R2, 3 cycle, or Pochmann for corners?

[alexc]

Why not try TuRBo?

Oh, I didn't think of that. It's kind of a lot of algos to learn and I don't know if it will be any better than R2.

 

[Erik]

It's only 18 algs of which I'm sure you know at least like 6 or 8

 

[cmhardw]

What does BH stand for?

Guessing "Beyer-Hardwick", maybe? Maybe this is the optimized commutators method?

maybe...

Chris

 

[cmhardw]

Why not try TuRBo?

Erik, I do recommend to people the TuRBo method, I think it has incredible potential to be fast. I like how it allows you to memorize just a few short, fast, sequences to use to permute and orient all the time. I may have convinced Andrew Kang to learn TuRBo, not sure if he stuck with it, but I at least got him very interested in the idea. So you may soon have another fast TuRBo solver added to the ranks.

I also like TuRBo method, but for my own solving I have to say that I prefer the BH method (wait for a bit to see it)

Chris

 

[Erik]

great Chris! I've got some improvements to it too I think now (though i have forgotten all algs now )

 

[Joël]

I tried out some methods. I am pretty sure I will still use 3 cycle w/ visual memory as my regular blindsolving method. For multi, I will use either commutators or just set up the pieces in A perms, etc. I tried a few sighted solves. I mostly used A perms, but I occassionally threw in an easy commutator. I really like commutators, but I have trouble doing them when the interchangeable slice isn't U or D. Using free setups with A perms was also good. The setup moves were sometimes short, but sometimes long, too. I think that I will use A perms or other corner algs when the setups are good, and use commutators when the setups are bad. Lastly, does anyone know good algs for the following corner cycles: (1 3 7) (5 7 3) Thanks!

I am assuming you use Macky's numbering scheme:

1: UFL, 2: UFR, 3: UBR, 4: UBL, 5: DFL, 6: DFR, 7: DBR, 8: DBL.

So (1 3 7) = UFL -> UBR -> DBR:

y U2 (R'F'R2 FR) U2 (R'F'R2 FR) y'

I use this commutator for pretty much all the 3 cycles with 2 corners in the U layer and 1 corner in the D layer. The (R'F'R2 FR) sort of swaps the URF and DRF corners. Then a U, U or U2, and (R'F'R2 FR) swaps two other corners. For example:

(R'F'R2 FR) U (R'F'R2 FR) U'

(5 7 3): I'd do y (RUR'U') * 3 D2 (RUR'U') * 3 D2 y'

 

[dbeyer]

Nice Joel, check this one out:

L2 U [R2 U L2 U' R2 U L2 U'] U' L2
Hence:
L2 U R2 U L2 U' R2 U L2 U2 L2, it's optimal in HTM, and pretty nice considering its RLU

 

[Joël]

Nice Joel, check this one out:

L2 U [R2 U L2 U' R2 U L2 U'] U' L2
Hence:
L2 U R2 U L2 U' R2 U L2 U2 L2, it's optimal in HTM, and pretty nice considering its RLU

R2 U L2 U' R2 U L2 U'... I've seen it before... So what is the logic behind that?

<U'R2U, L2> = U' R2 U L2 U' R2 U L2. So R2 U L2 U' R2 U L2 U' is just a shifted commutator...

I might use that too, now that I understand why it works .

 

[KConny]

(R2 D' R2' D R2 U2)*2 or (U2 R2 D' R2' D R2)*2 should do the trick.

 

[dbeyer]

That's 12 Moves though. It's a good substitute though. I'm actually amused by Joel's solution.

 

[Joël]

That's 12 Moves though. It's a good substitute though. I'm actually amused by Joel's solution.

I was actually not amused by it when I found it. My hands just could not get the order in which the faces where moving... But after some practice, I started to like it!

I don't really like KConny's move with the D D' moves, but I like it a lot when mirrod in the E layer, (R2UR2U'R2 D2') * 2.

I guess these things are very personal .

 

[Mike Hughey]

I really like KConny's move (got it from Macky's tutorial when I first started BLD). It's one of my fastest algorithms. But that probably just shows how slow all my algorithms are.

 

[MarcusStuhr]

How does the Turbo method work?

I've been developing a new system of my own -- it still has a few kinks to work out, but I've been able to get very good times with it. It is inspired by freestyle commutators but perhaps more braindead in nature. I've always had trouble finding a good method that is both fast and effective and yet easy to compress down in memory and easy to unpack. Does Turbo solve these issues?

 

[Lucas Garron]

How does the Turbo method work?

I've been developing a new system of my own -- it still has a few kinks to work out, but I've been able to get very good times with it. It is inspired by freestyle commutators but perhaps more braindead in nature. I've always had trouble finding a good method that is both fast and effective and yet easy to compress down in memory and easy to unpack. Does Turbo solve these issues?

TuRBo is freestyle 3-cycle with standard resolution for each cycle (what I like to call "restricted freestyle" ). 3 pieces are brought to 3 positions a layer, and their stickers are cycled with one of a few algs (depending on the location of the stickers to be cycled). It's on Erik's old site.

 

[KJiptner]

TuRBo:

for Edges: http://erikku.er.funpic.org/rubik/turboedges.html
for Corners: http://erikku.er.funpic.org/rubik/turbocorners.html

I'm working on it btw.

 

[alexc]

Kai, will you switch to TuRBo?

 

[KJiptner]

Kai, will you switch to TuRBo?

I'm experimenting.

 

[Johannes91]

<U'R2U, L2> = U' R2 U L2 U' R2 U L2. So R2 U L2 U' R2 U L2 U' is just a shifted commutator...

It's also a pure commutator: [R2, U L2 U'].

 

[Joël]

<U'R2U, L2> = U' R2 U L2 U' R2 U L2. So R2 U L2 U' R2 U L2 U' is just a shifted commutator...

It's also a pure commutator: [R2, U L2 U'].

Oh.. Of course .

 

[Inusagi]

Pochmann is nice, but the move count is just way too high to reach acceptable speeds.
R2 is faster, but you're right, you do need to learn several algorithms.
3-cycle is fast, but again, for multibld you need to memo twice as much.

Really, I can't decide for you. I've just said what I think.

When you say that 3-cycle and R2 is fast, which one do you mean is faster?