Fastest BLD method?

[rubiksczar]

whats the fastest BLD method and can someone give a link to it?

 

[Kirjava]

Methods don't have speeds.

 

[ianini]

Freestyle.

 

[Kirjava]

Haiyan uses speed-optimised BH btw.

 

[riffz]

Ideally you would want to gather a collection of about 800 algorithms to cycle any 3 corner stickers from a fixed buffer, and the same for edges. (Speed optimized BH as Kirjava said). But if you don't want to go that route, you could just learn M2 for edges and try learning how to do commutators for BH corners.

 

[iRiLLL]

Fridrich

 

[abr71310]

There's a method that's called "OMGWTFBBQGGNOREROFLMFAOKTHXBI".

The entire solution is guaranteed to be ONE solution, and it always works for every scrambled cube. The second you pick up that cube for inspection you instantly see the solution without much thought.

It's so godly that nobody except for Chuck Norris can learn it.
And when Chuck Norris looks at a cube, it solves itself in fear.
Think about how scared that cube would be if Chuck Norris was touching it BLINDFOLDED.

 

[Chuck]

It's so godly that nobody except for Chuck Norris can learn it.

Yes I am.

 

[Micael]

whats the fastest BLD method and can someone give a link to it?

The method that Haiyan Zhuang uses is the fastest. It is similar to BH. The difference is that the algs are more like "speed algs" than commutators. So you have to learn all those "speed algs". With all the special cases (mostly parity I guess), I think Haiyan Zhuang uses about 1000 algs. There was a post from him about that.

Good luck.

 

[MrMoney]

All of the above.

The thing to remember is that fewer moves does not always mean faster times.

BH is a commutator-method. This means the method cycles 3 pieces at one time using optimized move-count algorithms which range from 4-12 moves. The turns and slicemoves can be tricky for the fingers and therefore people are disencouraged from learning the method.

Some people (most noticeably Haiyan Zhuang) use what we call speed-optimized algorithms for all cases (440 edge-cases & 378 corner-cases + xyz parity) and with this achieves extreme BLD-times.

From my personal view, I would go forward in this fashion:

- Learn 3OP corners (Thus understanding 3cycles and "memo-method" for freestyle corners)
- Learn M2 edges (really fast method)

After you are fluent with this you can move on to BH corners and from then on decide if BLD is really your thing and worth the time learning 800+ algorithms.

Have fun BLD-solving!

 

[Anthony]

Haiyan uses speed-optimised BH btw.

Ideally you would want to gather a collection of about 800 algorithms to cycle any 3 corner stickers from a fixed buffer, and the same for edges. (Speed optimized BH as Kirjava said).

The method that Haiyan Zhuang uses is the fastest. It is similar to BH. The difference is that the algs are more like "speed algs" than commutators. So you have to learn all those "speed algs". With all the special cases (mostly parity I guess), I think Haiyan Zhuang uses about 1000 algs. There was a post from him about that.

lol 800 or 1000. iirc, Tim Sun told me that Haiyan only uses ~40 algs. But he's mastered those algs and just sets up to them in solves.

 

[Mike Hughey]

Haiyan uses speed-optimised BH btw.

Ideally you would want to gather a collection of about 800 algorithms to cycle any 3 corner stickers from a fixed buffer, and the same for edges. (Speed optimized BH as Kirjava said).

The method that Haiyan Zhuang uses is the fastest. It is similar to BH. The difference is that the algs are more like "speed algs" than commutators. So you have to learn all those "speed algs". With all the special cases (mostly parity I guess), I think Haiyan Zhuang uses about 1000 algs. There was a post from him about that.

lol 800 or 1000. iirc, Tim Sun told me that Haiyan only uses ~40 algs. But he's mastered those algs and just sets up to them in solves.

Yes, I know some people have said that, but Haiyan himself said he knows 800 or 1000 on here in the past. I suspect that the truth is a little of both - most of his 800 or 1000 are practically setups into the algorithms.

