The Future of Cubing?

[tfkscores]

Sorry I didn't know what thread to put this in. My question is asking for your opinion on where cubing will go from here. A little while after the Rubik's cube had come out the fastest times were getting down to around 20-30 seconds. At that time you had to usually make your own method since they didn't have YouTube to teach you. We have now progressed to some very complex methods with many algorithms and fast times such as I believe Erik's time of 6.2 or something. What do you think is the fastest time for a human to ever achieve on a Rubik's cube with a non-lucky solve. Will there be even more complex and faster methods to be discovered? It seems that we have progressed so quickly (thanks to the internet and technology) that we won't be able to progress much more. To sum it up, will it ever be possible for a human to get a sub 5 or sub 4 solve? Or could better methods be created for this. It just seems to unlikely for me.

 

[jacob15728]

I think that once the top cubers have more time to practice, sub-10 averages will become more common. Maybe someone can get sub 9 or even sub 8 with full ZB. Who knows?

 

[Cyrok215]

I think most current methods max out at sub-17 averages for most people but I believe, you could get sub-5 second solves with other methods that we haven't thought of.

 

[vgbjason]

Eric's PB is something like 5.05, and I'm certain that new methods and ideas will keep coming. Sub-5 doesn't seem impossible.

But I don't know how long we can keep making ideas. until there's a 30 move every time speedsolving method, we really can't say we've maxed out potential

 

[Cyrok215]

Eric's PB is something like 5.05, and I'm certain that new methods and ideas will keep coming. Sub-5 doesn't seem impossible.

But I don't know how long we can keep making ideas. until there's a 22 move every time speedsolving method, we really can't say we've maxed out potential

Fixed.

 

[Johannes91]

Single times are meaningless. Even if we had a good definition of "lucky" (we don't), it'd just change from getting the easiest solve possible to getting the easiest "non-lucky" solve possible, which would still require luck. If you want to know how long a normal solve takes, just use averages instead.

I remember saying that ~40 turns with ~5tps seems realistic to me, so sub-8 overall average. I still agree with that and guess someone will achieve it in a few years.

@Cyrok215: The optimal average is ~17.7 moves, not 22. http://kociemba.org/performance.htm

 

[Kolraz]

I think there is going to be a big improvement in averages, but there will come a point where single solves can't get much faster, without the production of new methods.

 

[piemaster]

Maybe we will invent a brain shock machine giving you every cube knowledge ever invented so that you can solve a cube in 20 turns or less with intuition in the future. Think big people, you guys aren't thinking outwards. Sub 4 or 5 are the basics.

 

[fanwuq]

Eric's PB is something like 5.05, and I'm certain that new methods and ideas will keep coming. Sub-5 doesn't seem impossible.

But I don't know how long we can keep making ideas. until there's a 30 move every time speedsolving method, we really can't say we've maxed out potential

Ever practiced linear FMC? It's not that easy. I usually get low 40's in about 5 min. Nobody is a master of it. Johannes is the best that I'm aware of and I don't think sub-30 is easy for him in 10 min. It's basically impossible in a few seconds unless you skipped quite a few steps.
Wlith ZBLL and good blocks or multislotting for F2L, Johannes's prediction might be possible in a decade or so.
If you want to talk science fiction, why don't you just make a time machine?

Edit: move this to off topic?

 

[IamWEB]

We can't determine how far it will go, because it will really depend on how many algorithms you know, and good you are at recognizing cases and applying the algorithms.

If someone learned All LL algs for Fridrich, then of course they could be faster than Harris Chan. But who has gone through the trouble to learn SO many algs?

 

[4Chan]

Eric's PB is something like 5.05, and I'm certain that new methods and ideas will keep coming. Sub-5 doesn't seem impossible.

But I don't know how long we can keep making ideas. until there's a 30 move every time speedsolving method, we really can't say we've maxed out potential

Ever practiced linear FMC? It's not that easy. I usually get low 40's in about 5 min. Nobody is a master of it. Johannes is the best that I'm aware of and I don't think sub-30 is easy for him in 10 min. It's basically impossible in a few seconds unless you skipped quite a few steps.
Wlith ZBLL and good blocks or multislotting for F2L, Johannes's prediction might be possible in a decade or so.
If you want to talk science fiction, why don't you just make a time machine?

Edit: move this to off topic?

Hmmmm, YES.

ZB is the future. It would take a long time to master obviously, but with practice.... you wouldnt even need to AUF and recognize almost as fast as PLL.

I had a J perm/T orientation, and thats a pretty fast ZBLL, very easy to recognize too. (all the colors on the sides are correct except 2 edges, and a block of color to the right.) I had a 2-3 second LL, and even though thats not too impressive (some people do that with fridrich), it made me confident that ZB is the future of cubing.

 

[fundash]

Sorry for being such a n00b, but...what is fmc?

and so you don't have to:
:fp at myself

 

[GermanCube]

Fewest Move Contest = FMC

 

[fundash]

thanks, i am so stupid, again...:fp at myself

 

[Anthony]

If someone learned All LL algs for Fridrich, then of course they could be faster than Harris Chan. But who has gone through the trouble to learn SO many algs?

Do you mean all 1-Look Last Layer algorithms?

 

[Cyrok215]

If someone learned All LL algs for Fridrich, then of course they could be faster than Harris Chan. But who has gone through the trouble to learn SO many algs?

Do you mean all 1-Look Last Layer algorithms?

I'm pretty sure that's what he means....

but that doesn't mean that they would be faster, they might have a 30 seconds F2L... maybe you could make up L2L methods that are faster though...

Ex: you would get cross, then corners, then fix the last 2 layer in one algorithm.

@Johhaness91: I knew that but I heard somewhere for some reason we didn't have proof that it was 20 so it's just 22 for sure right now.

 

[elcarc]

really the only way to get much faster now is to memorize the shortest move solve for every single combination of colors (rumored to be in the quintillions), which is almost impossible. you also have to factor in the human limitations of how fast we can physically move. i think about 4 seconds would be the fastest time possible.

 

[trying-to-speedcube...]

Fewest Move Contest = FMC

Haha Lol!

It's Fewest Move Challenge actually

 

[IamWEB]

If someone learned All LL algs for Fridrich, then of course they could be faster than Harris Chan. But who has gone through the trouble to learn SO many algs?

Do you mean all 1-Look Last Layer algorithms?

I'm pretty sure that's what he means....

but that doesn't mean that they would be faster, they might have a 30 seconds F2L...

They could be faster.
Take Harris Chan now, and then have him learn all algs for 1-look LL. With good recognition and execution, he's faster.

So if someone's F2L was as fast as his, but the they could do the LL in one alg, they could be faster.

 

[Shazbut]

I have to say that I think that cubing will continue to grow until someone discovers Gods Algorithm, then the 3x3 will be pushed aside, and be a mini contest like the Magic, where it is always the same thing over, and over, and over.....

Then, I think that the focus will go on the bigger cubes, or maybe the Megaminx.