Yeah, I counted them like this. You have 2dfr positions, in Dfr or above it. So you will have 4 eo cases if it is in place and 8 if it's not, because you are cancelling simmetries.
Then you have 6 CP of the 4 resting corners, this makes you have 4•6+8•6 EOCP cases.
For JBLL you have 12 EP cases, and 14 co cases, you can check the ribbon method algcount. It has 116 tols cases instead of 57 because of having one corner more to orient. This means that there are 14 ocll cases. So it's just 14•12 that makes 168 algs.
Oh, okay, I understand. EOCP also inserts DFR into it’s slot…
Yeah that makes sense, that also proves your 168 figure and it might explain EOCP? Haven’t checked. The reason I was mistaken is because I believed the DFR corner can be in the U layer too (I thought it was cool), so that made the alg count way higher.
Nice thinking, love the creativity!
So, yesterday I was making some solves on the 2x2, and this thought passed across my mind : making an entire layer kinda sucks.
So I theorised this method.
1) make a layer missing the Dfr corner (it doesn't need to be permuted correctly)
This takes at max 3 moves if you're color neutral.
2)orient all the corners. 14 algs
3) xpbl, which means permute everything, but two corners between d and u layer could be swapped, and there are 25 algorithms for that I think
The second idea was to
1) make the L shape
2) orient all the corners while separating u layer's from d layer's
3) permute everything (just pbl)
I am not sure about the xpbl's algcount though
Any suggestions?
I'm aware that there are some pretty new people on the forums who haven't been around for as long as some others have. I would recommend looking at Cubing History, a great resource.
A common strategy, with which you can create so many "methods", is by bashing steps together. Can you instead create a method by looking at shapes etc.? If you want some in depth things, there's this by Shadowslice and also this, this by me and this by Kirjava (which I found by trying to find a different thing, by Kenneth, but it's great and serves my purpose just as well). As you can see, people have thought about this a lot, which partly makes things more tricky for young, aspiring method developers, but it also means you can optimise further, come up with different ideas that people may not have seen. Go well and happy developing!
I'm aware that there are some pretty new people on the forums who haven't been around for as long as some others have. I would recommend looking at Cubing History, a great resource.
A common strategy, with which you can create so many "methods", is by bashing steps together. Can you instead create a method by looking at shapes etc.? If you want some in depth things, there's this by Shadowslice and also this, this by me and this by Kirjava (which I found by trying to find a different thing, by Kenneth, but it's great and serves my purpose just as well). As you can see, people have thought about this a lot, which partly makes things more tricky for young, aspiring method developers, but it also means you can optimise further, come up with different ideas that people may not have seen. Go well and happy developing!
I'm aware that there are some pretty new people on the forums who haven't been around for as long as some others have. I would recommend looking at Cubing History, a great resource.
A common strategy, with which you can create so many "methods", is by bashing steps together. Can you instead create a method by looking at shapes etc.? If you want some in depth things, there's this by Shadowslice and also this, this by me and this by Kirjava (which I found by trying to find a different thing, by Kenneth, but it's great and serves my purpose just as well). As you can see, people have thought about this a lot, which partly makes things more tricky for young, aspiring method developers, but it also means you can optimise further, come up with different ideas that people may not have seen. Go well and happy developing!
I'm aware that there are some pretty new people on the forums who haven't been around for as long as some others have. I would recommend looking at Cubing History, a great resource.
A common strategy, with which you can create so many "methods", is by bashing steps together. Can you instead create a method by looking at shapes etc.? If you want some in depth things, there's this by Shadowslice and also this, this by me and this by Kirjava (which I found by trying to find a different thing, by Kenneth, but it's great and serves my purpose just as well). As you can see, people have thought about this a lot, which partly makes things more tricky for young, aspiring method developers, but it also means you can optimise further, come up with different ideas that people may not have seen. Go well and happy developing!
Shape is broad. BCE is a way of generating shapes. The shape is what leads to the key step in what Shadowslice talks about. Ribbon could be counted as a shape, as is cross, or F2B in Roux. How can you get the cube into a state where the solution is nice from then on?
Shape is broad. BCE is a way of generating shapes. The shape is what leads to the key step in what Shadowslice talks about. Ribbon could be counted as a shape, as is cross, or F2B in Roux. How can you get the cube into a state where the solution is nice from then on?
Like, I don't know if already exists, to solve a 1x2x3 block in the left d layer and completing the whole belt, so then you insert Dbr while doing eo, then solving the rest of the dr block while influencing LL?
Like, I don't know if already exists, to solve a 1x2x3 block in the left d layer and completing the whole belt, so then you insert Dbr while doing eo, then solving the rest of the dr block while influencing LL?
