The New Method / Substep / Concept Idea Thread

Aerma said:
The steps of ZBRoux (also known as LLOB) are as follows:
1) Roux-style First Block
2) Roux-style Second Block
3) Solve DF and DB while solving EO
4) ZBLL
Click to expand...
So it's like ZZ but with blocks before EOL.
 
ProStar said:
Don't you need to learn (a modified version for DF and DB) ZBLS also?
Click to expand...
EODFDB (we gotta think of a better name for that...) isn't an algorithm set - it's completely intuitive. I either solve it by solving one of those edges and then inserting the other while solving EO, or do I do it more LSE style, but solving DF and DB instead of UL and UR.
 
Aerma said:
EODFDB (we gotta think of a better name for that...) isn't an algorithm set - it's completely intuitive. I either solve it by solving one of those edges and then inserting the other while solving EO, or do I do it more LSE style, but solving DF and DB instead of UL and UR.
Click to expand...

EOL2E? (I spent an entire 15.76 seconds of my life thinking of that name, you better not reject it lol)
 
ProStar said:
Ok, I got another bad idea:

F2L Pairs - Solve the corner-edge f2l pairs, but not the cross. The middle layer should now be finished, as well as the bottom corners.
CLL - Solve the corners on the last layer with one alg
Orient all edges - Orient the remaining 8 edges with 1 alg
Permute all edges - Permute the 8 edges with 1 alg

Probably not very practical, and I'd be willing to bet that OAE and PAE would take a lot of algs.
Click to expand...

Ok, idea for modified PCMS.

Pairs - Normal PCMS
CMSLL - Normal PCMS
EO - Orient L8E, this can be done intuitively with the help of a few algs; similar to Roux EO
B3E - Permute three bottom edges by using the M and S slices. Intuative
L5E - Solve the last bottom edge while solving the four top edges. This is mostly intuitive


Example solve time! (I'm bad at this :P)

z2 // Inspection

L U2 F' U F // 1st Pair

U' F U2 F' L' U2 L // 2nd Pair

U R' U R U' l U l' // 3rd Pair

y U M2 U2 R U' R' M2 // 4th Pair

U2 r' D' r U r' D r U' r U r' // CMSLL

U D' M D' M' // EO

z2 y D' U2 M U2 M' U' M' U2 M // B3E

y' M' U2 M U M' U2 M // L5E


While messing around with this method(I like doing timed solves with other methods), I got a 15 second solve, which is amazing for me with CFOP. Of course, it was an extremely lucky scramble, comparable to an XCross+Free pair. I'm not expecting this method to be any good, but it's fun nevertheless.
 
ProStar said:
Ok, idea for modified PCMS.

Pairs - Normal PCMS
CMSLL - Normal PCMS
EO - Orient L8E, this can be done intuitively with the help of a few algs; similar to Roux EO
B3E - Permute three bottom edges by using the M and S slices. Intuative
L5E - Solve the last bottom edge while solving the four top edges. This is mostly intuitive


Example solve time! (I'm bad at this :p)

z2 // Inspection

L U2 F' U F // 1st Pair

U' F U2 F' L' U2 L // 2nd Pair

U R' U R U' l U l' // 3rd Pair

y U M2 U2 R U' R' M2 // 4th Pair

U2 r' D' r U r' D r U' r U r' // CMSLL

U D' M D' M' // EO

z2 y D' U2 M U2 M' U' M' U2 M // B3E

y' M' U2 M U M' U2 M // L5E


While messing around with this method(I like doing timed solves with other methods), I got a 15 second solve, which is amazing for me with CFOP. Of course, it was an extremely lucky scramble, comparable to an XCross+Free pair. I'm not expecting this method to be any good, but it's fun nevertheless.
Click to expand...
L6E is probably better for this, especially considering that EO is already done in this. Also, inserting 2 edges is easier 3.
 
I was messing around with 3x3 yesterday and thought up a method so bad I had to try it: a mix of Petrus and Roux. The more I examined it though, the less hopeless it seemed.

