don't want to say better since at the end of the day recog is subjective. While ZBLL has higher alg count and and a weirder recog.

So for me CT is better.

What's your opinion and why is ZZ-A more efficient than CT?

 

[shadowslice e]

At the highest levels, alg count takes second seat to efficiency. In addition, insert+1 alg is better than alg to insert alg to solve due to reduced recog time.


CT is just a make do but not as good as if you lean the whole thing.

 

[LexCubing]

At the highest levels, alg count takes second seat to efficiency. In addition, insert+1 alg is better than alg to insert alg to solve due to reduced recog time.




CT is just a make do but not as good as if you lean the whole thing.

What? Can you explain that again. Like what the hell insert+1 alg stuff like that? I don't get what you mean.

Why is ZZ-A more efficient. It moves ~8 pieces while TTLL solves ~9 pieces?

 

[AlphaSheep]

Why is ZZ-A more efficient. It moves ~8 pieces while TTLL solves ~9 pieces?

There's no simple relation between the number of pieces solved and the efficiency of an alg set.

Anyway, my opinion after having learnt 25% of ZBLL and 30% of TTLL, I feel recognition for TTLL is slightly better because all of the stickers you need to look at are next to each other, whereas with ZBLL, you have to compare, for example a top sticker on a corner to a side sticker on an edge. You get used to this with practice though. On the other hand, ZBLL cases are more distinct from each other and have more features that stand out, so I find building an association between the alg and the case easier for ZBLL. The same applies to ZZLL, except with ZZLL, you have one less sticker that you need to look at so it's easier to recognise the case.

In terms of the quality of the algs, I prefer the ZBLL algs for the most part. With some exceptions, TTLL algs tend to be a couple moves longer than ZBLLs, and finger trickiness of many ZBLLs is better. With ZZLL, you lose some of the short and easy ZBLL algs (like Sune), but lots of the easiest ones stay (such as all of the 3 cycles). On average, the ZZLL algs are on par with the rest of the ZBLL algs.

With regards to alg count, I feel it does play a massive role. Maintaining 93 TTLL algs (+105 TSLE algs) is obviously less effort than maintaining 493 ZBLLs. However, if it's alg count, then 169 ZZLL algs is lower than both.

So my opinion is that ZZ-b is probably good enough for most.

 

[TDM]

Why is ZZ-A more efficient. It moves ~8 pieces while TTLL solves ~9 pieces?

Efficiency isn't about the number of pieces solved. Even if it was, why would solving 9 pieces require fewer moves than solving 8?

Also, the algs for TTLL are awful.

 

[shadowslice e]

What? Can you explain that again. Like what the hell insert+1 alg stuff like that? I don't get what you mean.

I mean that of you do a standard ZZ insert then 1 alg it is in general better than doing an edge insert, then an alg to insert the corner then an additional alg to solve everything else.

Why is ZZ-A more efficient. It moves ~8 pieces while TTLL solves ~9 pieces?

For TTLL you have more to preserve (orientation).

 

[LexCubing]

I mean that of you do a standard ZZ insert then 1 alg it is in general better than doing an edge insert, then an alg to insert the corner then an additional alg to solve everything else.


For TTLL you have more to preserve (orientation).

So if I understand correctly, ZZ-A is
F2L
ZBLL

While CT is
Edge in
Corner in
Solve LL??

Efficiency isn't about the number of pieces solved. Even if it was, why would solving 9 pieces require fewer moves than solving 8?

Also, the algs for TTLL are awful.

Any alg set has bad algs until they get optimised later on.

I knew it that orientation will make TTLL less efficient unlike ZBLL which has more freedom. Oh well.

 

[Malkom]

Isn't ZBLL recog kinda easy at the highest level? When you start it's pretty much COLL + 2 edges, seems pretty easy but may be hard becuase it may need a lot of AUFs. But after a while you should be able to find blocks and other things that makes distinguishe different cases and make ZBLL recognition "easy".
(I actually have no idea what I'm talking about, I don't even know COLL)

 

[AlphaSheep]

Any alg set has bad algs until they get optimised later on.

