The 20MM (20 Move Method)

[Calamity Strike]

It is widely known in the cubing community that, a cube, no matter what state it is in, can be solved in 20 moves or less. But it is also known, that this can only be achieved by a computer, well, I intend to change that. Now i know that this has been attempted multiple times by multiple people, and hasn't resulted in anything. But i believe we are overthinking this, i have found something very interesting, (look at the title and you'll see it) that i think may be something completely new, and could result in, if not understanding the method, then getting close to. However, since i am fairly new to cubing, i thought i might see if some more advanced cubers might want to help me with this project, anyone who is interested can add me on skype, where we have a group for this, or, just discuss it here.

 

[Mollerz]

If you really want this to be taken seriously you need to post everything you have currently done. If you haven't really come up with anything, then do so before asking for help.

 

[Sajwo]

What did you found?

 

[Calamity Strike]

What did you found?

I found that, rather than an algorithm, it is a variable method, meaning the method may change, but it is still the same concept. CFOP, for example, is an organized method, meaning you do the same thing everytime: cross, f2l OLL then PLL. But with the 20MM, you may do the steps differently, or in a different order.

 

[Goosly]

You should just peel the stickers off.

 

[Calamity Strike]

You should just peel the stickers off.

Takes to long

 

[Jaysammey777]

or take the pieces out and switch them. 20 pieces total = <20 moves everytime.

 

[Calamity Strike]

or take the pieces out and switch them. 20 pieces total = <20 moves everytime.

Do you know how hard it is to take a thunderclap apart? Or, put back together i should say.

 

[Sajwo]

I found that, rather than an algorithm, it is a variable method, meaning the method may change, but it is still the same concept. CFOP, for example, is an organized method, meaning you do the same thing everytime: cross, f2l OLL then PLL. But with the 20MM, you may do the steps differently, or in a different order.

You should abandon your researches at the start then. Optimal solution is nothing more than a random sequence of moves, you won't see the solution till the very last few moves. Try with ~10 moves scrambles and you will se

 

[DELToS]

It is widely known in the cubing community that, a cube, no matter what state it is in, can be solved in 20 moves or less. But it is also known, that this can only be achieved by a computer, well, I intend to change that. Now i know that this has been attempted multiple times by multiple people, and hasn't resulted in anything. But i believe we are overthinking this, i have found something very interesting, (look at the title and you'll see it) that i think may be something completely new, and could result in, if not understanding the method, then getting close to. However, since i am fairly new to cubing, i thought i might see if some more advanced cubers might want to help me with this project, anyone who is interested can add me on skype, where we have a group for this, or, just discuss it here.

I'm barely sub-20 on 3x3, I think that I would be considered intermediate, not advanced, but I'll still try to help, and I also have a skype

 

[guysensei1]

Do you have a method currently or do you want people to find a method with you?

 

[mark49152]

Is this for speedsolving or FMC?

 

[Siddharth]

Maybe both.

 

[RicardoRix]

So with FMC you have the current best cubing minds taking up to an hour to solve the cube, and the current world record average for this is 25 moves. I think you have a lot of people to convince.

 

[irontwig]

What you're looking for does not exist.

 

[Calamity Strike]

This is just to see if it can be done, in the future it may become a speedcubing method, but i honestly believe that cfop/roux will be faster. It would be very helpful for FMC however. And i don't want to be rude, but i made this thread for people interested in helping, not people interested in telling me it can't be done, so if you don't believe it can be done, fine, but please keep it to yourself. Just a heads up for anybody who doesn't know, you can send me a contact request on Skype by clicking on the Skype icon under my name, and choosing send a contact request.

 

[Ollie]

I wrote a proper answer, but it got deleted somehow.

First suggestion - read up on FMC methods. It is basically what you're describing.

You will find that the best solvers use a combination of techniques, depending on what the scramble looks like. If you break down the solution, it will probably have 'steps' but will not always look like a typical speedsolving method. The solution might use the inverse scramble, block-building, orienting edges, commutators etc.

