Fastest Method

[Zane_C]

Gradually dropping "freestyle" seems seems to be an easy way out of the confusion.

I personally use the term "freestyle" quite regularly, because as I previously pointed out; it's more convenient than saying:
"I use a combination of x and y, with a hint of z".

The problem with "freestyle" is its vagueness, if you're describing how someone solves the pieces, "freestyle" can be interpreted in a number of different ways.

 

[lucarubik]

I also call freestyle to a combination of methods, but how would I call algs like M' U M U M' U M U? I dont know

 

[riffz]

I've always considered freestyle to be any method that solves permutation and orientation of certain pieces at the same time, and that uses commutators/3-cycles to solve most of them.

I still like the term free-3-cycle more.

 

[aronpm]

There are some tricks that can be added to speed-optimal 3-cycles to make it even faster

 

[Stefan]

I'd be fine with either "non-orient first" or "non-redux". Of those two I personally like "non-redux" the best.

Ugh... I cringe. Those sound like "non-unnatural" to me, a double negation. I'd rather have a positive word instead. Or none, and just mark the redux methods instead. (I still have the opinion that "solving" is the natural thing to do, and "first orienting according to some arbitrary definition of orientation and then permuting" is artificial, unnatural). "Direct" is my clear favorite if you really need to speak of the class of these methods excluding others (though do you really ever need that?).

 

[Kirjava]

Yes, because clearly by this thread, it's obvious that we've established such a standard to define freestyle, correct?

No. This definition already existed.

Not sure how I can put this in the least offensive way, but you're being too presumptuous, let alone a bit ignorant of other's opinions.

I'm not being ignorant at all. I considered your opinions and think they are stupid.

 

[Erdos]

I choose correctness over simplicity. We already have criteria for it by the way - it's quite annoying that you'll ignore it for the sake of using the terms how you would like to instead of the general convention.

Also, if all methods are classified as freestyle the term becomes meaningless and redundant. You're being a bit stupid.

Yes, because by this thread, it's clear that we've established such a standard to define freestyle, correct? Not sure how I can put this in the least offensive way, but you're too presumptuous to assume that everyone agrees to, or even knows, your definition. You're being a bit stupid.

EDIT:

I'm not being ignorant at all. I considered your opinions and think they are stupid.

I'm not talking about ignorance in not accepting other's ideas (double negative but you get the point). I'm talking about the fact that you assume your definition has become standardized or even accepted by more than, say, 10 people.

 

[Kirjava]

lols, you deleted your post and made it again. see above for replies to the original version, see below for replies to the edit.

you're too presumptuous to assume that everyone agrees to, or even knows, your definition..

I'm talking about the fact that you assume your definition has become standardized or even accepted by more than, say, 10 people.

You're too ignorant to know that it's not *my* definition. It was a definition a group of cubers decided on. I'm not even sure if I was part of the discussion at the time. The fact that Chris knows exactly what I'm talking about implies that I'm not being too presumptuous at all.

You've been ignoring a few points I made too. I'd prefer you either countered them or acknowledged that I was correct.

 

[qqwref]

I've always imagined "freestyle" as a direct solving 3-cycle method where you don't deliberately try to follow any rules about which cycle algs to use. You do whatever you think of, and if you get the same cycle in multiple solves, you might not even do the same thing every time.

I think it would be normal for a sufficiently serious and dedicated blindsolver to eventually start applying rules to their cycles (like only using comms, or only using speed-optimal/move-optimal algs) but at this point they have moved onto a specific method. Freestyle is by definition unorganized and "free". Also, I think the reason 2-cycle methods don't have the same feel as freestyle is that everyone immediately decides or learns the best setup moves for each case (because there are only 24). So, nobody who does 2-cycle methods decently fast is really solving freely, but instead reading off their list of pieces and doing a very specific thing for each one.

I do like the term "direct solving", btw. It captures the concept of solving a piece into place immediately, without being derogatory or circumlocutory.

