ZZ-CT Thread

[Cale S]

TTLL alg: R U' R' (R' F R F')3 R U R'

 

[Dan Morris]

Some lower movecount solutions to of some of the edge in slot TSLE cases on gyroninjas site. Sorry if any of these are wrong/bad
1)
Old Alg: U R U R' U' R U R' U R U2 R' U' R U2 R'
My Alg: (R U' R' U') (R U' R' U) (R U2 R')

2)
Old Alg: (U R U R') (U' R U2 R') (U' R U' R') (U R U R')
My Alg: (U2 R U R') (U R U R') (U' R U2 R')

3)
Old Alg: (R U2 R' U) (R U R' U) (R U2 R' U') (R U' R')
My Alg: (R U2 R2 U2) (R U R' U) (R2 U' R')

Also if you use cyothekings G perms, you'll find this TTLL alg i found easy to learn : R' U R' U' D R2 U' R U' R' U R' U R2 U D' R2

 

[Rusca]

Is there any sort of 2 side recognition system for ttll?

Hi, I was asking myself that same question some days ago, so I thought about it and ended recording some explanations.

Maybe it helps with understanding all this recognition thing.

 

[Dan Morris]

Found another alternative TSLE alg for this case that's long, but easy to remember and fingertrick friendly, might be useful for forcing pll.
Gyroninjas Alg: (U R U R') (U' R U2 R') (U' R U' R') (U R U R')
My Alg: (U2 R U R') (U R U R') (U' R U2 R')
Alternative alg: 3x(R U R' U2) R U R'
The inverse is also kinda cool 3x(R U' R' U2) R U' R'

Also if you get the headlights OLL case as your TSLE case and the D corner is in the top left (when headlights face forward) you can do this alg to force pll. It worked everytime for me so far.
U2 3(R U R' U')

 

[phamtuanktdt]

I learn a lot from you guys. Thank you so much

 

[Dan Morris]

TSLE: U2 F (R U R' U') R' F' R

Use left index for the F' and its pretty fast

 

[4Chan]

TSLE: U2 F (R U R' U') R' F' R

Use left index for the F' and its pretty fast

Ooooh, I'll use this case along with my other case to force skips, this is NICE THANKS

 

[Dan Morris]

Is it possible to force PLL every time? learn multiple TSLE cases for every corner possible corner location, and you wouldn't need TTLL. must be like 400+ algs right?

 

[4Chan]

Yeaaap, that's ZZ-C.

It has some pretty gnarly cases for 4 and 5 corner misoriented, so that's why I think ZZ-CT is better on average.

 

[JTWong71]

Is it possible to force PLL every time? learn multiple TSLE cases for every corner possible corner location, and you wouldn't need TTLL. must be like 400+ algs right?

There is also the minor set back that after implementing some of those longer algorithms for ZZ-C, you will end up with a 15+ move PLL Diagonal Corner Swap, probably making Last Slot Layer Layer around 30+ Moves. I think the good thing would be to learn 3 different permutation algorithms for each TSLE case, allowing for an always 2-Gen TTLL (Including the y' rotation subset to make pseudo 2-Gen) and choices to transition to preferred cases like in 4Chan's Advance ZZ-CT Tricks video. The problem with this is of course algorithm count and recognition, possibly movecount, but not as much as ZZ-C in some cases.

 

[4Chan]

I came up with a new subset for ZZ-CT!
I call it TS3S or Tran Style 3rd Slot!

I've found that doing the backslots FIRST is better, because you won't have any pieces in your blind spot, leading to cleaner and faster solves.

So like, after your first two slots in back, what if the FL edge is already placed?
You have the following options with ZZ-CT:

1. Traditional: Finish 3rd LEFT slot and do TSLE and TTLL (3 steps)
2. Intermediate: Finish RIGHT slot, rotate and do one of the edge in place TSLE and TTLL (3 steps)

Option 1 is okay, but sometimes Option 2 has better lookahead and lower moves SOMETIMES.

So TS3S is an option that's only 2.5 steps!

TS3S involves just making the F2L pair for FR slot and then doing a special alg that inserts F2L and orients everything (includes FLD corner), skipping TSLE and going straight into TTLL!! Only 81 algs!