This may be true for alg sets in general, but it's certainly not true on the level of individual cases. For individual cases, you can spend a few hours generating and going through every possible candidate for a good alg (easily a couple hundred algs for a case), and be guaranteed that no better alg exists than the ones you tested.

I should also point out that people tend to do this for the worst cases first. I know I have.

 

[Teoidus]

ZBLL (~21 htm LS/LL) is more efficient than TSLE/TTLL (~25 htm LS/LL)

Simple counterexample to illustrate why "solving more pieces" =/= "more efficient algs":
Alg set of the pure 3-corner twists vs an alg set of pure corner 3 cycles

Both solve 3 pieces, but one is grossly inefficient compared to the other

Same with alg set of all pure L6EP cases vs an alg set of all pure L6EO cases

 

[mDiPalma]

zz-a's movecount is also offset by a host of short 7-8 move algs/commutators

 

[Malkom]

zz-a's movecount is also offset by a host of short 7-8 move algs/commutators

Wut?

 

[AlphaSheep]

Wut?

He means that there are lots of 7-8 move algs and commutators that are good ZBLLs that bring down the average movecount. Some examples:
R U R' U R U2 R'
r' F R F' r U R'
r U R' U' r' F R F'

Then there are all of the 9 move comms... There are lots of those too.

 

[shadowslice e]

Simple counterexample to illustrate why "solving more pieces" =/= "more efficient algs":
Alg set of the pure 3-corner twists vs an alg set of pure corner 3 cycles

Both solve 3 pieces, but one is grossly inefficient compared to the other

Same with alg set of all pure L6EP cases vs an alg set of all pure L6EO cases

These actually help proving the point more as you preserve more in pure twists than in 3 cycles (permutation is preserved in the first one but not the second).

 

[Teoidus]

These actually help proving the point more as you preserve more in pure twists than in 3 cycles (permutation is preserved in the first one but not the second).

Orientation can be preserved in 3 cycles, though (you could even just consider the corner 3cycles that preserve orientation and they'd still be more efficient), and L6EP preserves EO. I think the point is more that it's not just "more entropy = more efficient algs!" as was once advertised by some

 

[LexCubing]

Guys do you notice CT is like CFOP. Now we just need someone with 2 initials to turn 2LLSLL to 1LLSLL. How many 1LLSLL cases are there assuming EO is solved ? What do you think we can do in the beginning/ middle of to solve to reduce the cases just like how EO reduced a 4k alg count to a 500 alg count?

 

[4Chan]

Also, the algs for TTLL are awful.

Jabari and I are optimising them.

Thanks for the kind words Lex!

As someone who's used both, I think they're both tons of fun, with ZZ-A being more consistent, and ZZ-CT having better singles, full stop.

The funny thing is, after a year with both methods, I'm at the same speed.
Different reasons for the bottleneck though, and it's kinda funny, haha.

Ultimately, it comes down to identifying flaws in your style, reducing pauses, and ultimately practice more than anything leading to the best ZZ.

Just choose the most fun method and stick to it.
Cubes are supposed to be fun.
Let's just have fun.

 

[LexCubing]

Jabari and I are optimising them.

Thanks for the kind words Lex!

As someone who's used both, I think they're both tons of fun, with ZZ-A being more consistent, and ZZ-CT having better singles, full stop.

The funny thing is, after a year with both methods, I'm at the same speed.
Different reasons for the bottleneck though, and it's kinda funny, haha.

Ultimately, it comes down to identifying flaws in your style, reducing pauses, and ultimately practice more than anything leading to the best ZZ.

Just choose the most fun method and stick to it.
Cubes are supposed to be fun.
Let's just have fun.

Bruh... Tell me when you're done optimising them. I guess Learning TTLL would be put on hold, I'm assuming TSLE is also suboptimal.

I'm not trying to to start a fight here. I just really just want to know why ZBLL was more efficient. * Let's just have some fun * Sounded like you were worried this will turn into a flame war. Don't worry.