The New Method / Substep / Concept Idea Thread

TorbinRoux said:
Not even inefficient imo... I think this is really cool. How many algs for L5E? Also WVCP?
Click to expand...

Ha, he was referring to the "method" posted before. In the cons is the alg count. 142 for WVCP(?) and 120 for EOL5E. A different thread was merged around my post. It was really weird. Also look at the example solve
EDIT: I actually just finished my 17.68 AO12 with this.
 
Mollerz said:
FMC scoring using OBTM. M, E and S moves do not exist in this metric. If you use an M move in a solution, it should be a DNF.
OBTM is similar to HTM, but HTM includes M, E and S, which all count as 2 moves each.
STM is the same as HTM, but M, E and S are all counted as 1 move.

This is not a good FMC method.
Click to expand...

Thanks for the clarification, I always get confused with all the different metrics.
 
SFCuber said:
The first one I posted was mostly supposed to be intuitive so the solver can make it efficient...
Click to expand...

But in FMC, M-slices suck.
here is a very efficient roux solve, which has primarily slice moves.
F2 D B2 F2 L2 D' F2 D2 B2 U2 R' D F2 L2 D' U' B R' B2 U2
U2 D R L2 D' F//FB
R' U2 Rw U2 R U R Rw U Rw U' R//SB
Rw U' L' D L U L' D' x'//CMLL
Rw2 R2 U' R Rw' U R' Rw U' R' Rw U' R2 Rw2//LSE
40 HTM. While not awful, you can get much better solves even with pure petrus (No fancy LS, COLL, etc.)
 
2180161 said:
But in FMC, M-slices suck.
here is a very efficient roux solve, which has primarily slice moves.
F2 D B2 F2 L2 D' F2 D2 B2 U2 R' D F2 L2 D' U' B R' B2 U2
U2 D R L2 D' F//FB
R' U2 Rw U2 R U R Rw U Rw U' R//SB
Rw U' L' D L U L' D' x'//CMLL
Rw2 R2 U' R Rw' U R' Rw U' R' Rw U' R2 Rw2//LSE
40 HTM. While not awful, you can get much better solves even with pure petrus (No fancy LS, COLL, etc.)
Click to expand...

I see your point but you can get lower than that by not doing the LSE using Ms and doing L5E type stuff or edge cycle insertions.
 
Roux-inspired Blockbuilding Method for 5x5

Hi all, I just came up with a fun, maybe fast concept for 5x5.

Step 1: Base
Description: Solve a non-cross colour center + the corresponding cross edge

Step 2: First Block
Description: Solve a Roux-style FB by pairing up tredges with corners and inserting them by the craddle to form a solved block

Step 3: Second Block
Description: Solve a Roux-style SB by solving a center opposite to one of those we solved in the first step (I would solve red) followed by pairing up tredges and corners using slice moves and inserting them.

Step 4: Last 4 Centers
Description: Solve the remaining centers using slice moves and U moves.

Step 5: First 4 Layers
Description: Pair up the remaining cross tredges and insert them using slice and U moves.

Step 6: Reduction:
Description: Pair up the remaining tredges systematically. This is done using 2 easy algorithms. (one could learn 10-50? algorithms to be able to pair the tredges in one look)

Step 7: Last Layer
Description: Solve the last layer as a 3x3.

Pros:
- Direct solving is efficient
- Easy 3x3 stage
- Not too many substeps
- Good lookahead
- Ergonomic (mostly U and slice moves)

Cons:
- Direct solving restricts the user to particular pieces
- Last layer method is a bit slow unless you learn a lot of algs

There we go! Tell me what you think. I suck at 5x5 so I have nothing to lose by practising with this method. If I get decently fast, I'll update this post.
 
Last edited:
crafto22 said:
Hi all, I just came up with a fun, maybe fast concept for 5x5.

Step 1: Craddle
Description: Solve two adjacent centers (I do white and orange)

Step 2: First Block
Description: Solve a Roux-style FB by pairing up tredges with corners and inserting them by the craddle to form a solved block

Step 3: Second Block
Description: Solve a Roux-style SB by solving a center opposite to one of those we solved in the first step (I would solve red) followed by pairing up tredges and corners using slice moves and inserting them.

Step 4: Last 4 Centers
Description: Solve the remaining centers using slice moves and U moves.

Step 5: First 4 Layers
Description: Pair up the remaining cross tredges and insert them using slice and U moves.

Step 6: Reduction:
Description: Pair up the remaining tredges systematically. This is done using 2 easy algorithms. (one could learn 10-50? algorithms to be able to pair the tredges in one look)

Step 7: Last Layer
Description: Solve the last layer as a 3x3.

Pros:
- Direct solving is efficient
- Easy 3x3 stage
- Not too many substeps
- Good lookahead
- Ergonomic (mostly U and slice moves)

Cons:
- Direct solving restricts the user to particular pieces
- Last layer method is a bit slow unless you learn a lot of algs

There we go! Tell me what you think. I suck at 5x5 so I have nothing to lose by practising with this method. If I get decently fast, I'll update this post.
Click to expand...

there was a 4x4 method almost exactly like this..ross used it but I forgot what it was called
 
3x3 method: F3L
Step one - Solve the cube
Alg count - 43,252,003,274,489,856,000
Execution time - ~4-5 seconds, 3 if you drill algs in practice





my 100th post :D
 
DELToS said:
3x3 method: F3L
Step one - Solve the cube
Alg count - 43,252,003,274,489,856,000
Execution time - ~4-5 seconds, 3 if you drill algs in practice



my 100th post :D
Click to expand...

still slower than cfop, solutions aren't fingertrick friendly, easy to mess up on, and recognition, lots of cases that are extremely similar to one another
 
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
1,789
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
8,642
brododragon
brododragon
M
What method is this?
2
Replies
24
Views
6,123
~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