Heise Method Discussion thread

[Johannes91]

You need the top layer corners to be oriented correctly

No you don't.

I have to admit this is, in my opinion, one of the hardest parts in the heise method, because there is very little info on his site.

I don't think there's any need for info on this, it's so straight-forward. If you can't learn on your own without tutorials, Heise is probably not the right method for you.

The hardest part for me is definitely leaving just 3 corners.

A few examples:
1. R U2 R' U L' R U R' U' L U R U2 R' U L' R U R' U' L U L' R U R' U' L U L' R U R' U' L U L' R U R' U' L U L' R U R' U' L
U' R U R'
2. R U R' U R U2 R' U R U' R' U R U' R' U L' R U R' U' L U R U' R' U R U2 R' U R U2 R'
U R U R'
3. L' R U R' U' L U L' R U R' U' L U R U R' U R U2 R' U R U' R' U R U R' U R U' R' U L' R U R' U' L
R U R' U' R U2 R'
4. R U2 R' U R U R' U L' R U R' U' L U R U2 R' U L' R U R' U' L U L' R U R' U' L U R U2 R' U L' R U R' U' L
U' R U R' U2 R U' R'
5. R U R' U L' R U R' U' L U R U' R' U R U' R' U R U2 R' U R U R' U R U R' U R U' R'
R U R' U2 R U' R'

 

[Am1n-]

let me revise my previous post:
[How I learnt it] * previous post [/How I learnt it]

I'm still working on it.
Leaving just 3 corners can always be achieved by using pair 3-cycles, with sometimes an R conjugate.

You need the top layer corners to be oriented correctly

No you don't.

I meant after the sequence; otherwise you cant connect an edge to it... (exept if you are making the F2L pair, but the system remains)

1. R U2 R' U L' R U R' U' L U R U2 R' U L' R U R' U' L U L' R U R' U' L U L' R U R' U' L U L' R U R' U' L U L' R U R' U' L
U' R U R'

Here is the orietation sequence R; the edge manipulation before the orientation is U'

/*
I have to admit when I started learning this method, I knew little of cubing, I used the "edges first" for a long time, maybe I should have learnt another method first, but I wasn't very fond of the algs.
*/

still,
mvg

 

[Johannes91]

Leaving just 3 corners can always be achieved by using pair 3-cycles, with sometimes an R conjugate.

The hard part is doing it efficiently. Using several long algs or commutators solves the cube but misses the point. You could just as well use Petrus or some other method.

 

[Am1n-]

Yes indeed thats the hardest part, but therefore there is a sufficient amount of information on Heise's site. The recognition of the cases is, in my opinion, just a matter of experience.

EDIT: on topic: there isn't any tutorial or sample solutions, its like learning intuitive F2L, but harder (in my opinion)

mvg

 

[Wacky]

Thanks for posting those examples Johannes91.

But I disagree with you on it not needing further explanation. I find with intuitive methods it can be quite helpful to see some examples because sometimes it can make me go "what, you can do that!?"

I see in your examples you seem to prefer:

1) hide an oriented corner on U and shuffled the edge, similar to fridrich edge pairing (what I did as well)

2) taking a corner on D that's facing R and twisting it with R U' R' (neat). In the past I've only really used the corner on D when it's facing L.

Going through the two examples on Heise's website again it seems he likes pairing the edge in the F2L slot with a corner on the U layer. (seems to be more helpful than I originally thought)


Not having a 3 cycle at the end is annoying, it's a 14 move commutator instead of 10... hate getting caught in the two-edge swap parity before that though, although it's avoidable with better look-ahead.


"Also don't forget you can use the left layer for building the pairs and for permuting the pairs." - Am1n-

What left layer would you be talking about? Sorry, I don't quite understand.