 

[Kirjava]

I've always imagined "freestyle" as a direct solving 3-cycle method where you don't deliberately try to follow any rules about which cycle algs to use. You do whatever you think of, and if you get the same cycle in multiple solves, you might not even do the same thing every time.

I have never heard of anyone who solves like this. You use the best/easiest alg you know for that case, or make one up if you don't have one. Eventually you always end up using the same algs for each situation.

I really do think BH should be classed as a freestyle method.

 

[cmhardw]

I have never heard of anyone who solves like this. You use the best/easiest alg you know for that case, or make one up if you don't have one. Eventually you always end up using the same algs for each situation.

If we go by Michael's definition of "freestyle" then it seems that people would start off using freestyle when they first begin to solve pieces directly (without using an orient-first reduction step). However, once they begin to fall into patterns of using the same or similar cycles for specific cases, then at that point what they do is no longer "freestyle" but a defined non-standard method (i.e. personalized to them).

This "feels" right to me. I can definitely agree with this.

I really do think BH should be classed as a freestyle method.

Personally I view BH more like Classic Pochmann, and I definitely do not consider what I do to be freestyle. In the same sense that Michael mentions that "freestyle" should contain some amount of freedom of decision, I feel BH is almost the exact opposite. For every cycle of any 3 pieces, there is, in my mind, a correct cycle choice, the HTM (or WTM for edges) move optimal commutator that solves those 3 pieces. I sometimes do not choose the "correct" algorithm for speed reasons, but the times when I purposefully do not choose the move optimal algorithm are predictable and follow a pattern.

I would consider BH to be classified the same as Classic Pochmann, the only difference being that BH usually solves 2 pieces at a time compared to Classic Pochmann usually solving 1 at a time (cycle breaks being the exception to these rules).

Maybe this is where the confusion lies, as I don't know what term I would apply to the method type where I consider BH and Classic Pochmann to be essentially the same type of method. Perhaps they both could be called direct solving methods of "braindead application of pre-memorized algorithms for all possible cycle situations that could occur." This method type would be further broken down into two sub-classes: 2-cycle approaches, and 3-cycle approaches.

That's my take on it at least.

 

[Kirjava]

Method classification is always difficult.

Really the term 'freestyle' has evolved over time. When it first appeared it just meant "not 3OP".

I believe it originally gained this term because of the heavy use of commutators in the early freestyle methods. Commutators being "just make stuff up" and "do w/e".

Freestyle doesn't have to be free, and it's not the worst misnaming cubing has to put up with.

 

[qqwref]

Eventually you always end up using the same algs for each situation.

Maybe; maybe not. I know that when I do Fridrich I sometimes use different algs for the same case (even OLLs/PLLs), for no real reason other than that I feel like doing a specific one. It could be the same for a BLD solver, especially when there are so many cases that a casual solver wouldn't easily end up going "oh, it's this specific case, I must do it this way".

Maybe this is where the confusion lies, as I don't know what term I would apply to the method type where I consider BH and Classic Pochmann to be essentially the same type of method. Perhaps they both could be called direct solving methods of "braindead application of pre-memorized algorithms for all possible cycle situations that could occur." This method type would be further broken down into two sub-classes: 2-cycle approaches, and 3-cycle approaches.

We can call it something like "fixed" as opposed to free. So Classic Pochmann is a fixed, direct 2-cycle approach. (Whereas freestyle is a free, direct 3-cycle approach, and 3OP is generally a fixed, orient-first 3-cycle approach.)

 

[Kirjava]

Maybe; maybe not. I know that when I do Fridrich I sometimes use different algs for the same case (even OLLs/PLLs), for no real reason other than that I feel like doing a specific one. It could be the same for a BLD solver, especially when there are so many cases that a casual solver wouldn't easily end up going "oh, it's this specific case, I must do it this way".

Sure, but why should this be a 'thing'? Most of the cases will be performed the same way.

