Kirjava
Colourful
I read your post and understood it the first time, I'd prefer not to be told to read it again when my reply was perfectly valid.
VHF2L, COLL, & EPLL
- Thread starter EnterPseudonym
- Start date
eastamazonantidote
Veteran
[rant]Sorry? VHF2L and COLL are the parts of the solve slower than Fridrich, and that only slightly if you know what you're doing. VHF2L is clearly slower, but I think of it as ZZ with the EO near the end. No one gets on EOLine for being a slower start than the cross, so why get on VHF2L for that, especially when it's more fingertricky. Recognition for COLL is often faster than OCLL recognition (and always faster than OLL recognition) and the algs are very similar in speed (except a few odd cases and the Sunes). But then EPLL (or a skip) makes up for all of that. Recognition is instantaneous (most of the time I know that case before I'm done with COLL), execution almost as fast, and AUF is nothing to worry about.
My times with Fridrich are in the 21-22 range while my VH times are 19-21 (large range because of those few bad cases). I've probably put more work into VH, yes, because my goal is ZBLL, but I have spent a lot of time of Fridrich and I haven't been all that impressed. [/rant]
COLL takes more time to learn (mostly recognition for me- the algs are easy) and VHF2L is hard to practice. I would also note that you should learn OLL if you're going to learn the VH system because there are many cases that, erm, suck and that you're going to want to avoid. You should also know full PLL for that 1/27 OLL skip chance.
Sa967St
Not A Moderator
- Joined
- Mar 11, 2008
- Messages
- 3,793
- Location
- Waterloo, ON, Canada
- WCA ID
- 2007STRO01
- YouTube
- Visit Channel
13.64, 13.84, 15.34, 13.34, (12.43), 15.43, 15.77, (15.80), 13.78, 14.21, 13.09, 13.68 => 14.21avg12 (σ = 0.90)
It's not too great for 3x3 speed unless you only use VHF2L and COLL when you get easy/fast cases. It's pretty nice imo for the 3x3 stage of 5x5 and 7x7 (if you use reduction).
It's not too great for 3x3 speed unless you only use VHF2L and COLL when you get easy/fast cases. It's pretty nice imo for the 3x3 stage of 5x5 and 7x7 (if you use reduction).
Agreed.
Last edited:
EnterPseudonym
Veteran
how many VH algs did you learn? i was only planning on starting with pre-built pair algs, then maybe adding more.
eastamazonantidote
Veteran
I learned something like 80 COLLs as I was trying to find a decent recognition system, but I only use the standard 40 now. I learned the 32 base VHF2L cases but found that it wasn't enough. There are some really bad cases for VHF2L that really need ZBLL to do effectively, so I learned those. That was 36 more cases. Depending on the solve I skip some of the really nasty cases and just insert and OLL (which I count as a VHF2L subset). Learning the standard set of algs is fine, but you will eventually figure out that some cases need extra work.
trying-to-speedcube...
Veteran
How is COLL recognition faster than OCLL recognition?
That makes no sense to me, since OCLL is a subset of COLL. That said, COLL recognition is very nice, and ELL is much easier to see than PLL.
Weston
Premium Member
This doesnt really answer your question, but sometimes i recognize the OCLL case using COLL recognition. like if i see a checkerboard pattern on the top of the corners, i know that its pi.
trying-to-speedcube...
Veteran
Ah, I see. But still, isn't COLL recognition usually just OCLL recognition and then recognizing the permutation?
Weston
Premium Member
Yup usually.
Well, you kind of learn to see everything at once.
Its is slower in general though i think
Well, you kind of learn to see everything at once.
Its is slower in general though i think
4Chan
Premium Member
Escher
Babby
VHF2L + COLL + EPLL isn't faster than LS + OLL + PLL in my experience. I reckon you're best off learning OLL + PLL first and practicing it a lot, before moving on and adding COLL to your repertoire, and then if you really want to, branching out to VH/ZBF2L and ZBLL.
Imho, learning OLL and then just adding algs for extra angles and different CPs is more cost effective than learning COLL in it's entirety.
Imho, learning OLL and then just adding algs for extra angles and different CPs is more cost effective than learning COLL in it's entirety.
