Okay, I came up with something new yesterday, it has a few similarities with belt/equator method but I couldn't find any identical versions. So basically my first question is, is it a known method already? I'm not putting it into the guides category because it's not a guide (just a basic outline that I just came up with) and it's more of that question to begin with. Basically here are the steps. Because I'm used to it that way, let's say that the top color is yellow and the bottom color white.:
1. Do middle layer intuitively of course.
2. Now, the bottom and the top layer will only contain cubies that have either white or yellow along with the other colors (obv). Now, for this method to have a meaning, you will need to consider both as the same color. That is, suppose you're looking at the yellow face (top). We will consider it solved if it contains only yellow or white in it ; it shouldn't necessarily contain only the face color. More on step 3. Now, step 2 is what stumbles me about this method; you will need to bring either the top or the bottom face into an OLL state (can't word it better). That is, if we go and apply one of the fifty-so OLL algs, it will be "solved" (that is, the face will contain either yellow or white, not actually solved).
Note that if you bring one face into the aforementioned state (let's call it OLL state), the second one will be into that state too, so it doesn't matter which one you choose.
I'm trying to find a more efficient way to do this; for example, one time I managed to actually "solve" one top/bottom face, and then applying the OLL to the other one. Of course, this isn't always possible, so I'm working on that step. I don't know how would one go about it without full OLL knowledge.
3. Now, this step is like solving the last two centers on the 5x5. Kind of like what is shown here
http://cube.garron.us/big_cubes/L2C/
however I found that not all cases match and one didn't even work. There's a good chance I made a mistake though so don't quote me on that. If the cases don't match, then I imagine it will be really easy to come up with algorithms for all of the cases (select few), so no problem here. For now I do it intuitively with double turns.
Note: During the second and third step, I use double turns exclusively so I can't really destroy the middle layer (just swap a few corners maybe). I don't worry about that during step 2, however, after finishing step 3, the middle layer HAS to be solved. So if I use algorithms for step 3, it should also be solved after step 2.
4. The rest is obvious, just apply PLLs to the top and bottom layer. Likely there will be parity (although it isn't really parity, just the cube isn't solved) and if there is, I solve the two vertical layers (using PLLS) and put all the "bad" edges (in my experience only edges are the problem) into one layer, then do moves like M' U2 M to finish it. I don't think I have to say more. I move edges around with PLLs obv.
So, is this a known thing?
1. Do middle layer intuitively of course.
2. Now, the bottom and the top layer will only contain cubies that have either white or yellow along with the other colors (obv). Now, for this method to have a meaning, you will need to consider both as the same color. That is, suppose you're looking at the yellow face (top). We will consider it solved if it contains only yellow or white in it ; it shouldn't necessarily contain only the face color. More on step 3. Now, step 2 is what stumbles me about this method; you will need to bring either the top or the bottom face into an OLL state (can't word it better). That is, if we go and apply one of the fifty-so OLL algs, it will be "solved" (that is, the face will contain either yellow or white, not actually solved).
Note that if you bring one face into the aforementioned state (let's call it OLL state), the second one will be into that state too, so it doesn't matter which one you choose.
I'm trying to find a more efficient way to do this; for example, one time I managed to actually "solve" one top/bottom face, and then applying the OLL to the other one. Of course, this isn't always possible, so I'm working on that step. I don't know how would one go about it without full OLL knowledge.
3. Now, this step is like solving the last two centers on the 5x5. Kind of like what is shown here
http://cube.garron.us/big_cubes/L2C/
however I found that not all cases match and one didn't even work. There's a good chance I made a mistake though so don't quote me on that. If the cases don't match, then I imagine it will be really easy to come up with algorithms for all of the cases (select few), so no problem here. For now I do it intuitively with double turns.
Note: During the second and third step, I use double turns exclusively so I can't really destroy the middle layer (just swap a few corners maybe). I don't worry about that during step 2, however, after finishing step 3, the middle layer HAS to be solved. So if I use algorithms for step 3, it should also be solved after step 2.
4. The rest is obvious, just apply PLLs to the top and bottom layer. Likely there will be parity (although it isn't really parity, just the cube isn't solved) and if there is, I solve the two vertical layers (using PLLS) and put all the "bad" edges (in my experience only edges are the problem) into one layer, then do moves like M' U2 M to finish it. I don't think I have to say more. I move edges around with PLLs obv.
So, is this a known thing?
Last edited:
