Improvements for M2/R2

[martijn_cube]

What would your solution be for : DF --> DB --> FL?
The thing i can now think of is: B2 M2 (U'L'U) M2 (U'LU) B2, but i'm not sure if this is the right way? It is a bit shorter than normal M2, but maybe not faster.

 

[Mike Hughey]

What would your solution be for : DF --> DB --> FL?
The thing i can now think of is: B2 M2 (U'L'U) M2 (U'LU) B2, but i'm not sure if this is the right way? It is a bit shorter than normal M2, but maybe not faster.

That's exactly how I do it; I've been doing it that way for months. But you're right, it might not really be faster. It helps me because I'm slow at everything, so reducing number of moves almost always helps for me.

 

[trying-to-speedcube...]

Well, it's not really anything like M2, but if you don't think it's fast enough, I'd just do (y) D2 R' E' R D2 R' E R (y')

 

[Mike Hughey]

Well, it's not really anything like M2, but if you don't think it's fast enough, I'd just do (y) D2 R' E' R D2 R' E R (y')

Yeah, BH would probably be best for me, but I found that a variant of what deadalnix is doing worked really easily for me without too much extra work. I need to switch to full BH someday, I think. I suspect it will be fastest for me.

 

[deadalnix]

What would your solution be for : DF --> DB --> FL?
The thing i can now think of is: B2 M2 (U'L'U) M2 (U'LU) B2, but i'm not sure if this is the right way? It is a bit shorter than normal M2, but maybe not faster.

L'UMU2MU2M2U'L

Yeah, BH would probably be best for me, but I found that a variant of what deadalnix is doing worked really easily for me without too much extra work. I need to switch to full BH someday, I think. I suspect it will be fastest for me.

I Think that my solution have these advantages on BH :
1/ This is more finger friendly.
2/ This require less thinking.

BH need less moves.

Anyway, I think BH is really great on corners and center/wings on big cubes. But the edges is the bad side of BH in my opinion. Theres many different types of commutator for edges. With my method, you have edge grouped in category, and with the category of the two edges, you can know the pattern you have to use.

I'm not so fast at blind, I'm more a cube theoretician. And I'm probably not the most objective guy on this subject, but I think my method is at least as good as BH for edge, maybe better.

 

[martijn_cube]

What would your solution be for : DF --> DB --> FL?
The thing i can now think of is: B2 M2 (U'L'U) M2 (U'LU) B2, but i'm not sure if this is the right way? It is a bit shorter than normal M2, but maybe not faster.

L'UMU2MU2M2U'L

Wow that's a nice one. just a combination of the two M2's but alot easier

 

[Mike Hughey]

I Think that my solution have these advantages on BH :
1/ This is more finger friendly.
2/ This require less thinking.

BH need less moves.

Anyway, I think BH is really great on corners and center/wings on big cubes. But the edges is the bad side of BH in my opinion. Theres many different types of commutator for edges. With my method, you have edge grouped in category, and with the category of the two edges, you can know the pattern you have to use.

I'm not so fast at blind, I'm more a cube theoretician. And I'm probably not the most objective guy on this subject, but I think my method is at least as good as BH for edge, maybe better.

I've been using something pretty close to your method for edges for the past few months. I agree it's pretty good. But I'm slow at M2, so that keeps it from being really super good for me. It's finger friendly, but my fingers are slow and awkward.

As for BH requiring less thinking, not true if you've really got it down. BH should require zero thinking. If not, you haven't really learned it yet. But there's no reason why you can't use pseudo-BH where you replace certain minimum move algorithms with maximum speed algorithms. And of course your method can also require zero thinking, so I guess ideally they're the same on thinking.

Mainly I want to get where I've got everything memorized so there's no thinking. Your method requires little thinking, but there's still some involved, the way I do it. (And I also do it sub-optimally.) I guess I'm probably going to compare your "algorithms" for each pair with the standard BH algorithms, and see which one is best and memorize it for each case.

 

[martijn_cube]

I have another strange one during my practice solves. But i cant really come up with a solution.
DF-> FU -> BD.
But i red somewhere that BD was something special?
Maybe like this: B M’ (FR2F’) M (FR2F’) B’
Sorry for posting these questions again. I should just wait for you to complete your tutorial. But it's just a very cool method. especially when your used to M2. But indeed i don't understand all the cases.

 

[trying-to-speedcube...]

x F E R U R' E' R U' R' F' x'

It's the normal M2 alg for BD or FU.

 

[deadalnix]

U'RMUR'U'MURU'(M2R)U

 

[martijn_cube]

x F E R U R' E' R U' R' F' x'

It's the normal M2 alg for BD or FU.

wow, nice.

U'RMUR'U'MURU'(M2R)U

Alg isn't working. I think a wrong turn somewere. The last R should be R'. but then FU and BD are flipped. So your alg is for FD -> FU -> BD i think.

 

[Marcell]

Hi guys! No matter how much I liked R2 (it is absolutely braindead, also supercube-safe, and it goes well with M2), I was still looking for ways to improve it - mainly to get rid of the long and uncomfortable algs for the R-layer corners thus making the method more effective. I came up with some ideas, collected the neccessary algorithms for them and named the whole stuff R2+. I learned the algs, used R2+ in competition and won. I decided that I was satisfied so I wrote a description about it. You can read it at http://kerulo.uw.hu/rubik/R2+.html. Let me know what you think!