It is, however, super circumstantial, only for advanced ZZ-CT

 

[Y2k1]

So its tsle but doing it with the third pair when 4th edge is inserted? Sounds super cool, will learn after I finish tsle.

 

[genericcuber666]

4chan stop making these super cool 200+ alg subsets and please make a more viable version of ct so that more people will start learning.
because most of us are to lazy to even learn full tsle...

 

[Y2k1]

4chan stop making these super cool 200+ alg subsets and please make a more viable version of ct so that more people will start learning.
because most of us are to lazy to even learn full tsle...

This is only 81 algs, not a full 200+ subset, amking more viable than tsle itself for "lazier" people

 

[obelisk477]

from another user in another thread:

tbh it's a mediocre, overhyped, & dishonest method

1) they say your expected movecount is 40-50 htm.

TSLE = 10.37
TTLL = 15.21
AUFS = 3*3/4

that means your LS & LL is 28 moves on average

that means you have to build your EOF2L-1 in 12-22 moves in order to meet their expected movecount of 40-50 moves listed on the wiki. that's especially bogus considering you're gonna leave your FR slot for last.

2) they say you get a LL skip

so do people in the roux method - this is a silly claim

in ZZ-ct you have

-tsle (1 look)
-ttll (1 look)

in a regular ZZ solve you have

-last pair (you should see this during your 2nd/3rd pair - it's doesn't even deserve it's own look)
-oll/coll (1 look)
-pll/epll (1 look)

you aren't skipping the LL - you are only skipping your 4th pair look-ahead lol

3) they say it's better than WV/SV

but for WV:

set up case = ~6
WV = 8
PLL = 11.8 moves
AUF = 4*3/4

even if you round up your LS & LL is only 30 moves, that's only 2 more than ZZ-CT.....
with 4x fewer algs (and everyone already knows them all)
better recognition



tl;dr in my opinion, it's not good at all

 

[Cale S]

from another user in another thread:

tbh it's a mediocre, overhyped, & dishonest method

1) they say your expected movecount is 40-50 htm.

TSLE = 10.37
TTLL = 15.21
AUFS = 3*3/4

that means your LS & LL is 28 moves on average

that means you have to build your EOF2L-1 in 12-22 moves in order to meet their expected movecount of 40-50 moves listed on the wiki. that's especially bogus considering you're gonna leave your FR slot for last.

2) they say you get a LL skip

so do people in the roux method - this is a silly claim

in ZZ-ct you have

-tsle (1 look)
-ttll (1 look)

in a regular ZZ solve you have

-last pair (you should see this during your 2nd/3rd pair - it's doesn't even deserve it's own look)
-oll/coll (1 look)
-pll/epll (1 look)

you aren't skipping the LL - you are only skipping your 4th pair look-ahead lol

3) they say it's better than WV/SV

but for WV:

set up case = ~6
WV = 8
PLL = 11.8 moves
AUF = 4*3/4

even if you round up your LS & LL is only 30 moves, that's only 2 more than ZZ-CT.....
with 4x fewer algs (and everyone already knows them all)
better recognition



tl;dr in my opinion, it's not good at all

I agree with a lot of this, but in my opinion the purpose of ZZ-CT is to have a 2-look LSLL every time with as low of an alg count as possible while also having decent recognition. It also just looks cool

 

[genericcuber666]

from another user in another thread:

tbh it's a mediocre, overhyped, & dishonest method

1) they say your expected movecount is 40-50 htm.

TSLE = 10.37
TTLL = 15.21
AUFS = 3*3/4

that means your LS & LL is 28 moves on average

that means you have to build your EOF2L-1 in 12-22 moves in order to meet their expected movecount of 40-50 moves listed on the wiki. that's especially bogus considering you're gonna leave your FR slot for last.