It's ok to start with basic shapes, but it's more that it's easy to grind out lots of ideas, which is a good thing to get you thinking, but then you need to progress beyond that and these are the tools that will help.
It's ok to start with basic shapes, but it's more that it's easy to grind out lots of ideas, which is a good thing to get you thinking, but then you need to progress beyond that and these are the tools that will help.
I have two ideas, witch is the most amount of ideas I've ever had, one is a last layer method for 3x3 and on the last slot you put it in, in a way were you only get dot or cross cases, and ollcp for all dot cases and cross cases. My second idea is for 4x4 were you solve all the centers, the cross, f3l (first 3 layers) oll, and pll.
I think I remember there being an algset for this, but it wouldn’t be as easy as oll and pll because you haven’t solved edges. Lots more algs I think and also that makes me believe this might be already done, good idea tho!
Here is a new 3x3 Columns-First / PCMS-to-Roux hybrid method I developed called The VD Method. It skips the cross for unrestricted D/M F2L pairing, solves CLL to finish all 8 columns, and uses symmetrical <M, U> DL/DR insertions with a single y rotation before a Roux-style <M, U> top-edge finish. The VD Method (Columns-First / Rotational PCMS Hybrid)Creator: VD Flow: Crossless F2L → CLL → Opposite Edges (DL/DR) → y Rotation & Repeat → <M, U> ELL Finish Alg Count: ~16 (2-Look CLL + intuitive edges) to 42 (Full CLL) Estimated Move Count: ~55–65 STM Steps:
Crossless F2L (4 Pairs): Pair and insert all four corner-edge F2L pairs around the E-slice centers using free D and M moves.
CLL: Solve the top four corners while ignoring top and bottom edges, completing all 8 vertical columns.
First Opposite Edges (DL & DR): Use <M, U> triggers (like M' U2 M or M2 U2 M2) to insert the DL and DR cross edges.
Rotate & Repeat: Do a y rotation so the last two bottom edges sit at DL and DR, then insert them with <M, U> moves to finish the bottom two layers.
Top-Layer Edges (<M, U> ELL): Orient the final four top edges with M'/U triggers and permute them using M2/U moves.
Scramble (White Top, Green Front):R2 U' F2 D B2 L2 U' R2 U2 B2 F' R D' B L' B2 U' R' F U2(Solve starts with Yellow on top, White on bottom)Step 1: Crossless F2L
1st Pair (Orange-Blue): z2 y' R U' R' U' L' U L
2nd Pair (Red-Blue): U' R U R' U2 R U' R'
3rd Pair (Red-Green): y' U' R U2 R' U R U' R'
4th Pair (Orange-Green): y' U2 R U' R' U R U' R'
Step 2: CLL (2-Look)
Orientation (Sune): (U) R U R' U R U2 R'
Permutation (J-Perm): (U) R U R' F' R U R' U' R' F R2 U' R'
Step 3: First Opposite Edges (DL & DR)
Insert DR: U' M2 U2 M2
Insert DL: U M' U2 M U2 M' U2 M
Step 4: Rotate & Repeat
Rotate & Insert DR: y U2 M2 U2 M2
Insert DL: U' M' U2 M
Step 5: Top-Layer Edges (<M, U> ELL)
Orient Edges: U M' U M' U2 M' U M'
Permute Edges (U-Perm): U M2 U M U2 M' U M2 U2
Pros: Very low algorithm count, unrestricted D/M usage in F2L, and ergonomic <M, U> endgame. Cons: One mid-solve y rotation and D-layer blind-spot tracking during Steps 3–4. Any feedback on move efficiency or better ways to handle Steps 3–4 is appreciated!
Here is a new 3x3 Columns-First / PCMS-to-Roux hybrid method I developed called The VD Method. It skips the cross for unrestricted D/M F2L pairing, solves CLL to finish all 8 columns, and uses symmetrical <M, U> DL/DR insertions with a single y rotation before a Roux-style <M, U> top-edge finish. The VD Method (Columns-First / Rotational PCMS Hybrid)Creator: VD Flow: Crossless F2L → CLL → Opposite Edges (DL/DR) → y Rotation & Repeat → <M, U> ELL Finish Alg Count: ~16 (2-Look CLL + intuitive edges) to 42 (Full CLL) Estimated Move Count: ~55–65 STM Steps:
Crossless F2L (4 Pairs): Pair and insert all four corner-edge F2L pairs around the E-slice centers using free D and M moves.
CLL: Solve the top four corners while ignoring top and bottom edges, completing all 8 vertical columns.
