• Welcome to the Speedsolving.com, home of the web's largest puzzle community!
    You are currently viewing our forum as a guest which gives you limited access to join discussions and access our other features.

    Registration is fast, simple and absolutely free so please, join our community of 40,000+ people from around the world today!

    If you are already a member, simply login to hide this message and begin participating in the community!

Names of the Algs in the different subsets

MagicVince

Member
Joined
Sep 2, 2019
Messages
43
Location
Paris (France)
Hello,

I am not sure I am posting in the right thread.
For the website I introduced here, I am looking for "official" names for the different algorithms of the standard subsets.
For instance, the 57 OLL numbering is standard (even if it seems arbitrary to me, do you know who numbered them and on which criteria?), the names of the PLL are pretty standard too.
But what about the 42 CLL or the 29 ELL? I have seen different websites, and the names or the numbering do not seem consistent. Are there any official standard names? Or at least well accepted names?
I have the same question for LLEF or L4C, for instance.
Thanks,

Vincent.
 

MagicVince

Member
Joined
Sep 2, 2019
Messages
43
Location
Paris (France)
OLL and PLL are not an issue: I am more concerned with CLL and ELL for instance. There are reasonably sized sets (compared to ZBLL for instance), but probably because there are not as widely used as OLL and PLL I have difficulties to find standard names.
Let's take another example: COLL (42 algs).
- http://www.cyotheking.com/coll (no names, just the OCLL ones)
- https://www.cubeskills.com/uploads/pdf/tutorials/coll-algorithms.pdf (same)
- https://www.speedsolving.com/wiki/i...DB?mode=view&view=default&puzzle=3&group=COLL (here they are numbered but I don't know if this numbering is arbitrary, official, widely accepted)
 
Last edited:

Timona

Member
Joined
Feb 19, 2019
Messages
1,131
Location
Northampton, UK
YouTube
Visit Channel
Oops. This alg:
View attachment 20300
is called "BR Ub" in SpeedCubeDB (which makes sense) and "BF-Flip Ua" in AlgDB (which makes no sense). Should I expect more inconsistencies or mistakes?
From what I understand with SpeedCubeDB's naming system, if it's a U-perm, either a or b, then the "complete" side is placed at B (regardless of whether it has a flipped edge). I think AlgDB has the inconsistencies

1659978345952.png

According to SpeedCubeDB's system this case should be called RF Ua.
 

OreKehStrah

Member
Joined
May 24, 2019
Messages
1,048
YouTube
Visit Channel
Oops. This alg:
View attachment 20300
is called "BR Ub" in SpeedCubeDB (which makes sense) and "BF-Flip Ua" in AlgDB (which makes no sense). Should I expect more inconsistencies or mistakes?

Speedcubedb has cases labeled incorrectly (like squan U perms being named swapped).

Basically no alg set aside from PLL and to a lesser extent OLL have standard naming.

CxLL sets have commonly used named but lots of people don’t know them.
 

MagicVince

Member
Joined
Sep 2, 2019
Messages
43
Location
Paris (France)
Thank you very much for your answers. I will update my website soon with ELL and "rationalized" CLL and COLL. I found some inconsistencies between the CLL and COLL names that were given as a reference, So I had to make arbitrary choices. And I noticed, there is another way of naming them (which is more descriptive) for the CMLL subset (for instance here). I am convinced there is a need for standardization. I found this thread about OLL naming interesting. Another thing that would need some renaming attention is the name of the subsets themselves: OELL vs EOLL, COLL vs OCLL etc. Seriously... ;)
 
Joined
Aug 12, 2013
Messages
3,206
Location
Brazil
YouTube
Visit Channel
Another thing that would need some renaming attention is the name of the subsets themselves: OELL vs EOLL, COLL vs OCLL etc. Seriously... ;)
what's your suggestion? 🤔
I believe we are very used to "Edges of last layer", "Orient edges of last Layer","Corner of last layer" and "Orient corners of last Layer"...
 

MagicVince

Member
Joined
Sep 2, 2019
Messages
43
Location
Paris (France)
I would suggest slight modifications like O1-LL - Orient 1st the Last Layer- instead of OLL (meaning that Orientation is the first phase of your method, thus disturbing permutations, something that OLL does not say). O1-LL would be subdivided into O1-E1-LL (instead of EOLL) and O1-C2-LL (instead of OCLL).
Another example would P2-C2-LL for CPLL and P2-E2-LL for EPLL, the number 2 underlying that CPLL and EPLL can be done in any order on relatively to the other (there is no number one) and thus in particular at the end of the process (and additionally they preserve orientation since the P for Permutation is the second step). P2-C1 would be an algorithm that does not disturb orientation (since Permutation is number 2) that would fix the corner permutation (its a Px-Cx) but could potentially disturb the edge permutation since C1 is a first phase whose side effects can be fixed by the next phase (like a T-perm to swap two corners).
LLEF (Last Layer Edge First) would be E1-LL.
I will try to make a cartography of the current naming along with my suggestion.
 