 

[Am1n-]

well you have left (L L' L2) which is oposite to right (R R' R2).
Its used in Heise's examples ("move it out of the way" and the "R conjugation for oposite pairs"). In these examples its used for pair/edge permutation but I sometimes use it to pair up my pairs.

mvg

 

[Johannes91]

A simple solver found a better solution to scramble #4 above: R U' R' U' R U R'. And I missed a 4-mover in #5: U' R U R'. :fp

More scrambles and computer-generated solutions:

1. R L U' R' U L' R U R' L F2 R' F2 R L' U' L U' R U L' U2 R' F2 L' U' L U F2 L F2 R' F2 R L' U2 L F2 R' F2 R L'
U' R U2 R' U2
2. R L' U R' U' L B' R U' R' U B U B' R U' R' U B U2 R' L F2 R F2 L' R U' R' F' U2 L' U' L F R' L F2 R F2 L'
U2 R U2 R'
3. R L' U R' U' L R U2 R' U2 L' U R U' R' L B' R U' R' U B R U2 R' L F2 R' F2 R L' U B' R U' R' U B R L U' R' U L'
R U R'
4. R U2 R2 L F2 R F2 L' R U R' F2 U' L' U L F2 R' L F2 R F2 L2 R U R' U' L
U2 R U' R'
5. L F2 R' F2 R L' F' L' U L U2 F B' U' R U R' B R' L F2 R F2 L' F2 U' L' U L F2 B' U' R U R' B U F' U2 L' U' L F
R U' R' U2
6. F2 U' L' U L F2 L F2 R' F2 R L' F' L' U L U2 F R L' U R' U' L F' L' U L U2 F R L U' R' U L' R U R' F' L' U L U2 F
R U2 R' U
7. R L U' R' U L' F' L' U L U2 F L U' R U R' L' B' U' R U R' B R' L F2 R F2 L' U L U' R U R' L' U2 U R L' U R' U' L
U'
8. U R U' R' U2 R U2 R' L' U R U' R' L F' L' U L U2 F R U' R' U2 B' R U' R' U B L' U R U' R' L F' L' U L U2 F

9. B' R U' R' U B F' U2 L' U' L F B' U' R U R' B F2 L' U' L U F2 L U' R U R' L' F2 U' L' U L F2 F2 L' U' L U F2
R U' R' U'
10. R U' R' L F2 R' F2 R L' U2 L U' R U R' L' F2 L' U' L U F2 R L U' R' U L' F2 U' L' U L F2 L F2 R' F2 R L'
U2 R U R'
11. R U2 R' L F2 R' F2 R L' B' U' R U R' B U2 R U R' U' R U' R' F' L' U L U2 F R U2 R' F2 U' L' U L F2 U'
U2 R U R'
12. B' R U' R' U B L F2 R' F2 R L' U' F2 U' L' U L F2 F2 L' U' L U F2 L F2 R' F2 R L' F2 L' U' L U F2 U'
R L U' R' U L'
13. R U R' F2 U' L' U L F2 R L' U R' U' L R' L F2 R F2 L' U' F' L' U L U2 F R U' R' F' L' U L U2 F R U' R'
U'
14. F2 L' U' L U F2 L U' R U R' L' B' U' R U R' B F' U2 L' U' L F U' B' R U' R' U B R L' U R' U' L B' U' R U R' B
R U R'
15. R L' U R' U' L F2 L' U' L U F2 U' F2 U' L' U L F2 U2 B' R U' R' U B R U R' F2 U' L' U L F2 F' L' U L U2 F
U2 R U2 R' U
16. B' U' R U R' B R U R' L' U R U' R' L R U2 R' B' R U' R' U B U L' U R U' R' L R U' R' B' U' R U R' B
U R U' R'
17. R U2 R' L F2 R' F2 R L' F' U2 L' U' L F R U2 R' B' R U' R' U B R L U' R' U L' B' U' R U R' B R L' U R' U' L
R U2 R' U2
18. R L U' R' U L' R L' U R' U' L F' L' U L U2 F L' U R U' R' L U' B' U' R U R' B L' U R U' R' L R U R'
U R U R'
19. R U' R' L F2 R' F2 R L' R L' U R' U' L R U2 R' F' U2 L' U' L F R L' U R' U' L R U' R' F2 U' L' U L F2
R L U' R' U L'
20. F' L' U L U2 F L U' R U R' L' F2 U' L' U L F2 R U2 R' F' U2 L' U' L F R U2 R' F' U2 L' U' L F L F2 R' F2 R L'

 

[nigtv]

