• 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 35,000+ people from around the world today!

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

Intuitive CLS for Roux

Joined
Oct 9, 2012
Messages
64
Likes
1
Thread starter #1
Hello All,

I am looking into learning CLS, but I am only interested in the cases with the slot corner on U. Unfortunately, there are 3 sets of 27 algs for each orientation of the edge, so 81 algs in total. I feel like CLS can be solved intuitively like F2L, but you'd have to know the little tricks of course. I wrote an intuitive F2L tutorial a while back that had the steps: Edge on top -> Corner oriented -> edge parallel to corner -> RUR insert. I'm curious if there is some sort of a progressive method like this for CLS.

The reason I'm interested in CLS is because I've been toying around with Roux trying to find alternative ways to do things. I think that CLS could be very useful in the following way: Solve 1x2x2 on right block, slot edge on M, CLS, pair edge with corner right before insertion. This potentially could save some moves. Essentially, you would do CO and LS using CLS and 2-3 extra moves for the edge pairing. Edge pairing can be done at any time because it doesn't interfere with the corner orientation. This is a much better alternative to pairing, insertion, and OCLL. However, there are two things CLS does not address: feasible alg count and CP.

One easy answer to the high alg count is to confine the orientation of the slot corner to one orientation, however this yields more moves. Alternatively orientation could be worked into the 1x2x2 part of the right block. A better solution (here's where intuition comes into play) is to change the orientation of the corner while creating an easy case for that given orientation.

For example, if the slot corner was oriented up and you wanted it oriented counterclockwise, you would put it in FRU and then have the choice between either R U' R' or R U2 R'. If you wanted it to be oriented clockwise, you could put it in either FLU or BLU and do either R U2 R' or R U R', respectively. Rotating U and doing an RUR move for setup would yield more options for this. At this point, you would need to know which cases you want to shoot for so you can plan appropriately when orienting the slot corner. Are the easiest/fastest CLS cases ones where there is one corner oriented, two corners oriented, no corners oriented? And how do I set up to end up with one of these easy cases? Things like this I feel can be done without the use of 81 algorithms if you know the "tricks" of CLS just like F2L.

Alongside the alg count, there is the issue of CP. CP can be done before CLS or in conjunction with CLS using F moves instead of 2-gen algs. It shouldn't make that much of a difference in move count. I have already created a CP method with an average of 2.8 moves and 6 cases that could be translated into F moves (right now it uses f moves). Again, this would multiply the total cases by 6 and only save a few moves. One fix to this problem could be to orient the slot corner while doing CP(18 cases), then do a subset of CLS(27 cases) which would be more feasible, but probably not better in move count as well as being 2-look. Another option could be doing just CLS and PLL at the end of the Roux solve. Recognizing CO and CP with one D corner on BLU would be easier than CMLL because you can see 9 colors instead of 8 (at least 2 on each corner), requiring significantly less thinking.

Let me know how/if CLS could be incorporated into Roux.
 
Joined
Jan 3, 2009
Messages
147
Likes
0
#2
I've never checked this, but would WV work? If you were to use it in Roux, might there be a few cases with a lower move count, given that you get to disregard edges (both LL edges and M slice edges) at this stage of a Roux solve?
 
Joined
Oct 9, 2012
Messages
64
Likes
1
Thread starter #7
in my opinion, finishing f2b and having corners oriented would not be too much of a benefit. it takes the same amount of time to solve those cases as it does to do most cmll cases.
I'm trying to create a two-part system that would do CP (and maybe reduce CLS cases), then do CLS with edge insertion. Right now I'm coming up with CLS algs that work for just corners so the edge insertion can be optimized (regular CLS is a subset of MGLS so it isnt just the corners). Even if the system is around the same move count, it might be advantageous to switch to it b/c it makes more of the solve alg-based instead of intuitive-based. I'll post my results when i finish.
 
Last edited:

Kirjava

Colourful
Joined
Mar 26, 2006
Messages
6,122
Likes
50
WCA
2006BARL01
YouTube
snkenjoi
#8
And what if:
1) WV (last pair + CO)
2) CP + EO
This is something that many people have considered but haven't come up with any decent results.

The problems are that EO is very short anyway and gives good lookahead into the next step, possibly negating any advantage.

Also that most advanced Roux users are doing CP+EO without doing WV first these days anyway.
 
Top