Are there any websites or is anyone working on a web site that would be similiar to bigcubes.com which would talk about solving the 6x6x6 and 7x7x7, I've already solved both, but it is the centers that took me forever, and i was wondering if they would have better ways.

v-cubes is working on solution pages

Fair enough, but for the time being, what do you think is the most efficient way to solve centers on a 7x7x7 (and 6x6x6)? At first I have just been building straight (1x5) lines and putting them together, but this feels rather slow and inefficient. Lately, I've tried another method, but I haven't done it enough yet to see if its better. I got this idea from the bigcubes.com explanation of centers on a 5x5. I made a 1x3 including the center first:

00000
00000
00100
00100
00100

From there, I built a 2x3 section in the other layers and inserted it:

00000
00000
11100
11100
11100

And another 2x3 block to finish off the 5x3 section:

00000
00000
11111
11111
11111

Then I finished the center in whatever way seemed best, usually 1x5 lines. Any suggestions or more efficient ideas? Also, is there any more efficient way to do more than one piece at time during the last 2 centers? I use a commutator-like alg right now to swap 2 centers at a time, but its also rather slow.

I learned a center swapper that i use for my variation of cage method that allows me to swap more than 2 centers at a time as long as they follow certain rules. The best part is that the algorithm is about 8-10 turns, some of those are just setting up the centers for swapping, so they can be dropped out if the centers are already set up. For an example, I can easily place a group like this on the 7x7

00011
00011
00011
00011
00011

placing ten centers, and if the group that I swap them with has the same set up, I can solve as many as 20 centers (not a typo)at a time on the 7x7x7. I learned the swapper at the Berkeley Spring 08 competition from somebody named Mike, i believe. Forgot exactly. Sorry.

If you want me to show how this works, skype me at: rubiksguy. Or if I have enough requests, I could post a youtube video

Jameson

As I posted here before, I found a mirrored older version of the v-cube site which actually has a solutions page. For the centers, see here: http://www.rxdeath.com/vc/solution/centers.html
It may be worth printing out, who knows how long it'll be there.
Hope this helps
Takrl

Hi

Commutators are not all that slow, just practice them. 8 turns to solve 2 or more centers. In less than 3 secs. If you take longer you have to work on your wrist action. Most center commutators are not "fingertrickable"

You should also work more towards a hybrid, semi-intuitive center solving approach i guess. Use commutators where you have large blocks or singular centers to deal with. Especially on the last 2 faces.

- Per

For 7x7x7 center solving, I use 1x5 rows nearly exclusively. Sometimes I make 2x5 blocks if it's easy to set up. For the last two centers, I start by solving the leftmost three rows:
11100
11100
11100
11100
11100
You always have enough freedom of movement to do this. In fact you have enough freedom to now make a fourth row every time:
11110
11110
11110
11110
11110
For the last row, all you can do is get the T-center and the two X-centers (not the two oblique ones). So this means that you can always end up with at least this:
11111
11110
11111
11110
11111
The last two (or fewer!) centers can be done with Niklas-type commutators, which are quite fast indeed.
The same kind of thing works for bigger cubes, but the best you can get with this technique changes a bit depending on the cube's size. For example on a 9x9, using this technique, the best you can do without any commutators is
1111111
1111110
1111100
1111111
1111100
1111110
1111111

When did we convert to binary?
I'm just teasing

Would someone explain the 1s and 0s to me?

The problem i have is solving the last two set of centers for both...

For the 6x6x6, for the first 4 centers i set up a 3x3x3 then a 3x3x4 and insert the last 1x1x4.

But when i only have two centers left it gets hard.

SilverPotato- the 1s and 0s represent the two different colors of the last two unsolved centers.

qqwref- Awesome, your way of doing the last two centers is way better! Also, how do you go about solving the last 4 edges? As of right now, I solve one edge at a time, working from the inside out, and then solve any potential parities only on the last edge.

Also, I think it would be really cool and appreciated by many people if bigcubes.com would expand to include 6s and 7s. I understand that V-cubes is working on solution pages right now, but it is my understanding that they would be mostly intended for those first trying to solve the cube at all, whereas bigcubes.com could go into detail on optimization, additional algs, method variations, etc. In fact, I would be personally glad and honored to write such pages for bigcubes.com once I find the optimal way to solve these things, or maybe once someone whose had these for a while (perhaps from Idaho?) were to share some insight onto this matter.

and

Same people.

And even

I bet.

Yeah, kind of tricky... Guess I might as well explain what I do. The edges method I use nowadays is called Multi-Reduction. Basically the way it works is that you solve the edges, instead of one edge at a time, in multiple steps, and each step reduces it to the edges of a smaller cube. For example... on a 7x7x7 you have five pieces to each edge. The first step (which you do just like 5x5x5 edges) would pair up the middle three pieces on each edge. Then the second step would solve the rest of the edges just like a 5x5x5. For a 6x6x6, you'd pair up the middle two like 4x4x4 edges, and then do the same second step. So it turns out that you can solve all of the edges on any NxNxN cube with a bunch of steps where each step is just like 4x4x4 edges or 5x5x5 edges. As far as I can tell it's not any less efficient than doing the edges all at once, but the recognition/lookahead is definitely better.

Anyway, if you still want to do edges your way, I'd do it one edge at a time - do stuff like 3u' flip edge 3u, solving one piece at a time for the first three edges, and then for the last edge use the parity algorithm to flip everything you have to. It takes a while though, especially that awful all-in-one parity.

Chris, what I meant to say was that many of the speedcubers who want to get faster at the bigger cubes will want to go to bigcubes.com for help. V-cubes.com may have solutions up shortly, but they will likely be very beginner and won't cover many of the more advanced concepts. Some other source of information should be made available to those who already know how to solve the cubes, but want to become faster. Bigcubes.com seems like the most logical location to put this information. It really doesn't matter who takes the time to write up the actual descriptions, so long as they get published there. (And anyway, bigcubes.com seems rather incomplete without these new cubes)

What if the guys behind bigcubes.com are providing the solutions for v-cubes.com?

Don't worry, the solutions are on the way.

Frank Morris (From Idaho)

Hi

Are they going to be like the ones here?? Or are they working on something else ??

- Per