Sorry to bump a thread, but I feel your guy's pain with Heise. It took me forever to even be able to get it, and almost 2 months to get even sub 60 with it (although, it is much more fun than any other method). In a sense, the first step (well, finishing the four blocks) is the easiest, but it's probably also the hardest to get even close to efficient with it, I think it's because it gives a solver much more choice than they actually know how to use, so almost always completely miss the point (myself included)

Like others have said in here, you should try not to use algs. When I was first learning, it was hard to get that youre NOT just doing f2l minus a pair. There's also the bit about where you can and can't have any given block, since you don't have to have them TOTALLY in the right place. If it helps, I could say what I started using in my head, if not, oh well, but I think that anything is better than ONLY the heise site (it's kind of purposefully vague)

That first block is almost always easy if you aren't trying to look ahead. As you probably know on the site, it only needs to match up with one center, the center with the most colors in that set of cubes (the center where the '2x2' square is). I'm not a real big guy on doing this in as little moves as possible (it's usually trivial enough to just work it into the next block(s)). Without looking ahead, I'd say shoot for about 5. This block can be in any of its four 'spots', as long as that '2x2' square matches its center, so you probably should wait to move it (even though your entire gut is telling you to finish the 2x2x2).

Like Heise says on his site, for the second block, you should almost always go for the one that finishes the 2x2x2. I've never really wanted to look into it much further than that, more try to keep in mind that the 2x2x2 is block one and two, so they never HAVE to fully match up, and block 2 can be swapped into block 1, as long as an edge from 1 takes 2's place (not physically, just mentally).
(nigtv realizes this is getting long, sorry.)

From there, you should be able to tell which color you are going to want to leave off of the other two blocks, and if you know that, you can probably start to guess which one you want to leave off before you even start solving block 1, which can help with the turn count. It's hard to not make them fully match up, and it's hard not to use algs, but it's worth it in the end (sometimes making the ENTIRE rest of the solve <10-15 turns average). Alg's can really hurt your turn count, i think, because you're giving yourself that # of moves WITHOUT ANY tracking of the rest of the cube (well, as long as you dont know how the algs effect the entire cube), it's not as big of a deal later on (I use sune quite often actually, on really bad luck solves). The later on count can be helped a lot by using the fact that you dont have to fully match up all of those blocks at the start, and use them to orient some edges and move stuff around without wasting any moves.

Sorry that was long...
...oh, one more thing, I like that someone said "may not be the method for you". Heise is a pretty novel method. It's probably one of the hardest methods to really "master", and to be honest, I don't think that many people would find it all that rewarding. It's also a pretty 'macro' method, and isn't that much more efficient without extensive look ahead. I love it, but yrmv.

 

[Am1n-]

Thread bumping:

A few examples:
1. R U2 R' U L' R U R' U' L U R U2 R' U L' R U R' U' L U L' R U R' U' L U L' R U R' U' L U L' R U R' U' L U L' R U R' U' L
U' R U R'
2. R U R' U R U2 R' U R U' R' U R U' R' U L' R U R' U' L U R U' R' U R U2 R' U R U2 R'
U R U R'
3. L' R U R' U' L U L' R U R' U' L U R U R' U R U2 R' U R U' R' U R U R' U R U' R' U L' R U R' U' L
R U R' U' R U2 R'
4. R U2 R' U R U R' U L' R U R' U' L U R U2 R' U L' R U R' U' L U L' R U R' U' L U R U2 R' U L' R U R' U' L
U' R U R' U2 R U' R'
5. R U R' U L' R U R' U' L U R U' R' U R U' R' U R U2 R' U R U R' U R U R' U R U' R'
R U R' U2 R U' R'

I tried some full solutions for these scrambles:
1

• a. pairs: U' R U R'

• b. leave 3 corners:
  • U R U2
  • 3-cycle: (F' R' F) U2 (F' R F) U2
  • U2 R'

• c. Commutator: A9: B2 D2 (B' U2 B) D2 (B' U2 B) B2

2

• a. pairs: U R U R'

• b .leave 3 corners:
  • U
  • 3-cycle: M2 U {M2 U2 M' U2 M' U2} U M2

• c. Commutator: Cyclic shift: [(F R') U2 (R F')][(R' F) U2 (F' R)]

3

• a. pairs: U2 R U R'

