Speed Heise-2 Rubik's Cube Method [SH-2]

[CriticalCubing]

Speed Heise-2 is a different approach to Heise which is optimized for SpeedSolving and highly intuitive. SH-2 starts with a First Block, transition to 2x2x2 on D layer (either on DFR or DBR). Solves edges + Corners and finishes the last corners with commutators (also can be done intuitively). Due to the intuitive nature, lookahead is pretty simple and straight forward. Inspection is also used well as you plan out first block + part of 2x2x2 in inspection.
Speed Heise-2 Document: SH-2

Pure Heise vs Speed Heise vs Speed Heise-2
R2 D2 R2 D2 B2 D2 B' R2 B' F2 D2 L B R D B' F2 D' L D' F'

Pure Heise:
R' U2 F // 2 squares (3)
y2 F R2 U2 R2 U R // 3rd square (6)
y' U R' U R y U' L F' L' U // prepare last square (9)
R' F R2 // fix f2l, orient edges (3)
y U' R U' R' U' R U' // 5 edges + 2 corners (7)
y2 F' L' U R2 U' L U R2 U' F // commutator (10)
// = 38 moves pure Heise

Speed-Heise:
x2 D2 F2 U B' D // 2x2x2 (5)
y' R2 U' D' F D // 3x2x2 (5)
L' U L2' F' L' R U' R' // eo (8)
y' R2 U R' U' R // f2l-1 (5)
U2 R U' R' // set up (4)
R2 U' L' U R2 U' L U2 R U2 R' // speed-heise (11)
d' y R2 D R' U2 R D' R' U2 R' // l3c (10)
// = 48 moves Speed-Heise

Speed Heise-2:
x2 y R2 F U R u' R2 u' //FB (7)
U' M' r U' r U2 R2 U' M // 2x2x2 DBR (9)
U r U2 r' F' U F U' R U R' // Setup (11)
F2 U2 F2 U R' F2 U L' U' L F2 R // Finish (12)
// = 39 moves SH-2

Look at the document for steps, links, example solves and more info

 

[mDiPalma]

So it's just regular Heise except you do Roux in the beginning?

Sure, your initial steps may be slightly more ergonomic and rotationless than a Petrus-style block (which is what most people do), and they may be more suited for a speedsolve than the Heise-square approach, but you haven't addressed the primary issue with Heise Step-3 at all.

I think the primary issue with Heise is that 5e+2c is too hard of a step to do intuitively in a speedsolve - that's why Speed-Heise/Cardan Reduction/CR† seek to systemize the step with algorithms. Your "speed" method doesn't address that.

How long did it take you to find those Step 3 solutions?

 

[Pyjam]

I Haven't read the document, sorry.
Does "setup" = edge orientation + square on top ?
What set does the last alg come from?

There's a lot of easy blockbuilding with this sample.
Just for fun and for the sake of comparison, I've searched an "optimized" (pseudo) ZZ solution. I doesn't mean a lot, however.
U R L' U L' U' L' F // (8) Edge orientation (+some setups)
L' U L U' L' U R D' U' R2 // (10/18) F2L -1 pair
U' R U R' U' R U2 R' // (8/26) 4th pair
U R U2 R' U' R U' R' // (8/34) AntiSune - Solved

 

[CriticalCubing]

So it's just regular Heise except you do Roux in the beginning?

Sure, your initial steps may be slightly more ergonomic and rotationless than a Petrus-style block (which is what most people do), and they may be more suited for a speedsolve than the Heise-square approach, but you haven't addressed the primary issue with Heise Step-3 at all.

I think the primary issue with Heise is that 5e+2c is too hard of a step to do intuitively in a speedsolve - that's why Speed-Heise/Cardan Reduction/CR† seek to systemize the step with algorithms. Your "speed" method doesn't address that.

How long did it take you to find those Step 3 solutions?

Interesting question.
In the 3rd step, I solve the edges intuitively while influencing the corners. I do it using basic things like instead of inserting with R U' R', I'd do R U2 R' which solves corners differently. Instead of doing R' F R F', I'd do U' F' U F. So its intuitive things like this that I do for this 3rd step. For the above solution, I needed to insert DF edge and these are what I use (M' U2 M or r U2 r'). I went for r U2 r' as there was a good edge in the FR position which I preserved. F' U F does EO, and the final insert solves step 3. For me, this is pretty fast, however this is also kind of subconscious so I don't have a system for it. Personally, I don't think that intuitive is hard to do in a speedsolve as there are recurring patterns that you can influence. Take a look at Roux blockbuilding, it is intuitive yet fast. Initially it feels bad/slow but just needs some getting used to.
I also didn't want to make this an algorithm based method too much. We are already doing algorithm's for the last step anyway. For alg based approach, there are better approaches that we can take. This method preserves Heise's intuitiveness but the overall flow is better than a normal Heise solve.

 

[mDiPalma]

For me, this is pretty fast, however this is also kind of subconscious so I don't have a system for it. Personally, I don't think that intuitive is hard to do in a speedsolve as there are recurring patterns that you can influence.

I'd love to see a video of you executing a few of these SH-2 solves. I can only get ~20 seconds with pure Heise.

 

[CriticalCubing]

I'd love to see a video of you executing a few of these SH-2 solves. I can only get ~20 seconds with pure Heise.

Haven't learned commutators, and won't be learning them as I won't be using them. I can do reconstructions/walkthrough if you want.

 

[I_<3_SCS]

Haven't learned commutators, and won't be learning them as I won't be using them. I can do reconstructions/walkthrough if you want.

wydm you don't know commutators? Aren't they INTUITIVE? )))))

 

[CriticalCubing]

wydm you don't know commutators? Aren't they INTUITIVE? )))))

Nope. I'm RETARDED!!! )))))

 

[CriticalCubing]

I think the primary issue with Heise is that 5e+2c is too hard of a step to do intuitively in a speedsolve - that's why Speed-Heise/Cardan Reduction/CR† seek to systemize the step with algorithms. Your "speed" method doesn't address that.

I have to apologize. I'm terribly sorry as I completely misunderstood you and the 5e2c concept. The way I'm currently doing is Orienting+Permutation of edges, and Orientation of corners. The last step finishes and orients the remaining 3 corners while permuting all the corners simultaneously. This was the speed based approach which also keeps the solve efficient, however alg count is quite a lot with this (which I hadn't previously noticed as I didn't think anyone would use this).
However now that you mention it, 5e+2c (orientation + permutation) is a difficult step to do intuitively. Edges is easy, but permuting both edge+corners is difficult. (I only permuted the edges).
Now, here are some immediate ideas that come to mind.
a) Orient and permute the edges only. Solve the remaining corners as 2 3-cycle.
b) Orient edges and corners both and cycle them around intuitively (using simple coms)
c) Orient and permute edges only. Solve last 5 corners all at the same time (alg count will be high with this)
d) Orient and use your "speed-heise" algset to get L3C and solve it. (this also suits the speed heise-2 name haha as its using the speed heise alg set, using different approach for the previous steps xD)
Anyway, those are the ideas that come to mind. Tell me what you think I'll edit the doc accordingly.
I'm sorry for all the misunderstand on my part, and all the confusion that it created. My apologies.

Edit: Formatting