Order of Memorization

So, I was thinking about how to make memorization more efficient. So, assuming we think about 3x3 BLD for the moment. What I do is:

Memo Edges
Memo Corners
Solve Edges
Solve Corners

I think it might be more efficient to do it like this:

Memo Corners
Memo Edges
Solve Edges
Solve Corners

You wouldn't have to spend a long time memorizing edges, and it would be a lot easier to remember during the edges solving stage. Applying this to 4x4 BLD, what I do is this:

Memo Centers
Memo Edges
Memo Corners
Solve Centers
Solve Edges
Solve Corners

and sometimes I will do this, depending on the case

Memo Edges
Memo Corners
Memo Centers
Solve Centers
Solve Edges
Solve Corners

But if 3x3 works better in an ABBA pattern, shouldn't 4x4 work better in ABCCBA? Such as:

Memo Corners
Memo Edges
Memo Centers
Solve Centers
Solve Edges
Solve Corners

Maybe I'm the only one that does ABAB and ABCABC, but what's your opinion on this?

Symmetric FTW.

The only glitch is that on 4x4 I memo and solve corners last, because I use my 3x3 method for them, and I'm unsure about the center-preserving properties for these algorithms.

Yeah, I don't know anyone (although I'm sure there are some) who do centers last (or even second). Most people just do them first to make corners a whole lot easier.

For 3x3:
I do
memo edges, memo corners, solve corners, solve edges.

For 4x4:
I would
memo edges, memo corners, memo centers, solve centers, solve corners, solve edges.

It is purely one's own personal strengths. I gave this exact same thing some thought recently and decided that ABBA with B being edges might be nicer than ABAB. Then, like you said, it would be easier to speed memo edges. Unfortunately, my corners memo is extremely fast (often, sub-7) ONLY because I do BAAB. Thus, I have to decide whether the fast edge memo outweighs fast corner memo in advantages. Also, I have to take into account that edges would be harder to recall when speed-memo'ed because of the larger number of stickers.

I'm so different

for 3x3x3:
memo flipped edges (if applicable)
memo corner permutation
memo rest of edges
memo corner orientation

solve corners (orientation then permutation)
solve edges (permute them, then flip any remaining edges)

So, I'm going to standardize this, for this discussion at least, maybe other places:

A = Corners
B = Edges
C = Centers

So it seems like a ABBA would be the ideal approach to me, because there are more edges, and you have to remember them for less time. Tyler (happa95) uses BAAB, which makes sense, so you can memorize corners super fast and solve them right away.

For 4x4, I think right now that a ABCCBA is still fastest.

@Sa967St : Is that method like a Old Pochmann/3OP Corners kind of thing?

byu said:

@Sa967St : Is that method like a Old Pochmann/3OP Corners kind of thing?

Click to expand...

not really.
corner orientation: freestyle
corner permutation: a mix of old pochmann and 3-cycle
edge permutation: a mix of old pochmann, 3-cycle and TurBo

I likes.

byu said:

Yeah, I don't know anyone (although I'm sure there are some) who do centers last (or even second). Most people just do them first to make corners a whole lot easier.

Click to expand...

serrusgod. the one who made the blind turorial for 4x4

I didn't like his tutorial. The parity fix was much too long, for one thing

the way I'm learning it it's memo corners, memo edges, solve edges, solve corners
mostly because I only have to remember max 8 corners so I can remember that longer

oh yeah, and because the parity alg is shorter

Here's what I do. Mostly I solve in reverse order to my memorization, so I can do shorter-term memory for the last ones I memorize. With 6x6x6 and 7x7x7, I have some slightly strange orders so the memorization can pack and unpack well in my Roman Rooms.

3x3x3
Memorization: edges, corner permutation, corner orientation
Solve: corner orientation, corner permutation, edges, parity
4x4x4
Memorization: edges, centers, corner permutation, corner orientation
Solve: corner orientation, corner permutation, centers, edges, edge parity, corner parity
5x5x5
Memorization: central edges, + centers, wings, X centers, corner permutation, corner orientation
Solve: corner orientation, corner permutation, X centers, wings, + centers, wing parity, central edges, corner parity
6x6x6
Memorization: outer wings, left obliques, inner wings, right obliques, outer X centers, inner X centers, corner permutation, corner orientation
Solve: corner orientation, corner permutation, outer X centers, inner X centers, left obliques, right obliques, inner wings, inner wing parity, outer wings, outer wing parity, corner parity
7x7x7
Memorization: central edges, outer wings, left obliques, inner wings, right obliques, outer X centers, outer + centers, inner X centers, inner + centers, corner permutation, corner orientation
Solve: corner orientation, corner permutation, outer X centers, outer + centers, inner X centers, inner + centers, left obliques, right obliques, inner wings, inner wing parity, outer wings, outer wing parity, central edges, corner parity

Wow, I think I'm going to have to extend my "notation" to support what Mike's doing and properly understand it.