Forte
Veteran
- Joined
- Jul 12, 2009
- Messages
- 1,151
- Location
- Waterloo, Ontario
- WCA ID
- 2009SHIN02
- YouTube
- Visit Channel
Well there is a pi case like that
but I agree with the above. Sometimes you can just look at the colours and deduce, but most of the time it's slower. (It might even be slower anyway)
FatBoyXPC
Veteran
What about for those who are still 2LOLL? I realize that there are only 57 OLLs and most of them have triggers that make them easy, but there are still those (like me) who worked too much to have time to do everything in their day, improve F2L, and learn a new OLL every day (I tried to learn the 6 dot cases before Indiana Winter 2010 but that didn't work out so well). Now I just do partial edge control whether it's a built pair or the RUR' case (I just RU'R2FRF'U2R'FR'F'). My point though is since I'm already going to orient edges, then corners, why not make the most of it and learn COLL and then I only have to remember H, Z, Ua, Ub? Yes sometimes I'll get an OLL skip but I already know all 21 PLLs.
Forte
Veteran
- Joined
- Jul 12, 2009
- Messages
- 1,151
- Location
- Waterloo, Ontario
- WCA ID
- 2009SHIN02
- YouTube
- Visit Channel
If you want to cut down on effort, OLL will be easier. COLL requires you to know a different type of recognition that needs more repetitive practice than OLL, where you can just look at the shape.
FatBoyXPC
Veteran
Fair enough. Somebody mentioned COLL can help learning 2x2 CLL easier, couldn't I just use all COLL algs on a 2x2 anyway? Solve a layer then coll? I think if that was the case at least I could argue it's "more worth it."
EnterPseudonym
Veteran
yes but optimally you would learn CLL instead of COLL for 2x2
rubiknewbie
Veteran
- Joined
- Jun 27, 2009
- Messages
- 593
For a 2-handed solve, I find the 20 seconds for a normal solve too rush for the extra recognition and moves for VHF2L + COLL, though I have not tried it for 2-handed. Especially if my U, H and Z are not way faster than my other perms.
However, I try it for OH as the U, H and Z (via COLL) are very nice for OH and edge-oriented OLL (via VHF2L) are also very nice for OH, relative to other cases generally. Sometimes you are rewarded more than other times.
The best case scenario is when you don't need to do much if anything for VHF2L and it leads to a easy to recognise and execute COLL and then a U perm which would normally have resulted in a bad OLL and like a E perm.
However I only learn full COLL for the L, T, U and H cases as sune, anti-sune and Pi each has an outstanding algorithm which I always use and would not sacrifice time for.
I look at VHF2L as 16 cases as I ignore mirror but I am only learning 14 of the 16 cases (not the dot cases which I convert to 2-edge orientation).
At the same time I am also learning Winter variation (forced OLL skip) as 4 of the 16 VHF2L cases (8 of 16 if you add moves) can easily convert to WV cases. When I do timing for last slot + last layer often the most dramatic improvement is when I get a nice WV + lucky perm.
However, I try it for OH as the U, H and Z (via COLL) are very nice for OH and edge-oriented OLL (via VHF2L) are also very nice for OH, relative to other cases generally. Sometimes you are rewarded more than other times.
The best case scenario is when you don't need to do much if anything for VHF2L and it leads to a easy to recognise and execute COLL and then a U perm which would normally have resulted in a bad OLL and like a E perm.
However I only learn full COLL for the L, T, U and H cases as sune, anti-sune and Pi each has an outstanding algorithm which I always use and would not sacrifice time for.
I look at VHF2L as 16 cases as I ignore mirror but I am only learning 14 of the 16 cases (not the dot cases which I convert to 2-edge orientation).
At the same time I am also learning Winter variation (forced OLL skip) as 4 of the 16 VHF2L cases (8 of 16 if you add moves) can easily convert to WV cases. When I do timing for last slot + last layer often the most dramatic improvement is when I get a nice WV + lucky perm.
eastamazonantidote
Veteran
For many cases there is only one case that the COLL could belong to. Believe it or not, looking at the front and top faces is faster than looking at the top a side. Also, you are not looking for a specific color (yellow most of the time) but pairs of colors, which your eyes can pick up much faster. There are several overlapping cases (like in Pi and H or in T and U) but for the most part it is faster to find the sticker pairs on U and F than to find the shape from U and the orientation from R, L, F, or B.
joey
Member
You cannot always recognise COLL from just U/F though, just like OCLL.
