What makes a LL-Skip method a true LL-skip?

[N's-cvt]

Recently I've been working and thinking about what a LL-skip means and building a method around it however I don't know what really defines a LL-skip so I just wanted to see what the speedsolvers out there think what makes a LL-skip a true LL-skip?

 

[ProStar]

I just define an LL skip as finishing F2L and being the rest of the cube being finished(assuming CFOP), even if special sets like VLS is used. Many would define it as finishing F2L with a normal insert(i.e: not trying to force any special cases) and having the rest of the cube

 

[Cuberstache]

Recently I've been working and thinking about what a LL-skip means and building a method around it however I don't know what really defines a LL-skip so I just wanted to see what the speedsolvers out there think what makes a LL-skip a true LL-skip?

ZZ-CT is probably the best possible LL-skip method

 

[GenTheSnail]

There was a really cool method around a couple years ago that was a proof of concept LL skip method. Not efficient, at like 70-80 moves, but it was a true LL skip method. I think the mods merged it with the new/concept method/idea thread though, so I cant find it atm, but maybe someone else who remembers it can.

 

[Etotheipi]

I don't know why, but ZZ-CT doesn't feel like a LL skip method to me, even though it technically is. I've always thought of it as forcing a Pseudo OCLL skip, at the cost of a larger PLL set. But I suppose it counts.

 

[ProStar]

There was a really cool method around a couple years ago that was a proof of concept LL skip method. Not efficient, at like 70-80 moves, but it was a true LL skip method. I think the mods merged it with the new/concept method/idea thread though, so I cant find it atm, but maybe someone else who remembers it can.

Is this what you remember?

 

[Hazel]

There was a really cool method around a couple years ago that was a proof of concept LL skip method. Not efficient, at like 70-80 moves, but it was a true LL skip method. I think the mods merged it with the new/concept method/idea thread though, so I cant find it atm, but maybe someone else who remembers it can.

Or maybe this? (Although this isn't really a LL-skip method, it's a 2-algorithm-1-look-last-slot-last-layer method that requires knowledge of full ZBLL...)

 

[Sub1Hour]

I'm very surprised this thread has gone for as long as it has without the mention of Cfinity.

 

[xyzzy]

I just define an LL skip as finishing F2L and being the rest of the cube being finished(assuming CFOP), even if special sets like VLS is used. Many would define it as finishing F2L with a normal insert(i.e: not trying to force any special cases) and having the rest of the cube

I think I mostly agree with your latter definition. If I do something like VHLS/ZBLS that forces EO to be done on the last slot, and I get a LL skip then, I'd call it a "ZBLL skip" instead (likewise for LL skips in a ZZ solve).

Hmm. I never really put much thought into this (because LL skips are rare, spend your mental effort on things that aren't) but thinking more about it, I guess my personal definition is more like: based on what I see while starting to finish F2L, what do I expect to be solved in LL? If I don't have any expectation of anything being solved because I'm an idiot and not looking ahead, and I get an LL skip then, that's a "full LL skip" to me. If I had been doing edge control (not necessarily just last slot) and know that there'll be EO done, that'd be only a "ZBLL skip". And so on.

Then again I almost always do edge control (and to a smaller extent, corner control) on the last slot so that would also mean "full LL skips" basically don't exist for me. I guess I'm fine with that.

---

On another note, I find that method designers have been placing undue emphasis on "LL skips" as if that's the holy grail of cubing. I have less of an aversion to considering something like square-1 CSP to be forcing parity skips since it does so without increasing the average move count of the original cubeshape step much (like less than half a move?), but to date every 3×3×3 method advertised as being an "LL skip" method fails that litmus test: either it's a method that straight-up has no last layer step at all (Roux, unless you really want to count the M slice as the last layer; 3-style) or it's a variant of a more well-known method that "distributes" the EO, CO, EP, CP aspects of the last layer into some earlier steps, making them each significantly longer.

You're not "truly" skipping the last layer then; you're just solving it at a different time.