3x3 Corners First Help

Poke2000

Member
Joined
Jul 15, 2012
Messages
21
Location
Edmonton, Alberta
Ok, so, I'm Poke2000 and I'm kind of new here. I average 35 sec. on the 3x3, using beginner f2l, 2-look oll, and 3-look pll. I average 5 sec. on the 2x2, using CLL, and over 3 min. at the 4x4. So, since I've learned CLL, I was looking for ways to put that into other cubes, and I realized that I can solve the corners by solving it like a 2x2. So, I solve the 4 corners of 1 face and then CLL for the other 4 corners. I think 5-6 sec. to solve the corners is possible with practice. But... how can I do the edges without disturbing the corners? I only want to learn 10-15 algs. Thanks.
 
The way I do corners first is by basically doing Roux.. Start by getting the L/R centres to match the corners. Then I place the three edges on the bottom of the left face, making a 3x2 block; then the three edges on the bottom of the right face, making a block on the other side. Then I finish up the last six edges by using the LSE step of Roux.
 
Well... 've heard of Roux, but think it's a blockbuilding method, and don't like those. was JUST asking if anyone knew how to do all the (12) edges with only 10-15 algs, after solving the corners.
 
Well... 'm not sure Ell will work, since need to solve 12 corners, in different positions.


ELL won't really work, and I think 15 algorithms isn't enough, unless you do many looks. You could always use M2 :D

Counting there are 12 edges, and all of them have 2 possible orientations / parity, we get the amount of possible permutations of the edges.

(12!)*(2^12)/(2*2)=490,497,638,400


My opinion: Go for Roux.
 
Ok, so, I'm Poke2000 and I'm kind of new here. I average 35 sec. on the 3x3, using beginner f2l, 2-look oll, and 3-look pll. I average 5 sec. on the 2x2, using CLL, and over 3 min. at the 4x4. So, since I've learned CLL, I was looking for ways to put that into other cubes, and I realized that I can solve the corners by solving it like a 2x2. So, I solve the 4 corners of 1 face and then CLL for the other 4 corners. I think 5-6 sec. to solve the corners is possible with practice. But... how can I do the edges without disturbing the corners? I only want to learn 10-15 algs. Thanks.

The typical CF method goes the following way, after solving the corners:
- put L/R centers in place and solve 7 L/R edges (intuitive, no algs);
- solve the 8th edge (2 algs);
- orient M layer (1 alg);
- permute M layer (3 algs)

That is for the most basic version of the method. You can decrease the number of steps required at the expense of increasing the number of algs.

If you prefer to go for Roux, be warned that some of the CLL you know may not be usable as CMLL because you now have to preserve 6 edges in addition to the bottom corners.
 
I only want to learn 10-15 algs.

Even simplified version of CF can be found at this site (it is only in czech so i suggest you to translate it using google translator. Principal are applets, anyway.) Described system can be improved by a lot. Still, my pb avg12 is less than 19 seconds.
 
Ok, then. think i'll continue with CFOP. i was thinking of somehow putting (2x2) cll into 3x3, but it doesn't seem that good.

And... I'm really new to Corners First. I mean I've cubed since 2011, but never with a corners first method. I just wanted to know if it was possible to solve ALL THE CORNERS FIRST, and then all the edges. I guess it's not a normal corners first method, but it's the only way I could think of to put 2x2 cll in.
 
Last edited:
I wouldn't recommend this method. It's not a speedsolving method and I think you're getting lazy by implementing other methods on other cubes. I would learn a completely different method. Try learning CFOP. That's the world record method.
 
I wouldn't recommend this method. It's not a speedsolving method and I think you're getting lazy by implementing other methods on other cubes. I would learn a completely different method. Try learning CFOP. That's the world record method.

Just because CFOP is world record method doesn't mean it's best, 5BLD has UWR 1/5/12/100 with Roux.

I'm not saying Roux is better though, I'm just telling you can't compare methods based on how much records a cuber makes with that method.
 
Back
Top