Last edited:

OreKehStrah

Member
Joined
May 24, 2019
Messages
1,048
YouTube
Visit Channel
I would suggest slight modifications like O1-LL - Orient 1st the Last Layer- instead of OLL (meaning that Orientation is the first phase of your method, thus disturbing permutations, something that OLL does not say). O1-LL would be subdivided into O1-E1-LL (instead of EOLL) and O1-C2-LL (instead of OCLL).
Another example would P2-C2-LL for CPLL and P2-E2-LL for EPLL, the number 2 underlying that CPLL and EPLL can be done in any order on relatively to the other (there is no number one) and thus in particular at the end of the process (and additionally they preserve orientation since the P for Permutation is the second step). P2-C1 would be an algorithm that does not disturb orientation (since Permutation is number 2) that would fix the corner permutation (its a Px-Cx) but could potentially disturb the edge permutation since C1 is a first phase whose side effects can be fixed by the next phase (like a T-perm to swap two corners).
LLEF (Last Layer Edge First) would be E1-LL.
I will try to make a cartography of the current naming along with my suggestion.
This sounds unnecessary and confusing
 

MagicVince

Member
Joined
Sep 2, 2019
Messages
43
Location
Paris (France)
OK-OK, I understand you are not easy to convince :).
Of course, if you are familiar with CFOP, you know what OLL and PLL mean, and if you ever wanted to simplify anything, you would rather put efforts on simplifying the 57 OLL names than these simple two phases.
What I pointed out was only the inconsistency of the naming conventions. There mainly three 2-look methods for last layer: OLL/PLL, CLL/ELL and LLEF/L4C (I don't think any one begins with positioning all the cubbies and orient them afterwards, but I can be wrong).
I am just saying it would be more logical to call them O1LL/P2LL, C1LL/E2LL and E1LL/C2LL. Then, if one or both of these two main phases are decomposed to make a 3-look or 4-look method, the same logic can still apply.
I made a mapping of the current names (let me know if I made any mistake, I know ZBLL is missing but I did no want to have arrows crossing :)). I added the number of algorithms so that we are sure we are speaking about the same subset.
1660166108958.png
Would you say it is obvious to understand the phases from their names?
Here is what I was suggesting:
1660166228846.png
I failed to fit COLL, ZBLL and of course 1LL in this naming convention. But the other ones are rather easy to understand.
P2LL (Permute second = PLL) can be split into P2C1LL (Permute Corner First after full orientation = CPLL) and P2E2LL (Permute Edge after corners, after full orientation = EPLL). Note that P2E2LL is the same as E2P2LL (Permute Edge after orienting them, after corner full solving) but I don't know if it can be generalized.
Does it make more sense, now?
 

OreKehStrah

Member
Joined
May 24, 2019
Messages
1,048
YouTube
Visit Channel
OK-OK, I understand you are not easy to convince :).
Of course, if you are familiar with CFOP, you know what OLL and PLL mean, and if you ever wanted to simplify anything, you would rather put efforts on simplifying the 57 OLL names than these simple two phases.
What I pointed out was only the inconsistency of the naming conventions. There mainly three 2-look methods for last layer: OLL/PLL, CLL/ELL and LLEF/L4C (I don't think any one begins with positioning all the cubbies and orient them afterwards, but I can be wrong).
I am just saying it would be more logical to call them O1LL/P2LL, C1LL/E2LL and E1LL/C2LL. Then, if one or both of these two main phases are decomposed to make a 3-look or 4-look method, the same logic can still apply.
I made a mapping of the current names (let me know if I made any mistake, I know ZBLL is missing but I did no want to have arrows crossing :)). I added the number of algorithms so that we are sure we are speaking about the same subset.
View attachment 20322
Would you say it is obvious to understand the phases from their names?
Here is what I was suggesting:
View attachment 20323
I failed to fit COLL, ZBLL and of course 1LL in this naming convention. But the other ones are rather easy to understand.
P2LL (Permute second = PLL) can be split into P2C1LL (Permute Corner First after full orientation = CPLL) and P2E2LL (Permute Edge after corners, after full orientation = EPLL). Note that P2E2LL is the same as E2P2LL (Permute Edge after orienting them, after corner full solving) but I don't know if it can be generalized.
Does it make more sense, now?
I understood what you meant but I don't see the point in making these really long names like P2C1LL + P2E2LL when you can just say CP EP 2LPLL. If you make the case that someone might not understand what CP EP means, one can just as easily make that same argument back at P2C1LL + P2E2LL only now it's longer and has numbers thrown in.
 