• b. leave 3 corners:
  • U' R U R'
  • 3-cycle: M2 U {M2 U2 M' U2 M' U'} M2

• c. Commutator: A9: U' (F D F') U' (F D' F') U2

4

• a. pairs: R U' R' U' R U R'

• b. leave 3 corners:
  • U2 R U R'

• c. Commutator: A9: (D2 B D) F2 (D' B D) F2 D

5

• a. pairs: U' R U R'

• b. leave 3 corners:
  • (U' L U2) {R U' R'} (U2 L U')
  • 3-cycle: (B R B') U2 (B R' B')

• c. Commutator: pure:U' (F D F') U (F D' F')

I'm happy about all solutions exept the 5th. I think i could be shortened a lot, but I don't know how... Can anyone try this one?

mvg

 

[Wacky]

Failing to improve on 5 without not using Heise...

5.

20 Moves:
a. Pairs U' R U R' (4)
b. Shuffle around U' R U' R' (4)
c. Y Perm (12)

Okay, that was a cop-out and not heise....

28 Moves:
a. Pairs U' R U R' (4)
b.1. Solve 3 of 5 edges R U' R' U2 (4)
b.2. Solve other 2 edges B R B' U B R' B' (7)
c. Solve last two corners R' D2 R F D2 F' U' F D2 F' R' D2 R (13)

(I think it's pretty much the same length you had)

 

[Am1n-]

I found another one (with B2 D' conjugate, where I was looking for) but I failed in making it shorter
- pairs: U' R U R'
- solving pairs: F' d' F { B2 D' R2 D B2 } F' d F (U')
- corners: [U' B U R B R' : F]

EDIT: by replacing the d and d' by E and E' you create a 3-cycle at the end; wich is shorter than flipping

- pairs: U' R U R'
- solving pairs: F' E' F { B2 D' R2 D B2 } F' E F U'
- corners: [F R F': L2] (8)

mvg

 

[nigtv]

I've been having much better results with this as of late when I already have a pair somewhere, or if only two edges are swapped on the LL, as well as if there is one edge/pair swap between the slot and the LL.

EDIT: The two pair approach also seems to work if you have more than one unsolved block in your f2l.

 

[milkolate]

Heise for Beginners?

I've been speedcubing for a couple of years. Now, I want to teach my friend to solve but she doesn't want to memorize any algos. She wants to solve the cube completely intuitively. I think the Heise method is the only way to go. But I can't figure out the method!

I'm stuck with the "Edges First" step. I can put the edges in their right places but I dont understand how to do the 3-corner cycle for the corners (I'm not yet referring to the last step of the method). I understand the concept but I can't do it on a cube. Any help would be appreciated!

 

[Cyrus C.]

You can do CF with no algorithms.

 

[milkolate]

You can do CF with no algorithms.

Really? I thought CF has algos... Can you point me to a tutorial with no algos?

 

[Cyrus C.]

You can do CF with no algorithms.

Really? I thought CF has algos... Can you point me to a tutorial with no algos?

Solve the first layer of corners, use commutators for the last layer of corners, Ortega with commutators.

Basically for the OLL use (RUR'U'), PLL use (RUR') & (RU'R'). For the middle layer at the end of ortega I'd use (M'UM'UM'), although it's turned into more of an algorithm now & I use it for OLL now.

 

[milkolate]

But I just dont understand commutators (in practice)

 

[Cyrus C.]

But I just dont understand commutators (in practice)

Basically in a commutator, you use a certain move like (RUR'U')x2, change something, like D, & Reverse the commutator (URU'R')x2. This is how you could solve the LL's corners. I also find commutators useful in FMC & as pure algorithms.

 

[Baian Liu]

But I just dont understand commutators (in practice)

Basically in a commutator, you use a certain move like (RUR'U')x2, change something, like D, & Reverse the commutator (URU'R')x2. This is how you could solve the LL's corners. I also find commutators useful in FMC & as pure algorithms.

Yes, but no need to use such a long sequence to do that.

 

[Zarxrax]

Badmephisto did a tutorial on commutators: http://www.youtube.com/watch?v=7yZoDi_B1lI

But I have to admit, I still only *kinda* understand it after watching that.