Freestyle method

[mr6768]

Hi cubers ...
I've heard about some speedBLD solvers using a method called freestyle . I was wondering what's this method all about . generally what's the Idea ?!
and also I know a little bit about Commutators and i think some people use this for corners .
any Ideas would be appreciated ...
Thank you

 

[cmhardw]

I was going to link you to the Wiki page about Freestyle, but I was surprised by this. Basically freestyle is where you solve the position and orientation of pieces at the same time. You can use a structured method like Pochmann or M2/R2 to do so, or you can just "freely" perform setup moves and PLL algorithms to do this more organically. BH is a type of free style method where you pre-memorize every possible case, and solve always using 3 cycles. Try looking up Classic Pochmann, or M2, as well as BH for an idea of some types of freestyle methods. Most people just use setup turns and PLL algorithms or other algorithms to solve each case.

Chris

 

[deadalnix]

Usually, we use the term freestyle for method that orient and permut two piece at time. This isn't a method in itself, more a category of methods.

 

[cmhardw]

Hey Amaury, how come DIADEM is not mentioned here? Didn't you post a website about it recently?

Chris

 

[Weston]

The "freestyle" method I use is just to solve orientation and permutation of two pieces at a time by using 8 move commutators and setups + a few other algs for bad cases.
I don't use PLLs except for solving parity.

 

[Joker]

Are freestyle users generally faster than M2/R2 users? Just curious.

 

[Zane_C]

Yes.

 

[deadalnix]

Hey Amaury, how come DIADEM is not mentioned here? Didn't you post a website about it recently?

Chris

Yes, here it is : http://www.francocube.com/deadalnix/DIADEM_en.php

But I didn't wanted to advertise too much as long as this is not exactly the topic.

 

[cmhardw]

While this thread is here, and from rereading the OP this question is actually still on topic as far as I can tell.

What exactly do people consider to be freestyle? I think that most people would answer "An unstructured method whereby people solve position and orientation of either 1 or 2 pieces at a time, depending on the situation, using any means necessary (i.e. algs, commutators, double swaps, etc.)." But, my opinion is that although people think their method is unstructured, they are actually always solving similar situations in similar ways. So, in my mind, there is structure to all of people's varying versions of freestyle BLD. So a better wording might be:

"An structured method whereby people solve position and orientation of either 1 or 2 pieces at a time, depending on the situation, using any means necessary (i.e. algs, commutators, double swaps, etc.). Some particular variations of Freestyle may decide to chose one type of means (i.e. commutators, PLL algs with setups) and use this as the backbone for the method."

To me, this style definition makes freestyle a meta-method, not a single method. So, for example, Classic Pochmann, M2/R2, DIADEM, BH, speed optimizing algorithm lists for either corners or edges, these are all freestyle methods if we go by the above definition. They are simply variations on the main freestyle idea by choosing one particular means of solving, and making this means the backbone to the method.

This seems the most logical way to categorize methods, based on how people have been defining freestyle in the past. Do people agree? Am I a minority here, or do others feel roughly similar? Anything we can change for the better in terms of terminology? Does someone have a suggestion as to how we could define what freestyle is precisely, and how it fits into the scheme of other blindfolded methods besides the "meta-method" one I've proposed?

I was honestly a little dismayed by the fact that Freestyle BLD is a blank page on the wiki!

Chris

 

[Mike Hughey]

Freestyle can mean anything we want it to. I guess the question is: what do we want freestyle to mean? It's our jargon, we can make it mean anything we want to.

I wonder if the freestyle "kings", such as Ville, use floating buffers. That would distinguish those people from things like Classic Pochmann, M2/R2, DIADEM, and BH. Somehow a method doesn't seem all that free if you're locked into specific buffers. But I don't know if floating buffers should be included in a definition of freestyle or not. Should it?

The thing that got me thinking about floating buffers again was square-1 BLD; it seems like many small cycles are more common on square-1 than they are on regular cubes, so I'm thinking maybe it would be better to use floating buffers instead of a BH-like approach, which is what I have been going for so far.

 

[cmhardw]

Freestyle can mean anything we want it to. I guess the question is: what do we want freestyle to mean? It's our jargon, we can make it mean anything we want to.

Yes, I completely agree! And, what better time to come up with a set definition than now?

I wonder if the freestyle "kings", such as Ville, use floating buffers. That would distinguish those people from things like Classic Pochmann, M2/R2, DIADEM, and BH. Somehow a method doesn't seem all that free if you're locked into specific buffers. But I don't know if floating buffers should be included in a definition of freestyle or not. Should it?

