Welcome to the Speedsolving.com, home of the web's largest puzzle community! You are currently viewing our forum as a guest which gives you limited access to join discussions and access our other features.

I think ATM is a fun metric to play with but STM tells more about actual speedsolving execution. Of course with training and decent fingertricks it's possible to execute something like (U' D) as one move but for something like (R2 M2 R'), which I actually used in my solve, it's not that easy. Technically R2 M2 can be performed as L2 but we want to avoid rotations and I believe the following PLL alg is faster with U turns instead of D turns.

The thing is that I have faith in your post-belt+orientation method anyway already. I just don't want to be too overhasty with my hype. Also my solves are generally not speedsolves, we have to keep that in mind. I will try to create more example solves and even if it turns out the average movecount will be around 41 moves it's pretty great.

EDIT: I'm stupid, R2 M2 can be performed as r2 of course and especially for a Roux speedsolver it will be quite easy to use such things in a real solve. No impact on my point though, I'd be curious to see R2 M2 R' being performed as one move.

Thank you, but as can be seen by reading my edit I already noticed myself.
There's still the R' move in the end, and no matter how smart the execution is, I doubt R2 M2 R' can be performed as fast as a single move in, say, HTM.

I love how the current variant being experimented with here involves reducing to a square-one-esque state, and then solving with a method similar to Roux-and-Screw. It's like we've gone full circle.

Ok, I have an idea for continuation after orientation stage. Here's how the solve would look as a whole:

1) Orient everything (OL5C method), ~15-17 STM
2) Solve 1x2x3 on D (like a sideways FB), ~6-8 STM
3) Solve 1x1x2 on D, ~4-5 STM
4) TTLL, ~7-20 STM
~= 32-50 STM fullstep (if only TTLL move count were lower...)

Total alg count: 24 (OL5C) + 72 (TTLL) = 96

Other idea was to solve a 1x2x2 on U for 3) instead of a 1x1x2 on D, which would give you 4 options and hopefully shorter algorithms for 4), the idea being to reduce to a tripod-like state using pseudo-2x2x3 and -square, with all pieces oriented.
Unfortunately the case count is probably around 120, since you can't take advantage of rotational symmetry like TTLL does.

Edit: On second thought, for 3), you could solve the D layer corners, leaving, yes, HKPLL (god, that name). 7-17 STM according to this, though I tend to think the actual move count stats may be even lower, given how rushed this algset looks (one of the algs, for example... R' U2 R U R U R2 U' R'. The entire set is <R,U,L> as well, which means executing it as <R,U,D> may lead to even lower ATM)

This would push estimated move count to 32-47 STM, though would have more algs to learn.

If anyone wants EPBL algs here are the ones discounting M2 U2 M2 parity. THey are organised by the alg on the bottom.

Spoiler: H

H: M2 U2 M2 y M2 U2 M2 (7)
Ua: (D) M2 U M' U M2 D' M2 U M U M2 (11)
Ub: (D) M2 U2 M2 D' M2 U M' U2 M U' M2 (11)
Z: M2 U' L2 (U D') M2 (U' D) l2 U M2 (9)
avg: 9.5

Spoiler: Ua

H: (U) M2 D M' D M2 U' M2 D M D M2 (11)
Ua: M' (U' D) M2 D2 M' U' M' (U' D) M (U') (9)
Ub: L2 (U' D) R2 U M2 U' L2 (U' D) L2 (9)
Z: M2 D M' U2 M' (U D') M2 U M2 (9)
avg: 9.5

Spoiler: Ub

H: R2 (U' D) M2 D M' U2 M' U' M2 D2 L2 (11)
Ua: L2 (U' D) r2 U M2 U' l2 (U' D) L2 (9)
Ub: M' (U' D) M2 U2 M' D M' (U' D) M (D) (9)
Z: M2 D' M' U2 M' (U' D) M2 U' M2 (9)
avg: 9.5

Spoiler: Z

H: M2 D' R2 (U D') M2 (U' D) L2 U M2 (9)
Ua: M2 U M' D2 M' (U' D) M2 D M2 (9)
Ub: M2 D' M2 (U D') M D2 M U' M2 (9)
Z: M2 D' M2 (U D) M2 U' M2 (7)
avg: (8.5)

total avg (with EPLL): 8.5 ATM (this does not take into account the frequency of the cases)
with AUFs: ~10 ATM

I'm working on it, but can't promise to finish it within a few weeks.
Also, there are two current versions of SSC (1. CP --> LEE, 2. Roux blocks --> PLL --> M-perm).

I'm working on it, but can't promise to finish it within a few weeks.
Also, there are two current versions of SSC (1. CP --> LEE, 2. Roux blocks --> PLL --> M-perm).

If anyone wants EPBL algs here are the ones discounting M2 U2 M2 parity. THey are organised by the alg on the bottom.

Spoiler: H

H: M2 U2 M2 y M2 U2 M2 (7)
Ua: (D) M2 U M' U M2 D' M2 U M U M2 (11)
Ub: (D) M2 U2 M2 D' M2 U M' U2 M U' M2 (11)
Z: M2 U' L2 (U D') M2 (U' D) l2 U M2 (9)
avg: 9.5

Spoiler: Ua

H: (U) M2 D M' D M2 U' M2 D M D M2 (11)
Ua: M' (U' D) M2 D2 M' U' M' (U' D) M (U') (9)
Ub: L2 (U' D) R2 U M2 U' L2 (U' D) L2 (9)
Z: M2 D M' U2 M' (U D') M2 U M2 (9)
avg: 9.5

Spoiler: Ub

H: R2 (U' D) M2 D M' U2 M' U' M2 D2 L2 (11)
Ua: L2 (U' D) r2 U M2 U' l2 (U' D) L2 (9)
Ub: M' (U' D) M2 U2 M' D M' (U' D) M (D) (9)
Z: M2 D' M' U2 M' (U' D) M2 U' M2 (9)
avg: 9.5

Spoiler: Z

H: M2 D' R2 (U D') M2 (U' D) L2 U M2 (9)
Ua: M2 U M' D2 M' (U' D) M2 D M2 (9)
Ub: M2 D' M2 (U D') M D2 M U' M2 (9)
Z: M2 D' M2 (U D) M2 U' M2 (7)
avg: (8.5)

total avg (with EPLL): 8.5 ATM (this does not take into account the frequency of the cases)
with AUFs: ~10 ATM

I've been lurking here because I'm very intrigued about this method! I made an account on here because of this method.

I have a couple of questions for crafto just to clarify some things. When you proposed the idea for roux blocks you said it could be done in about 10-12 moves. Can you explain how you do that? Also, you said that you're supposed to do roux blocks whilst preserving co. Did you mean to preserve co AND belt? It seems more logical, but I'm just making sure.

I've been lurking here because I'm very intrigued about this method! I made an account on here because of this method.

I have a couple of questions for crafto just to clarify some things. When you proposed the idea for roux blocks you said it could be done in about 10-12 moves. Can you explain how you do that? Also, you said that you're supposed to do roux blocks whilst preserving co. Did you mean to preserve co AND belt? It seems more logical, but I'm just making sure.

Hi, welcome on speedsolving.com. I'm not crafto but when the Roux blocks are built, the 4 belt edges are forced to be preserved as they're part of the blocks. However, we can adjust the M slice (and have to, because the next step uses M to bring up some edges).
So, if you don't count the centers as part of the belt the answer is "yes", belt will be preserved, but we don't need to take care of it.