If you know BH, you know that it seems reasonable to think that both statements are a little bit true. An awful lot of BH algorithms are so similar that they're almost exactly the same algorithm. That's why BH is so easy to learn.

Edit: Oh, and one more thing. In the two reconstructed solves by Haiyan I've seen, I noticed that almost all of the corner algorithms were essentially identical to the (BH) ones I use. And about half of the edges algorithms were the same as mine. So it certainly seems like he's essentially using speed-optimized BH, based on those sample solves. For the algorithms that were different from mine, they usually resembled typical speed-optimized OLL or PLL-type algorithms, being one or two moves longer than a BH algorithm, but quick to execute. This certainly doesn't mesh with using only 40 algorithms - it's more complicated to remember than standard BH.

 

[Swordsman Kirby]

Edit: Oh, and one more thing. In the two reconstructed solves by Haiyan I've seen, I noticed that almost all of the corner algorithms were essentially identical to the (BH) ones I use. And about half of the edges algorithms were the same as mine. So it certainly seems like he's essentially using speed-optimized BH, based on those sample solves. For the algorithms that were different from mine, they usually resembled typical speed-optimized OLL or PLL-type algorithms, being one or two moves longer than a BH algorithm, but quick to execute. This certainly doesn't mesh with using only 40 algorithms - it's more complicated to remember than standard BH.

Which reminds me, I've seen him use two T-perms (with setups) in a row during corners phase. He might use those for those ugly 2 2-cycle cases.

 

[Jani]

RF Method???

 

[Mike Hughey]

Edit: Oh, and one more thing. In the two reconstructed solves by Haiyan I've seen, I noticed that almost all of the corner algorithms were essentially identical to the (BH) ones I use. And about half of the edges algorithms were the same as mine. So it certainly seems like he's essentially using speed-optimized BH, based on those sample solves. For the algorithms that were different from mine, they usually resembled typical speed-optimized OLL or PLL-type algorithms, being one or two moves longer than a BH algorithm, but quick to execute. This certainly doesn't mesh with using only 40 algorithms - it's more complicated to remember than standard BH.

Which reminds me, I've seen him use two T-perms (with setups) in a row during corners phase. He might use those for those ugly 2 2-cycle cases.

Wow, that seems weird. I'd like to see an example.

I guess that means he's not quite matching typical BH in those cases, because I generally don't notice 2 2-cycle cases. I just see it as 2 3-cycles (breaking a new cycle), often never even realizing that it was actually 2 2-cycles (or at least never really thinking about it).

 

[EVH]

It seems to me that speed is based on the algorithms you use and how fast you memorize.

 

[dbeyer]

For Parity, what I used to do was cycle
Buffer Edge -> Last Edge -> Adjacent Edge.

So I had a 1x1x2 block of the Buffer Edge and Corner.
I would then create a two swap of the two edges in a 1x2x2 block.

This leaves 21 cases of parity left. Buffer to any sticker of any of the other 7 cubies (3x7=21).

I used conjugations of the Y perms. Essentially, Y perms, J perms, and D/F/L turn setups.

At the expense of one extra commutator, you save setups. I'm not training for blind solving anymore, but its the method I've used to get sub 8 on 4x4 and sub 16 on 5x5 back in 2008. I've been doing jiu jitsu tournaments since the summer of 2008. So back then those times were quite good.

The beauty of this system is its big cube applicable, allowing the parity fixes to be easy, and its center safe, because you only use outer slice turns.

Later,
DB

 

[dbeyer]

The BH system is based off of 5 major techniques, and for corners there are 6 case types.

Inverting
Cancellations
Setups
Cyclic Shifting
5-move insertions.

You can use these techniques, and solve all of the cases.
Pure, A9s, Orthogonal, Cyclic Shifts, Columns, and Pyramids.

Its a nice system. Havent touched in a while, but its quite complete.