MagicVince

Member
Joined
Sep 2, 2019
Messages
43
Location
Paris (France)
one can just as easily make that same argument back at P2C1LL + P2E2LL
Well, assuming that one knows what O, P, C, E and LL mean, my hope was that by seeing P2E2LL, one would understand that :
  • P: the current main phase is P (Permute)
  • 2: this is the 2nd phase (Permute 2nd), meaning you did the opposite - Orientation - before in the 1st phase (so all is already well oriented and your current permutation algorithm won't disturb what you did previously, that is orientation)
  • E: what you are currently permuting are Edges
  • 2: this is the 2nd sub-phase (or subset), meaning that you took care of the opposite of Edges - Corners - before in the current phase, so your edge permutation algorithm will not either disturb what you did before in the current phase, that is the corners permutation (since the second phase is P, Permutation).
So basically the letters say what you are doing, as in the current naming. The numbers add some crucial information: they say what you can disturb or cannot not disturb. It is self-explanatory.
With the same two letters and two extra numbers you know what you phase or algorithm is doing without the need to know to which method this phase belongs to (for instance you have to know that OLL+PLL are part of CFOP to understand that OLL don't care about corners whereas PLL preserve edges).

Example of precision you have to add, taken from the ELL page of this wiki:
"This page is about the ELL that solves edges preserving corners, for the method that does not, see LL edges first"
ELL fails to say something important: it preserves corners. This is what E2LL says.

So for me the advantage is obvious. Now, I agree that the fact of increasing the length of the names only for that is arguable.
 
Last edited:

xyzzy

Member
Joined
Dec 24, 2015
Messages
2,669
Re: systematic naming for last layer substeps:

One deficiency with the proposed scheme is that it cannot handle methods that aren't O-P, P-O, C-E or E-C. For instance, CPEOLL into 2GLL (CP+EO then CO+EP) cannot be described by this scheme, and as you've already noted, COLL and ZBLL don't fit into this scheme either.

(Side note: defining "permutation" sensibly is nonobvious. You can have the edges oriented and permuted, and the corners oriented and permuted, but the last layer still unsolved: an H perm.)

But besides that, it is too verbose for casual conversation. If I say "OLL" and "PLL", it's nearly guaranteed to be understood to be orientation first, permutation second, because that's a much better method than the other order. The algs are better, the recognition is better. When it comes to "CLL" and "ELL", that's more of a mixed bag, but corners first has priority simply for having been proposed earlier. Nobody is going to go out of their way to say "E2LL" when they can say "ELL" and be understood just as well.

Another problem with systematic naming is that it makes you sound like a robot. (… Or a mathematician.) I consider precision in speech to be a virtue, but sometimes the precision gets in the way of delivering a point; you're bogging down the audience with details that either don't matter or are already understood.
 

MagicVince

Member
Joined
Sep 2, 2019
Messages
43
Location
Paris (France)
Thank you for you detailed and constructive answer.
One deficiency with the proposed scheme is that it cannot handle methods that aren't O-P, P-O, C-E or E-C. For instance, CPEOLL into 2GLL (CP+EO then CO+EP) cannot be described by this scheme, and as you've already noted, COLL and ZBLL don't fit into this scheme either.
This true. I did not know CPEOLL/2GLL, Thank you for pointing out this other counter-example.
(Side note: defining "permutation" sensibly is nonobvious. You can have the edges oriented and permuted, and the corners oriented and permuted, but the last layer still unsolved: an H perm.)
True again, but I don't think this is an issue. It translates into more algorithms in some case like L4C (C2) vs CLL (C1).
Another problem with systematic naming is that it makes you sound like a robot. (… Or a mathematician.)
Touché! I guess this could perhaps helps the beginners that do not have the insights you mentioned, but obviously it can bother the others.

OK, I understood and agrees with your points. If these names are here to stay, I think I will make the mapping above (the "LL algs subsets" one with the legacy names) more comprehensive. What would be the best section in the forum to discuss such a mapping?
 
Top