Is VHLS+COLL bad?

[qboi]

So i watched Jperms video on algorithms worth learning, and he mentioned VHLS. After he explained what it was (solving the last F2L slot while making the cross on top), I rushed to learn it. I found a file with all the 32 algorithms. And i was planning on learning COLL and, eventually, a set of ZBLLs after. But then, after learning a few VHLS cases, i watched the rest of his explanation, and he said that the combination is bad. Can anyone explain why it would be bad?
P.S. these two have a similar number of algorithms as OLL+PLL, and yes, i do know COLL sometimes gives you an EPLL
P.S. no.2 i know all of the EPLLs

 

[Milominxx]

there's just no point of it 'cause learning oll and pll is easier and is atleast the same speed but most likely faster.

 

[qboi]

Altho i do think its is ieasier to learn VHLS than oll,since it has 20 less algorithms, and most of the OLL algorithms are confusing (at least for me). But thanks for answering, ill reconsider both options.
Another question so that i don't have post another thread for it:
In my eyes VHLS+2-look OLL seems just fine, with of course full pll.
What do you think of this combination

 

[Milominxx]

Altho i do think its is ieasier to learn VHLS than oll,since it has 20 less algorithms, and most of the OLL algorithms are confusing (at least for me). But thanks for answering, ill reconsider both options.
Another question so that i don't have post another thread for it:
In my eyes VHLS+2-look OLL seems just fine, with of course full pll.
What do you think of this combination

personally i think that if you're gonna learn vhls you should use it with coll but with 2look oll it could still be pretty good but oll and pll is still gonna be a bit faster

 

[OreKehStrah]

It’s worse. OLL is ridiculously optimized as well as PLL so the speed difference in OLL vs COLL is higher, and the speed difference between EPLL and the other PLLs is so small it’s not worth forcing.
Then you are also making your last slot solution worse to do worse OLL.
Imagine doing 5 extra moves to force a sune COLL into Z perm. Sound fun? Nope.
If you really want to do COLL use ZZ. That would be the better approach

 

[Thom S.]

I've made this post very recently but whatever.
If you care about your 3x3 times, use OLL/PLL. The most boring thing is the fastest. That's just the way it is. Most development over the whole community is dedicated to standard CFOP.
From my research(which consists of reading posts from smart people) as it currently stands, doing OCLL/PLL is less moves than COLL/EPLL on average of course.
As for me, I use VHLS/COLL/EPLL. I know, I know hypocritical, but I am at like 35 seconds after 6 years of speedcubing, so my word doesn't count.
I mostly just post this because many here just dismis anything that isn't strictly CFOP and recommend that to others and I think it's sad.

 

[IsThatA4x4]

The full sets aren't worth learning if you want to go fast, but it is definitely worth learning some algs from each set.
For example simply just learning to use a sledge/hedge (or more for some cases) to insert the last F2L pair when appropriate can allow you to skip dot OLLs pretty much all the time.
And 3/4 of the H COLLs are worth it, as they are short/easy to learn. So are some T, U, and L cases, and a couple Pi cases.
Basically just pick whatever you feel comfortable with. If you really want to learn the full sets, go ahead, no-one's stopping you.

 

[abunickabhi]

I agree with most of the discussion above. We do not need to learn the algsets entirely. That being said knowing algs from more algsets is always good.

In a solve, you should have the option of "option select" to choose from say WV+PLL or ZBLS+ZBLL and with experience, we tend to choose the faster option.

 

[qboi]

Thanks everyone for your opinions

 

[xyzzy]

Is VHLS+COLL better than 2-look OLL? Yes.
Is learning VHLS+COLL a good use of your time? No.
Should you learn it anyway? If you want to!

1. VHLS

VHLS is a method where you set up the last F2L corner and edge so that they can be solved in 3 moves (with e.g. R U' R' or R U R'), then use a single alg to simultaneously finish F2L and orient the last layer edges.

Most cases are bad, and if you're learning it from older websites, you might miss out on good algs even for the cases that aren't bad (e.g. S' R U R' S'). For the bad cases, there's barely any improvement over just solving the pair normally then doing EO on its own. There are also a handful of F2L cases where the best algs don't involve forming a pair first, and either you (i) form the pair first anyway, which is inefficient or (ii) you don't get to use VHLS in those cases.

It's still worth learning for people pursuing full ZB, but if you're the kind to shy away from learning full OLL, I doubt full ZBLL would suit your tastes.

2. COLL

COLL is a method where you solve the last layer corners while preserving edge orientation (assuming EO is already solved).

Unlike VHLS, there's a more even mix of good cases and bad cases. The ones with two twisted corners (T, U, L) are mostly good. The others (pi, H, S, AS, O) are also mostly okay if considered in a vacuum, but the normal OLL algs are already very fast and it's better on average to just do the normal OLL alg and take whatever PLL you get. (It may still be useful to learn how to avoid diag PLLs (E, N, V, Y perms).)

Without VHLS or some other early-EO method (e.g. ZZ, Petrus, LEOR), you'll only be able to use COLL one-eighth of the time, which reduces the ROI on learning it. (Crucially, it doesn't change the sign: if it was worth it before, it'll still be worth it, and if it wasn't worth it before, it still won't be. It just changes the magnitude. A lot of people miss this when they say that COLL or Winter Variation aren't worth learning because they're only applicable one-eighth of the time.)

3. but the 1/12 chance of PLL skip and 1/27 chance of OLL skip

At what cost? You're spending more time on the last slot or an earlier part of the last layer. The reason skips are notable is that when they happen without having to force them, you save a lot of time. "Forcing" skips is defeating the purpose entirely, and "forcing" skips of already-fast steps is doubly defeating the purpose.

(this is my personal beef with how a lot of methods used to be advertised as "LL skip methods")

3.5. imagine Z perms

Yeah, Z perms suck, but they're only 1/6 of the EPLLs, and they don't suck that much. You should always be looking at the big picture: does the 1/12 chance of getting EPLL skip + 8/12 chance of getting U perms offset the 2/12 chance of getting Z perm + worse OLL alg? Maybe, maybe not; it depends on the algs in question.

4. Why do you cube?

For me, cubing is a thing I do for fun, and if I want to learn an alg set, nothing's going to stop me. It doesn't matter if it's "not worth it".

I know and use almost all of COLL. Yes, I use the sunes, have been for years despite knowing it's a waste of time, and you can't stop me.

But if your ultimate goal is to get faster, you should look into smarter ways of spending your time – there are only so many hours in a day, probably most of which you can't even use for cubing. If you don't like the idea of using bad methods, then you should spend your time learning things that are more useful, like full OLL.