But optimal solutions tend to not look like anything 'human'. You may struggle to see any sense in those solutions without a computer and a good knowledge of FMC methods, which is exactly what some of the most computer-savvy cubers are doing.

This is just to see if it can be done, in the future it may become a speedcubing method, but i honestly believe that cfop/roux will be faster. It would be very helpful for FMC however. And i don't want to be rude, but i made this thread for people interested in helping, not people interested in telling me it can't be done, so if you don't believe it can be done, fine, but please keep it to yourself. Just a heads up for anybody who doesn't know, you can send me a contact request on Skype by clicking on the Skype icon under my name, and choosing send a contact request.

You need to explain exactly what you mean. You're basically just calling for a new method, without a proper knowledge of previous methods for speedsolving and FMC, how they differ, what they're good for. You're saying that people can't do it but you have found something interesting and new, but you haven't actually said anything interesting or new.

You're saying that the steps of speedsolving methods don't change, but that a new method might exist where the steps aren't always the same, which is what the best FMC solvers do now.

You've just said "x exists, who wants to help me find it?"

 

[irontwig]

Look, people have a hard enough time to solve a 2x2x3 block optimally. It's not like plenty of pretty clever people have not tried to solve in as few moves as possible.

 

[shadowslice e]

This is just to see if it can be done, in the future it may become a speedcubing method, but i honestly believe that cfop/roux will be faster. It would be very helpful for FMC however. And i don't want to be rude, but i made this thread for people interested in helping, not people interested in telling me it can't be done, so if you don't believe it can be done, fine, but please keep it to yourself. Just a heads up for anybody who doesn't know, you can send me a contact request on Skype by clicking on the Skype icon under my name, and choosing send a contact request.

I would be interested in helping but I don't have Skype.

Exactly what *is* this "interesting thing" you have found?

Cause until then, *cough* Snyder *cough* vapourware *cough*

And what do you mean that people are overcomplicating things? We have people here whose job it is to simplify very complex things (like mathematics professors and software engineers) and people whose job it is to solve complex problems (again above) and a lot of them have been trying for years.

You should abandon your researches at the start then. Optimal solution is nothing more than a random sequence of moves, you won't see the solution till the very last few moves. Try with ~10 moves scrambles and you will se

This, however, I disagree with. The moves are not random and there are patterns you can find in them. This is the essential nature of groups within a fixed range.

 

[Calamity Strike]

I wrote a proper answer, but it got deleted somehow.

First suggestion - read up on FMC methods. It is basically what you're describing.

You will find that the best solvers use a combination of techniques, depending on what the scramble looks like. If you break down the solution, it will probably have 'steps' but will not always look like a typical speedsolving method. The solution might use the inverse scramble, block-building, orienting edges, commutators etc.

But optimal solutions tend to not look like anything 'human'. You may struggle to see any sense in those solutions without a computer and a good knowledge of FMC methods, which is exactly what some of the most computer-savvy cubers are doing.



You need to explain exactly what you mean. You're basically just calling for a new method, without a proper knowledge of previous methods for speedsolving and FMC, how they differ, what they're good for. You're saying that people can't do it but you have found something interesting and new, but you haven't actually said anything interesting or new. You've just said "x exists, who wants to help me find it?"

I'm not saying other people haven't found anything, obviously people have been able to solve the cube in 20 moves, and also, i have a good idea how the method works, but im not positive. I have come close to intuitively solving the cube within 30 moves just by observing how computers solve it. What im trying to do, is find a way to solve the cube in 20 moves, in a relatively short time period, just being able to pick up a cube, and being able to solve it in 20 moves, now your welcome to say this isn't possible, and im not sure it is, but based on what Ive observed, i believe it is. If your willing to help, i can walk you through it and show you, if your just going to say "what you've found doesn't mean anything", I'm not going to waste my time.