Yes, I know what you're thinking: "It's another guy with some 'new method' which is either just a simple variation of something else or it's been done before." Well, (as you'll see soon) it's not the first. And, to the extent of my knowledge, it isn't the second either. I've scoured the web but have been unable to find this or a similar method. So, before I ramble on too long, let's get to the good stuff.

Method:
  1. Solve two 1x2x2 blocks on right lower front and left lower front (i.e., by putting the M slice centers in their places and solving the DF edge you would solve a 2x2x3 block.)
  2. Put one U or D edge (an edge with yellow or white on it) in the DF slot. It doesn't have to be the right edge, just make sure it is oriented correctly (white/yellow facing down).
  3. Do a y move. Now, solve all edge orientation Petrus-style. (I won't cover that here — if you don't know how to do that, check this tutorial).
  4. Solve F2L minus the DR and DL edges using only <R, U, S2> turns. S2 is the important thing: provided you only use the S slice in increments of 2 you will keep edge orientation. (The S2 also makes F2L much easier than in Petrus.)
  5. CMLL (look up any Roux tutorial)
  6. L4E (same as above)

Why this might be good
I was thinking about what benefits this method might have, and although I couldn't come up with much I was thinking that since Roux and Petrus are both blockbuilding-based, low alg count methods that they might lend well to each other. Maybe I'm wrong, but just a thought.

So... now it's your turn. Tell me what you think! Questions, comments, suggestions, anything. I don't care whether you tell me that this is absolute garbage or the next CFOP, any input is welcome!

(P.S. I really do apologize if I have stolen someone's method. I promise I have come to this conclusion independently and was not aware in any way of your method.)

Whoops, was a bit groggy when I wrote this. There should be a "Solve UL/UR Edges" step after CMLL.
 
Last edited:
fuzzballcat said:
I was messing around with 3x3 yesterday and thought up a method so bad I had to try it: a mix of Petrus and Roux. The more I examined it though, the less hopeless it seemed.

Yes, I know what you're thinking: "It's another guy with some 'new method' which is either just a simple variation of something else or it's been done before." Well, (as you'll see soon) it's not the first. And, to the extent of my knowledge, it isn't the second either. I've scoured the web but have been unable to find this or a similar method. So, before I ramble on too long, let's get to the good stuff.

Method:
  1. Solve two 1x2x2 blocks on right lower front and left lower front (i.e., by putting the M slice centers in their places and solving the DF edge you would solve a 2x2x3 block.)
  2. Put one U or D edge (an edge with yellow or white on it) in the DF slot. It doesn't have to be the right edge, just make sure it is oriented correctly (white/yellow facing down).
  3. Do a y move. Now, solve all edge orientation Petrus-style. (I won't cover that here — if you don't know how to do that, check this tutorial).
  4. Solve F2L minus the DR and DL edges using only <R, U, S2> turns. S2 is the important thing: provided you only use the S slice in increments of 2 you will keep edge orientation. (The S2 also makes F2L much easier than in Petrus.)
  5. CMLL (look up any Roux tutorial)
  6. L4E (same as above)

Why this might be good
I was thinking about what benefits this method might have, and although I couldn't come up with much I was thinking that since Roux and Petrus are both blockbuilding-based, low alg count methods that they might lend well to each other. Maybe I'm wrong, but just a thought.

So... now it's your turn. Tell me what you think! Questions, comments, suggestions, anything. I don't care whether you tell me that this is absolute garbage or the next CFOP, any input is welcome!

(P.S. I really do apologize if I have stolen someone's method. I promise I have come to this conclusion independently and was not aware in any way of your method.)
Click to expand...

Really like that method. Well thought out, especially for your first post!