 

[cmhardw]

Wow Marcell, that is very neat! I wonder what Stefan will think of R2+, I hope he sees this!

Chris

 

[MarcusStuhr]

Wow, impressive stuff... do you guys find that M2/R2 fits well with memory mnemonic systems (more easily than other methods?)

 

[deadalnix]

I just read abour R2+ . This is really interesting, but I think that the alg set have to be reworked. Many of them are really finger hostile.

The method itself look very promizing. However, I do prefer BH on corners because of the small number of cummutator types (as opposite to edges).

 

[Sakarie]

Marcell:

I haven't read the whole page (since I'm not interested in using it over BH), so I don't really understand the notation. Well, doesn't really matter, cause I can still do the algorithms, and compare to what I would (want to) use.

You say you learnt the algs, so you learnt an algorithms for every case, not "inventing" commutators? That would make it possible for you to learn non-optimal algorithms, suited better for speed right?

I went through the algorithms in the first block, to see if I could find anything faster than what you've written. Decide for yourself what you like the best, but I think "mine" is faster. OBS, I don't use all of these myself, since it's easier to find move-optimal than speed-optimal. But I wish I did.

My list:
FRU + z L' U2 L D' L' U2 L D z'
FRU - D' R' D R U2 R' D' R D U2
FRU 0 x U L2 U' R' UL2U' R x' (The same)
UFR + L2 U' R2 U' L2 U R2 U' L2 U2 L2
UFR - z' R2 D' R'U2R D R'U2R' z (The same)
UFR 0 z'x R2 D R U2 R' D' R U2 R x'z
RUF + x' D' R U' R' D R U R' x
RUF - x' U' D' R U R' D R U' R' U x
RUF 0 R' D'R U2 R'DR U2 (The same)

Your list:
FRU + z' R'D2R U' R'D2R U z
FRU - z R2U' L2U R2U' L2U z'
FRU 0 x UL2U' R' UL2U' R x'
UFR + LD'L F2 L'DLD' F2 D L2
(FR'F'R)3 U2 (FR'F'R)3 U2
UFR - z' R2 D' R'U2R D R'U2R' z
UFR 0 D R2 F'LF R2 F'L'F D'
RUF + z'x' RU2R' D RU2R' D' xz
RUF - x R'D2R U R'D2R U x'
RUF 0 R'D'R U2 R'DR U2

 

[Marcell]

@Chris: the fact that you commented positively on my post actually made me jump up and shout out loud.

@Marcus: I don't know, but I'd say it's more likely the cuber and not the solving method that defines what memo system would be optimal.

@deadelnix: I forgot to write "suggest better algs" at the end of the page. I corrected that now.

@Sakarie: I learned algs for case types rather than individual cases, thus reducing the number of algs to learn and making it possible to really know them by heart and execute them without thinking rather than inventing them on the fly. This algs are usually two moves longer than the optimal algs (setup and resetup).
Also, thanks a lot for your list. I will update my page when I have some time.

 

[riffz]

Here's what I use. Some are the same as yours or Sakarie's. I don't use R2, but I use all these algs for corners anyway.

FRU +: x R U' L U R' U' L' U x'
FRU -: z R2 U' L2 U R2 U' L2 U z'
FRU 0: R2 U' L' U R' U' L U R'

UFR +: R' F' R2 F R U2 R' F' R2 F R U2
UFR -: z' R2 D' R' U2 R D R' U2 R' z
UFR 0: D x R2 U' L U R2 U' L' U x D'

RUF + : x' D' R U' R' D R U l'
RUF -: l' D2 R U R' D2 R U' x'
RUF 0: R' D' R U2 R' D R U2

 

[Gavin]

Hi guys! No matter how much I liked R2 (it is absolutely braindead, also supercube-safe, and it goes well with M2), I was still looking for ways to improve it - mainly to get rid of the long and uncomfortable algs for the R-layer corners thus making the method more effective. I came up with some ideas, collected the neccessary algorithms for them and named the whole stuff R2+. I learned the algs, used R2+ in competition and won. I decided that I was satisfied so I wrote a description about it. You can read it at http://kerulo.uw.hu/rubik/R2+.html. Let me know what you think!

R2+ is too many algs for my taste, but you're 12-15 sec execution with it is impressive.

 

[Sakarie]

Here's what I use. Some are the same as yours or Sakarie's. I don't use R2, but I use all these algs for corners anyway.

FRU +: x R U' L U R' U' L' U x'

That one is perfect. Don't understand that I didn't see it.​

FRU -: z R2 U' L2 U R2 U' L2 U z'
FRU 0: R2 U' L' U R' U' L U R'

That one is really nice too!​

UFR +: R' F' R2 F R U2 R' F' R2 F R U2
UFR -: z' R2 D' R' U2 R D R' U2 R' z
UFR 0: D x R2 U' L U R2 U' L' U x D'

RUF + : x' D' R U' R' D R U l'
RUF -: l' D2 R U R' D2 R U' x'
RUF 0: R' D' R U2 R' D R U2

What method are you using? This modified commutator-method?