2) they say you get a LL skip

so do people in the roux method - this is a silly claim

in ZZ-ct you have

-tsle (1 look)
-ttll (1 look)

in a regular ZZ solve you have

-last pair (you should see this during your 2nd/3rd pair - it's doesn't even deserve it's own look)
-oll/coll (1 look)
-pll/epll (1 look)

you aren't skipping the LL - you are only skipping your 4th pair look-ahead lol

3) they say it's better than WV/SV

but for WV:

set up case = ~6
WV = 8
PLL = 11.8 moves
AUF = 4*3/4

even if you round up your LS & LL is only 30 moves, that's only 2 more than ZZ-CT.....
with 4x fewer algs (and everyone already knows them all)
better recognition



tl;dr in my opinion, it's not good at all

Not really
1) it does have a low move count do some solves and check
2) ok youre Kinda right but after even if it's not a ll skip it's still a 1lll
3) it has higher skips and no setup so one less look

 

[mDiPalma]

1) i have done tens of thousands of ZZ solves, i have checked

consider, for example, ZZ-a's expected movecount: 44 moves
Last F2L pair = 6.7 moves
ZBLL = 12 moves
AUF = 2*3/4

that leaves 24 moves for f2l-1 (f2l with the freedom to do any slot at any time and do entire blocks at a time [1x1x3->1x2x3])

now explain to me how you plan to get 12-22 moves doing a restricted version of f2l-1? bogus lol

Spoiler: here's an actual zz-ct movecount approximation

assuming you're gonna leave FR for last (or BL and do a y2)

eoline = 6.13
full block = 11.44
square = 6.61
TSLE = 10.37
TTLL = 15.21
AUF = 3*3/4

= 52 moves on average

for reference, CFOP is ~55 moves.

the only way you're gonna get 12 moves for EOF2L-1 is with premoves on the inverse scramble lol


3) You're right, WV has an extra step, but it has 4x fewer algs and is not many more moves. Also ZZ-ct actually has 5x lower skip chances than WV


I'm not here to debate - the variant is generally decent, but largely overrated. However, anytime hard work and dedication is put into something, good things are bound to happen. Good luck to whoever uses this variant.

 

[Cale S]

this TSLE alg seems better than what's on the site

R U' R' U2 R2 D R' U' R D' R2

alg listed is
U R U R' U' R U U R' U' R U' R' U R U R'

 

[AlphaSheep]

I'm not here to debate - the variant is generally decent, but largely overrated.

I agree with this. It's just one more viable ZZ variant out of many. Honestly I even think it's advantage over simple ZZ+COLL+EPLL is marginal. I like the method because it looks cool, but the hype after the US Nats seminar caught me by surprise.

I've learned about half of TSLE and about a third of TTLL. I don't know them well enough to speed solve yet (put off learning the rest to practice for back-to-back comps), but I've done tons of slow solves which I've reconstructed. The move count for a normal solve does seem to be about low to mid 50s (I average high 50s in slow solves with ZZ+COLL+EPLL, and about 3-5 moves more in speed solves). Sure, there are lucky solves with sub-40 move solutions, but they are balanced out by the nasty cases where theres an AUF, 12 move TSLE, AUF, ADF, 18 move TTLL, AUF, ADF for 35 moves for LS+LL alone...

I also think the idea of "skips" in ZZ-CT is a little exaggerated. A lot of skips are like a CFOP solver claiming an EP skip when they get an E perm. Sure, its a skip in a sense, but many of the best cases occur when there is no "skip".

• If the TTLL case is a PLL (1 in 5) then it feels like a skip because you're much more familiar with the PLL algs than the TTLL algs. If you practiced the TTLL algs until they become as natural as PLLs, I'm sure a V perm wouldn't feel at all like a skip.
• Another example is that the TSLE case is an OCLL (1 in 15 chance), then that feels like a skip, even though 2/3 of those are cases I'd use a 12 move alg for (i.e the worst cases). Sure Sune and Antisune feel super fast, but there are another 20 something other 7 move TSLE algs the same length that are just as fast. So even Sune isn't that special.

Genuine skip chances are:

• 1 in 405 for TSLE
• 1 in 360 for TTLL

Every other "skip" is not really a skip but just an easy case.

That said, ZZ-CT is a cool method and it feels really fast (in a subjective sense). I don't think a person using ZZ-CT would be at any major disadvantage over users of other ZZ variants. I'm planning on finishing learning all of TTLL and TSLE over December, so I'll see if my opinion changes after that.