I would argue that this still makes Freestyle the meta-method, it is just qualified as either "floating buffer freestlye," or "fixed buffer freestyle." Just because we have not used these terms before, doesn't mean we can't now.

The alternative, of course, is to come up with a different term for "methods that solve position of orientation of pieces at the same time." But, since we're already using freestyle in a context like this, it seems to make sense to use the same term. Agreed or no?

The thing that got me thinking about floating buffers again was square-1 BLD; it seems like many small cycles are more common on square-1 than they are on regular cubes, so I'm thinking maybe it would be better to use floating buffers instead of a BH-like approach, which is what I have been going for so far.

The plot thickens If you reorient as well (with x2 rotations), then you'll have the most ridiculously amazing SQ-1 BLD method ever conceived! (Not that you don't already )

Chris

 

[Mike Hughey]

If you reorient as well (with x2 rotations)...

That made me LOL.

Maybe...
No, it's just too ridiculous.

 

[Stefan]

My opinion: Freestyle is not a method but rather a lack thereof. I just don't like the term, I think there's always a better one, for example "3-cycles" if that's what is meant. I especially disagree with Chris' above "Basically freestyle is where you solve the position and orientation of pieces at the same time". If I orient and permute separately and make up how on the fly, how is that not freestyle?

 

[Kirjava]

Freestyle methods are traditionally considered to be non-orient first.

I agree that the meta-method classification suits freestyle quite well.

 

[cmhardw]

My opinion: Freestyle is not a method but rather a lack thereof. I just don't like the term, I think there's always a better one, for example "3-cycles" if that's what is meant. I especially disagree with Chris' above "Basically freestyle is where you solve the position and orientation of pieces at the same time".

I disagree slightly with this. I disagree with the idea that freestyle is a lack of method. I've heard this said before as well, but I think that people in fact do solve similar situations in similar ways every time they see it, regardless of whether or not they believe they have a set method. Perhaps a better way of of describing freestyle is a "loose" method that can be changed depending on the situation/grip-hand/memorization recall time/etc.? Although there are multiple options to solve each case, I'll bet that people will fall into patterns of using similar approaches in similar cases based on their style. This will eventually result in solving each individual case the same way, or one of two ways, every time they see it. I would still classify this loose association of case to solving approach as a fixed, purposeful method.

If I orient and permute separately and make up how on the fly, how is that not freestyle?

That is a very good point. I suppose freestyle is a qualifier for the method, meaning that one has a loose or organic flow through each of the steps. Similar to how I have defined it above, they have possibly multiple ways of solving some or most of the cases, rather than exactly 1 way. M2/R2, DIADEM, BH, Classic Pochmann all have exactly one way to solve every case, thus they are not freestyle. However, if some or most of my cases can be solved in various ways, then that should be freestyle?

Based on this, you could call your description of a loose orient first approach as "freestyle 3OP."

Freestyle methods are traditionally considered to be non-orient first.

I agree that the meta-method classification suits freestyle quite well.

I like the meta-method classification as well, but Stefan brings up a good point that "freestyle" is perhaps a poor choice of name for this meta-method. We need a good term for "methods that solve position and orientation of one or more pieces at a time." Perhaps direct solving? Seems logical.

Chris

 

[MatsBergsten]

"Sticker solving" as opposed to "piece solving"??

 

[Stefan]

"Sticker solving" as opposed to "piece solving"??

Opposed? Those seem equal.

 

[MatsBergsten]

Opposed? Those seem equal.

Piece for me is a corner piece or edge piece. Sticker for me is one side of one piece.
And I thought that came from your bld page (Or Joel's description of your "Classic" method).

 

[Stefan]

If that stems from me, then it means one specific sticker, for example the red sticker on the red/blue edge is not the same as the red sticker on the red/green edge. So when I solve one sticker, the whole piece gets solved. And of course if I solve one piece, then all stickers on it get solved. So "sticker solving" and "piece solving" are the same. Unless you think really only of stickers, not taking into account the others on the same piece. Like when speedcubing newbs start solving one side and not one layer. But I don't quite see what that has to do with orientation vs permutation.

You meant it as a reply to Chris asking for names to distinguish methods separating or not separating orientation and permutation, right? Which one would be which?

 

[qqwref]

I think in a BLD context people generally use "piece" to mean "permutation only". So piece cycling or piece solving would be just solving the permutation (doing the orientation earlier, probably).