I've tried to create an example solve that can be found here, so this is what I've noticed:
  1. Not sure if it's fully plannable in inspection, but first 1x2x2 block and DR are definitely possible. I'd put them in the back instead of the front for better lookahead.
  2. This step can be removed in my opinion. Only advantage I could see is that it prevents 8 flips, but you solve EO immediately after, so just EO would actually be more efficient.
  3. Good step, though it's much more ergonomic to do the rotation afterwards (most cases are <RULF>, with a y' it'd be more like <RUBF>).
  4. While S2 does make the last two pairs more efficient and easier than Petrus, it isn't really fast to execute. So I'd recommend to only really use S2 when it's way shorter to use it than only <RU>.
  5. Since you have EO already solved, you'd have to do COLL (but ignoring the permutation of the S-slice) here.
  6. Because of EO, you could do LSE in one algorithmic step, L6EP. For this you should probably rotate, since M is a lot faster than S.
For the method as a whole, I believe that Roux is still better than that since algs (preserving EO takes more moves) and ergonomics (Roux has all moves, <RrMU>, <RUF> (or whatever CMLL algs you use) and <MU> while your method has all moves, <RULF>, <RUS2>, <RUF> and <MU>).

This still method does have some potential though, so keep on improving it!
 
RedstoneTim said:
Really like that method. Well thought out, especially for your first post!

I've tried to create an example solve that can be found here, so this is what I've noticed:
  1. Not sure if it's fully plannable in inspection, but first 1x2x2 block and DR are definitely possible. I'd put them in the back instead of the front for better lookahead.
  2. This step can be removed in my opinion. Only advantage I could see is that it prevents 8 flips, but you solve EO immediately after, so just EO would actually be more efficient.
  3. Good step, though it's much more ergonomic to do the rotation afterwards (most cases are <RULF>, with a y' it'd be more like <RUBF>).
  4. While S2 does make the last two pairs more efficient and easier than Petrus, it isn't really fast to execute. So I'd recommend to only really use S2 when it's way shorter to use it than only <RU>.
  5. Since you have EO already solved, you'd have to do COLL (but ignoring the permutation of the S-slice) here.
  6. Because of EO, you could do LSE in one algorithmic step, L6EP. For this you should probably rotate, since M is a lot faster than S.
For the method as a whole, I believe that Roux is still better than that since algs (preserving EO takes more moves) and ergonomics (Roux has all moves, <RrMU>, <RUF> (or whatever CMLL algs you use) and <MU> while your method has all moves, <RULF>, <RUS2>, <RUF> and <MU>).

This still method does have some potential though, so keep on improving it!
Click to expand...
Thanks! Your suggestions are very helpful. For step two, how would you do EO if the edge in the place mentioned is flipped (is there an EO algorithm I don't know about?)

Thanks again!
 
So I came up with a pretty bad skewb method, but it may become mediocre with some more development. The steps are as follows:
1. Solve a face, with regard to corner permutation.
2. With sledge hammers and hedgeslammers, solve the F and B centers or the L and R centers in each others spots, so if you had blue for your solved face on bottom, then you could solve the white center in the yellow centers spot, and the white in the yellows spot.
3. Do a sledgehammer to finish the centers, there will always be one solved center after step 2, and if it is on the sides, You can put in the back and one sledge will finish the centers. Heres one of the main flaws with this method, if the solved center is the top center, then you get a bad case for edges that needs three sledges and two y2s to solve, so im wondering how to prevent this.
4. Solve the last four corners.
 
Last edited:
You must log in or register to reply here.

Similar threads

D
The physics of Rubik's Cube to Substep Method Platform
2
Replies
33
Views
1,787
Christopher Mowla
Christopher Mowla
D
Incomplete Substep Method LL Megaminx
Replies
2
Views
299
Deleted member 27249
D
Dane man
Calculating Method/Substep Efficiency
2
Replies
27
Views
8,641
brododragon
brododragon
M
What method is this?
2
Replies
24
Views
6,121
~Adam~
~Adam~
AK51
2.51 world record used 33 moves, it shall be 28 moves using AK-cubie method
Replies
0
Views
224
AK51
AK51
Back
Top