First Opposite Edges (DL & DR): Use <M, U> triggers (like M' U2 M or M2 U2 M2) to insert the DL and DR cross edges.
Rotate & Repeat: Do a y rotation so the last two bottom edges sit at DL and DR, then insert them with <M, U> moves to finish the bottom two layers.
Top-Layer Edges (<M, U> ELL): Orient the final four top edges with M'/U triggers and permute them using M2/U moves.
Scramble (White Top, Green Front):R2 U' F2 D B2 L2 U' R2 U2 B2 F' R D' B L' B2 U' R' F U2(Solve starts with Yellow on top, White on bottom)Step 1: Crossless F2L
1st Pair (Orange-Blue): z2 y' R U' R' U' L' U L
2nd Pair (Red-Blue): U' R U R' U2 R U' R'
3rd Pair (Red-Green): y' U' R U2 R' U R U' R'
4th Pair (Orange-Green): y' U2 R U' R' U R U' R'
Step 2: CLL (2-Look)
Orientation (Sune): (U) R U R' U R U2 R'
Permutation (J-Perm): (U) R U R' F' R U R' U' R' F R2 U' R'
Step 3: First Opposite Edges (DL & DR)
Insert DR: U' M2 U2 M2
Insert DL: U M' U2 M U2 M' U2 M
Step 4: Rotate & Repeat
Rotate & Insert DR: y U2 M2 U2 M2
Insert DL: U' M' U2 M
Step 5: Top-Layer Edges (<M, U> ELL)
Orient Edges: U M' U M' U2 M' U M'
Permute Edges (U-Perm): U M2 U M U2 M' U M2 U2
Pros: Very low algorithm count, unrestricted D/M usage in F2L, and ergonomic <M, U> endgame. Cons: One mid-solve y rotation and D-layer blind-spot tracking during Steps 3–4. Any feedback on move efficiency or better ways to handle Steps 3–4 is appreciated!
Core Philosophy: Solve all eight corners alongside the middle equator first, reducing the second half of the solve into a fast <M, E, U2> domino-style puzzle.<br>
Primary Move Set:<R, U, M, E, F2><br>
<br><br>
<br><br>Step 1: The Corner Pillars (10–12 moves)<br>Instead of starting with edges, build four vertical 1x1x2 Corner-Edge Pillars around the middle layer (E slice).<br>
<br>
Goal: Pair all four bottom (D) corners with their matching middle-layer (E) equator edges and place them in their correct relative positions.<br>
The Trick: Leave the D-layer edges completely unsolved. With top and bottom edges free, you can use U, D, R2, and L2 moves to drop pillars in 2–3 moves each.<br>
<br><br>Step 2: Top Corners + Belt Lock (10–11 moves)<br>Solve the remaining four top (U) corners without disturbing the pillars below.<br>
<br>
Goal: Orient and permute all four U corners simultaneously.<br>
How it works: Use CLL algorithms (42 short algorithms, ~9 moves average). Because U and D edges are ignored, these algorithms are much shorter than standard OLL/PLL.<br>
Cube State: All 8 corners and all 4 equator edges are solved. Only 8 edges remain.<br>
<br><br>Step 3: Edge Separation — Domino Gate (8–9 moves)<br>Separate the remaining 8 unsolved edges by layer and orient them simultaneously.<br>
<br>
Goal: Place all 4 U-color edges in the top layer and all 4 D-color edges in the bottom layer, with all stickers oriented (facing up or down).<br>
How it works: Use simple <M, U2> triggers (such as M' U2 M) to swap misoriented edges between layers in pairs or sets of four.<br>
<br><br>Step 4: Dual-Layer Permutation — DLP (13–15 moves)<br>With the cube reduced to a Domino state, finish the solve using purely <R2, L2, U, D, M2> moves.<br>
<br>
Goal: Permute both the top 4 edges and bottom 4 edges at the same time.<br>
How it works: Execute 1 algorithm from a set of 22 DLP algorithms operating on both layers simultaneously.<br>
The BEE method for speedsolving (highly experimental)
step 1: make a 2x2x3 block on the DR position.
step 2a: make a 1x2x2 block on the DBL position while making it so that the UL edge and the UBL corner are touching.
step 2b: solve the UL edge and the UBL corner to form a 1x2x3 block.
step 3: solve the remaining 4 corners (DLF, UFL, UFR, UBR) with one algorithm, this step is called L4C.
step 4: solve the remaining 4 edges (FL, FU, RU, BU) with one algorithm, this step is called L4E.
As I said, this method is highly experimental, so the algorithms aren't fully developed yet