 

[cmhardw]

We can call it something like "fixed" as opposed to free. So Classic Pochmann is a fixed, direct 2-cycle approach. (Whereas freestyle is a free, direct 3-cycle approach, and 3OP is generally a fixed, orient-first 3-cycle approach.)

Wow! I really like this!

In my mind BH would be a fixed, direct 3-cycle approach. It has in common with Classic Pochmann that both methods are fixed, direct solve methods.

Free-3-cycle or Freecycle would be a free, direct 3-cycle approach under this naming.

Honestly, I feel that the combined use of the three categories of:
1) "free" or "fixed"
2) "direct solve" or "orient-first/reduction"
3) "2-cycle" or "3-cycle"

cover most of the current BLD methods.

TuRBo would be a fixed, direct solve 3-cycle approach, so TuRBo is covered. M2/R2 would be fixed, direct solve 2-cycle approaches.

Are there any common methods I'm missing that can't be classified using the combination of these three categories? If people agree that this works, then I feel that a page should be created in the Wiki, as well as the naming for each of the current BLD methods. This would certainly help with some of the naming convention confusion for BLD methods.

Freestyle could then just mean "free" solving, with the other two parameters being left open to either "direct" or "orient first" as well as "2-cycle" or "3-cycle"

Sound good? Comments? Suggestions? Criticisms?

 

[Marcell]

Honestly, I feel that the combined use of the three categories of:
1) "free" or "fixed"
2) "direct solve" or "orient-first/reduction"
3) "2-cycle" or "3-cycle"
cover most of the current BLD methods.

I really like these! I too think that these cover the aspects of our current popular methods, and that's kind of how I'm thinking about them.

 

[Mike Hughey]

Chris, for wings, I still find myself sometimes using different move-optimal commutators for the same pieces on different solves. So move-optimal doesn't always mean fixed. It probably should always mean fixed, but if you're bad at it, which I still am with wings, it might be somewhat freestyle. Or would you consider someone who does BH commutators freestyle to not be doing BH, simply because they're coming up with them on the fly?

(An example: FDr -> LDf -> FUr might be done [D', f U f'] or [F' u' F, D]. Both perfectly valid BH commutators.)

I really need to work on this; wings are holding me back so much on big cubes.

 

[Erdos]

You're too ignorant to know that it's not *my* definition. It was a definition a group of cubers decided on. I'm not even sure if I was part of the discussion at the time.

I said "your" definition for simplicity's sake. Of course, there's a philosophical argument behind whether someone can even have his or her own definition since a definition's an idea. Who really has a definition? I wouldn't imagine you to jump to this argument, but whatever; to go over every technicality really defeats the purpose of the post to begin with.

The fact that Chris knows exactly what I'm talking about implies that I'm not being too presumptuous at all.

Look more closely at his post. There's no indication that he "knows exactly what you're talking about." He simply agrees with your definition (again, I'm using "your" instead of using an abstraction without a modifier), or more accurately, he agrees to "your" definition with slight modifications. That hardly means he knows exactly what you're talking about. And even if you take in only one person to account, that hardly counters your presumption that the definition was standard, or something that everyone knows.

You've been ignoring a few points I made too. I'd prefer you either countered them or acknowledged that I was correct.

I really wasn't going to address your entire post since it's pointless to get into an argument I really wasn't planning to.. But I guess since you called me on it, I'll address the only thing I didn't so far, and that's on "my" definition of freestyle. My point was to make it meaningless and redundant (redundant much?). That way, it'd eliminated all such vagueness and destroy the use of the term altogether, and I'm sure I'm not the only one in favor of abandoning the term in favor of clearer ones.

I'm doubtful that arguments across the internet can even make a difference, so this'll be my last post on the topic. Feel free to bash and have your last word on whatever you want. I'll try to be the mature one and stop this pointless digression.