So we have:

A: Corners
B: Central Edges (3x3 Edges)
C: X Centers (4x4 Centers)
D: + Centers
E: Inner Wings (4x4 Edges, 5x5 Wings)
F: Outer Wings
G: Outer X Centers
H: Outer + Centers
I: Left Obliques
J: Right Obliques

I'm actually now going to use Uppercase letters for Memorizing, and Lowercase for solving

Looks like Mike is doing BAab (like Tyler) for 3x3
For 4x4, looks like ECAace (again, a symmetrical order)
For 5x5, seems to be a BDECAacedb (once more symmetrical)

And on 6x6 and 7x7 (I'm not going to write it all out) it's no longer symmetrical. Why is this Mike?

Memo Edges
Memo Corners
Solve Edges
Solve Corners

I'm a BLD noob (I've only solved it 4 times out of 25 attempts) but I do this as my memo is pretty accurate, my execution is the problem Out of those 25 attempts, 10 were solves where only 2 edges were flipped. That's what I get for trying to be as fast as I can

Yay! Another BABA user, like me. You should join me in the switch to ABBA.

HAHAHA, ABBA, the pop group from the 80's. i do BAAB for 3x3, then CBAABC for 4x4. I think chris hardwick does this, and puts up a great reason why. Centres usually have more of them solved, so you memo the one that has less things to memo the hardest, so then its easier to recall the harder edges.

Rubixcubematt said:

I think chris hardwick does this, and puts up a great reason why. Centres usually have more of them solved, so you memo the one that has less things to memo the hardest, so then its easier to recall the harder edges.

Click to expand...

Yes, Chris mentioned doing this in something he wrote a long time ago. I thought it made sense, so I adopted it.

byu, I hinted why things get different with bigger cubes in my previous post. I can just barely fit 2 different sets of pieces in each of my memorization rooms. In order to fit a 6x6x6 in 3 rooms and a 7x7x7 in 4 1/2 rooms, I have to pack them in such that they fit. Since wings usually have more pieces to solve than centers, I pack each set of wings with a set of obliques. Then for 7x7x7 I pack one set of X centers with one set of + centers. Then when I unpack them, I go ahead and do the last two rooms I memorized first in the order I memorized them. I do the outer ones first and the inner ones second because the outer ones go quicker. I guess I could have memorized them last, but I didn't do that at first, so I got used to this. Then I do the obliques next because I want to be able to do the wing parity fixes as I go, rather than going back and fixing them. I should definitely do the same thing on 5x5x5 (solve X and + centers, THEN wings), but I got used to doing it the other way so I never switched. You should definitely solve all centers first, so you can fix other parities as you go. Except that I still like solving corners first, so I don't have to keep them in long-term memory.

Mike Hughey said:

Rubixcubematt said:

I think chris hardwick does this, and puts up a great reason why. Centres usually have more of them solved, so you memo the one that has less things to memo the hardest, so then its easier to recall the harder edges.

Click to expand...

Yes, Chris mentioned doing this in something he wrote a long time ago. I thought it made sense, so I adopted it.

byu, I hinted why things get different with bigger cubes in my previous post. I can just barely fit 2 different sets of pieces in each of my memorization rooms. In order to fit a 6x6x6 in 3 rooms and a 7x7x7 in 4 1/2 rooms, I have to pack them in such that they fit. Since wings usually have more pieces to solve than centers, I pack each set of wings with a set of obliques. Then for 7x7x7 I pack one set of X centers with one set of + centers. Then when I unpack them, I go ahead and do the last two rooms I memorized first in the order I memorized them. I do the outer ones first and the inner ones second because the outer ones go quicker. I guess I could have memorized them last, but I didn't do that at first, so I got used to this. Then I do the obliques next because I want to be able to do the wing parity fixes as I go, rather than going back and fixing them. I should definitely do the same thing on 5x5x5 (solve X and + centers, THEN wings), but I got used to doing it the other way so I never switched. You should definitely solve all centers first, so you can fix other parities as you go. Except that I still like solving corners first, so I don't have to keep them in long-term memory.

Click to expand...

How do u memo with rooms? do u put 2 pieces of info at each location, then go round the room twice? I think that I may adopt this 4 multi so i can have one cube per room.

Memo corners
Memo edges
Solve corners
Solve edges
---> visual memo, freestyle, pb=1:12.37 not happy when successful solves are over 2:00. approx 70% success rate.

memo edges
memo corners
solve corners
solve edges

I'm a BAAB I wanted to be ABBA!

I like it this way though, easier to remember the corners visually if I memo them last and solve them first.