7x7 centers by reduction

[AvGalen]

I also change direction sometimes so we need to change the name of this method to "growing rectangle staying as close to square as possible while if you see something nice; Use it!". From now on this can be shortened to SPAVG Centers which reflects both you and me while retaining a SPiral like name

It's a little like a freestyle continously inverted Fermat spiral, but SPAVG Centers sounds good to me

 

[shvak1]

good idea but you only really need to do that on the last two centers

 

[bodom]

There are algorithms to move just two center pieces (which is technically a three cycle move), but they are slow and not advisable for doing the last two centers.

Stefan has the best approach. It took me a bit to warp my mind around the concept of center tredges, but I'm used to it. Had to keep straight when to do 1, 2, or 3 slice moves. The problem for me now is I have to mirror some of my algorithms as I now have "left" and "right" tredges. I can also mirror them with swapping "forward" and "back" as well.

I could get away with always tredging on one side using edges, because each edge has two sides. But with centers, they are each independent, so I have to do the mirror algorithms as well.

1x5 center matching is not good for long run.

Basically, your plan should be this:
1. 3x3 centers
2. center "left" tredges
3. center "right" tredges
4. 5x5 centers (done just like 3x3 centers)
5. inner edge tredge pairs
6. outer edge tredge pairs
7. do 3x3

Your centers are like this (going one step further)

GEDFG
FABAE
DBCBD
EABAF
GFDEG

There are 7 separate centers. Each center has four possible locations (outside of the main center (C)). Your 3x3 centers match up your A, B, C centers. I then do the DE tredge pairs. Then I do another 12 pairs for the DF pairs. It's important to differentiate early on between E and F centers. Once you build your DEF tredges, then ABC becomes your super center and the G centers are just like 5x5 corner centers.

 

[StachuK1992]

I am really having alot of trouble with my new(got it today) 7x7
I have worked with it for 4 hours straight and have gotten this far on my own:
first 4 centers
quintedges
3x3
could someone please supply me with a commutator or 2 that I could play around with
a quick video would be greatly appreciated
ummm. Thanks!

 

[StachuK1992]

any1???????

 

[rjohnson_8ball]

The methods above for solving the last 2 centers are probably better for real big cubes, but I haven't mastered them yet. Instead, here is a simple algorithm to swap one cubie on the F face with one cubie on the U face:

U? ( r f ' r ' f ) U? ' ( f ' r f r ' )

The r and f indicate the particular slices that contain the bad center cubies.
Before you begin the algorithm, prepare the F and U faces so that the bad cubie on the F face would move to the bad cubie position on the U face if r is done. The first U? moves the bad cubie out of the way on the U face. The (r f' r' f) does a 6 dot pattern. Note the bad cubie on the F face has moved up to the U face. The U?' moves the bad cubie back into position. The (f' r f r') undoes the 6 dot pattern, and returns the bad cubie on the U face to the F face.

I prefer to complete the last 2 centers before connecting edge cubies.

 

[AvGalen]

I just PM'ed this to Stachuk1992. I think it is more usefull to keep this type of info on the forum:

I am making a video tutorial this weekend, but for now you can try these 2 algs:
(3R means the 3 rightmost layers. 3r means the 3rd layer from the right)
R (U' L' U) R' (U' L U)
R (F B') U2 (F' B) R'

Now try doing them on multiple layers (like 2R) or innerslices (like 2r)
3R (U' 3L' U) 3R' (U' 3L U)
3R (U' 2L' U) 3R' (U' 2L U)
3r (U' 3l' U) 3r' (U' 3l U)
3r (U' 2l' U) 3r' (U' 2l U)

3r (3F 3B') U2 (3F' 3B) 3r'
2R (3F 3B') U2 (3F' 3B) 2R'
2R (2F 2B') U2 (2F' 2B) 2R'

And slice-sunes are usefull as well:
3r U 3r' U 3r U2 3r'
3R U 3R' U 3R U2 3R'

 

[StachuK1992]

this looks farmilliar
thanks..this really helped alot...and I only tried a few so far

 

[LarsN]

I'm not likely to be one of the fastest on the 7x7 with an average of little more than 9 minutes, but here's how I do the last two centers.

At first I look for simple slice swaps. If a slice has at least 3 wrong centers and there's a similiar slice on the other face with at least 3 wrong centers, I swap them using: slicemove U2 slicemove'

I do 3 or 4 swaps, and end up having on average 3-5 centers wrong on each face. The I do single centerswaps like in this example:

U-face center:
xxxxx
xxxxo
xxxxx
xxxxx
xxxxx

F-face center:
ooooo
oooox
ooooo
ooooo
ooooo

Usings Arnauds notation
Commutator (or wosname...): 2R' F 3r' F' 2R F 3r

Often you get cases where you can switch more than one center:

U-face center:
xxxxx
xxoxo
xxxxx
xxxxx
xxxxx

F-face center:
ooooo
ooxox
ooooo
ooooo
ooooo

Usings Arnauds notation
Commutator (or wosname...): 2R' M' F 3r' F' 2R M F 3r

I'm not so sure of the 7x7 notation, but I hope it makes sense.

 

[mrCage]

Hi Arnaud

All of these are intuitive if i have understood notation correctly ...

- Per

 

[rachmaninovian]

i have to do a lot of block cycles while doing centres last i think i know all of these cases