 

[Kirjava]

I said "your" definition for simplicity's sake. Of course, there's a philosophical argument behind whether someone can even have his or her own definition since a definition's an idea. Who really has a definition? I wouldn't imagine you to jump to this argument, but whatever; to go over every technicality really defeats the purpose of the post to begin with.

You said that I assumed that my definition had become standardised.

I was using the standard definition.

The rest of your paragraph is flowery ********.

Look more closely at his post. There's no indication that he "knows exactly what you're talking about." He simply agrees with your definition (again, I'm using "your" instead of using an abstraction without a modifier), or more accurately, he agrees to "your" definition with slight modifications. That hardly means he knows exactly what you're talking about. And even if you take in only one person to account, that hardly counters your presumption that the definition was standard, or something that everyone knows.

Look at this quote from Chris;

"I remember a debate about this happening a few months ago"

This is exactly what I was talking about. All of my definitions and references come from this debate.

this'll be my last post on the topic. Feel free to bash and have your last word on whatever you want. I'll try to be the mature one and stop this pointless digression.

Grow the hell up.

 

[cmhardw]

Chris, for wings, I still find myself sometimes using different move-optimal commutators for the same pieces on different solves. So move-optimal doesn't always mean fixed. It probably should always mean fixed, but if you're bad at it, which I still am with wings, it might be somewhat freestyle. Or would you consider someone who does BH commutators freestyle to not be doing BH, simply because they're coming up with them on the fly?

(An example: FDr -> LDf -> FUr might be done [D', f U f'] or [F' u' F, D]. Both perfectly valid BH commutators.)

I really need to work on this; wings are holding me back so much on big cubes.

This is good, as this issue means we need to solidify what it means when a method is "fixed." I sometimes do this as well, and it can depend on how I am holding the cube at the time. For example, if the previous alg had me do a y' from the initial orientation I would definitely use the [D', f U f'] alg (or, since I'm already at a y' rotation from solved I would really be doing [D', r U r'] y).

If the previous alg had me do a y rotation from solved, then I would use the [F' u' F, D] alg (only, since I would already be at a y rotation from solved it would really be [L' u' L, D] y')

I would say that since both are move optimal, and you alternate between the two, that this method is still "fixed." Even if one was move optimal and the other was not, say D2 l U2 l' D l U2 l' D, as long as you alternate between those two it is still fixed.

If lots of algs have mixes between move optimal variations and non-move optimal variations, and you swap between the two on any given case, then I would argue that your method is becoming more of a "free, direct solve, 3-cycle method" and thus more "freestyle" than BH.

So for example, for the cycle you gave if I consistently alternated between the 9 move alg and the 8 move alg (and I did this on a number of other cycles as well), then that is more of a "free" solving approach in my opinion than a "fixed" solving approach.

As to you saying that "you're bad at wings," what makes you say that? If using an approach that's a bit more free makes more sense to you, then do that. If there's one thing I've learned, it's that free and fixed approaches seem to be a preference to the individual. I'm the kind of person that does better with fixed approach methods. Like Stefan, I think the "braindeadness" of a fixed approach is an advantage for ease of solving. For some people it seems that having a freer method allows for more innovation during the solve, which can lead to solves with better flow, and thus faster solves. Mike, from what you're saying it seems like your method is still fixed, or are you alternating a lot between different algs on many different cases?

I say "free, direct solving, 3-cycle" variations of BH are maybe not strictly BH. In my opinion they're a modified version, or variation of BH - simple as that. I personally use a "fixed, direct, 3-cycle" variation of BH - I do not use strictly BH myself. I often use 9 move algs with better flow in place of the move optimal 8 movers. I do this for all piece types, but I do it in a predictable way, and I don't alternate between the 9 mover with better flow and the 8 mover with less flow. I just use the 9 mover with better flow. So, this is still a "fixed" approach, in my opinion, but it's also still a variation of what could be defined as "strictly BH." That's what works for me, though.