What's the fastest method for 3x3?

[Stefan]

Are you blind? Or just trolling?

All of those units are in distance/time (or some are nondimensionalized by dividing by some known speed at those conditions [sound, light]).

I bolded the part where you realized your mistake.

Either way, "one post" is not a speed, and you are fully aware.

Correct. Which is why I said it's not a measurement of time.

If it takes you "one post" to learn something, then the quantity of ONE POST has been READ for you to attain some understanding. There is no distance. There is no speed.

There is speed, and there certainly doesn't need to be distance. That's even clearer than time not being necessary. For example, "frames per second" or "megabytes per second" should sound familiar, and neither frames nor megabytes are distances.

my major is relevant to Newtonian Dynamics

That might explain the close-mindedness.

a vast quantity of members of the speedsolving community profess negative opinions about you

I doubt that, but feel free to demonstrate it. In any case, there are also plenty of people with negative opinions about for example democracy, so I'm not sure what you're trying to say here.

[more rhetoric]

...

When proven factually incorrect, you simply disagree.

The latest I remember, from three days ago:

Ah, ok, sorry. Indeed, those cases are probably inner-slice heavy and corners are (almost) solved most of the time, so I don't see it being any good for that, either.

Yeah.... I'm totally disagreeing there, and certainly not apologizing at all.

But rather, it assigns a TIME to the APPLICATION of the method.

I'd call it measuring the time of the application. And then we can take that time and assign/attribute it to the method. And apparently you mind that we use times to describe speeds, but that's just the normal way in speedcubing. When asked "how fast", saying something like "one cube per 5.55 seconds" would just be idiotic and worse than the perfectly legitimate "5.55 seconds" (or even just "5.55").

 

[uberCuber]

I bolded the part where you realized your mistake.

:rolleyes: Your statement that started that little argument was "Speed doesn't necessarily involve time." Both of those dimensionless units still involve time. Each of them is a ratio between the speed of the object in question and another (constant) speed. So in order to calculate the speed of an object using either of those units, you still need to know the speed, as in distance/time, of the object in question to perform that calculation. Speed always involves time, it is the definition of speed. Stop arguing about this point, you really are doing a great job of supporting mDiPalma's claim (and a claim I have seen made by others [note: not claiming "a vast number"]) that you post just for the sake of arguing, and continue arguing even after you have been shown to be wrong.

 

[GuRoux]

But there is a certain level of subatomic randomness that occurs at every infinitely small time step. Simply assuming that you start at the same state doesn't mean that the states will be the same after even 1 time step into the future.

You are right though; the times that you and your clone get will be nearly identical. To the accuracy of a timer, they will be the same, but in reality they will be slightly different. And that slight difference is what magnifies as you change environments, solver, solver background, and the numerous other legitimate factors that affect speedcubing times.

If you produce a good experiment, those problems will not be too significant. A large sample population should balance uncontrollable aspects such as personality, culture, intelligence, upbringing, etc. Solving enviornment will be controlled. I don't really see anything that would significantly sway the results.

How do you know that there is subatomic randomness. Could not this perception of randomness be determined; its randomness only a illusion. It seems to me that there is no way to prove or disprove that "random" particles are not determined. It's like flipping a coin. Heads or tails is not really determined by chance, but external factor. In fact, chance does not exist but simply refers to uncertainty of external factors.

 

[Stefan]

That is the method's speed (that number is a time, by the way, not a speed).

As just pointed out in my previous post, I consider it a legitimate and much better way to describe a speed. But if you insist, simply prepend the obvious "one cube per" (and btw don't forget to append "seconds").

What if Feliks breaks this world record at a future comp? Then by your definition, CFOP's speed would then be lower than before. But the speed of CFOP is 6.54!

No, it was 6.54.

Without changing the method at all, how could its speed possibly change? Answer: it wouldn't.

If I drop my cube, its (falling) speed also changes, even though the cube doesn't change.

The only thing changing is Feliks' proficiency with the method.

Or he got lucky.

If a method itself actually has a speed, then that speed is a property of the method. If you don't change anything about a method, one particular property of it can't spontaneously change.

No, the speed is changed by an outside force (Feliks), just like the speed of the falling cube is changed by an outside force (gravity).

3. Say this experiment was carried out for CFOP, and the average ended up being 7.00 flat, with the fastest person's avg100 being 6.66 (made-up numbers, no statistical calculations made here). By your third definition, the speed of CFOP is 7.00. That is the method's speed (that number is a time, by the way, not a speed). But the fastest person in that group of 100 actually averages faster than that. How is it possible to use a method and be faster than what the method's speed is? How can you say that "the speed of the CFOP method is 7.00" if there is actually a person averaging below 7.00 over a whole lot of solves?

Because we're talking about the speed of the method. As a whole. If you "drive your car at 60 mph", most if not all of the car is actually not going 60 mph. The atoms move relative to each other, making that impossible. And the current bottom part of the tire is going about 0 mph while the top part of the tire is going about 120 mph. Other big pieces also move inside the car, giving them different speeds. Still, we say the car is moving 60 mph, no?

Now just relate the cubing method to the car, the method's people to the car parts, and the individual solves the car's atoms, and I hope this will answer your question.

4. Okay, hold up. Your first and third definitions state that CFOP is faster than all other methods (though each definition states that CFOP has a different speed than the other). Your second definition states that Roux is faster than all other methods. Your fourth definition says that, for example, Heise is faster than both CFOP and Roux. But these statements are all mutually exclusive. It's also pretty much guaranteed that the kind of model proposed in your fifth definition would declare a different "Fastest" method than your fourth definition does, since an extremely low movecount method will inherently have bad grips (as compared to CFOP/Roux) and recognition. And if this sophisticated model did in fact declare the same "Fastest" method as any of your other four definitions, I strongly doubt it would also come up with the same specific number as did the other definition. Therefore, all five of your definitions for the speed of a method are mutually exclusive. So which, if any, is actually correct? (The answer of course, is none, since they aren't even giving speeds, but rather other quantities such as time or number of moves).

Which of the definitions are correct? All of them. Have you never seen different definitions for one thing? Is "eagle" an animal or a coin? An animal is not a coin and a coin is not an animal, so are those two definitions wrong?

It's just wrong to misuse the definitions, by saying stuff like "CFOP is fastest" instead of "CFOP is fastest as determined by ...".

What is your current speed, the speed at which you are moving? If you just sit in front of your computer, you can legitimately say "Zero mph", but at the same time I can legitimately think you're moving "about 67000 mph". When I take the Sun as reference. So you do have several speeds at the same time and I don't see a problem with it (though in this case I'd find it ok to leave out the explanation for the "zero mph" because that's the straight-forward perspective, while the "67000 mph" is unusual and thus the Sun reference should be mentioned).

 

[DeeDubb]

At least we all agree that hijacking threads is fun.

 

[Stefan]

:rolleyes: Your statement that started that little argument was "Speed doesn't necessarily involve time." Both of those dimensionless units still involve time.

I know. If they didn't, I would've pointed that out earlier, when I first replied to him showing that page.

But he didn't say "involve time" there. He made a very specific claim, that claim is wrong, as even shown by the page he had shown, and that's all I said there.

Speed always involves time, it is the definition of speed.

That's a definition of speed. Not the only one. Especially that wikipedia one he showed is not "the one and only", as its "a length divided by a time" for example doesn't fit "frames per second" or "megabytes per second", both of which wikipedia itself, on other pages, calls speeds.

[you] continue arguing even after you have been shown to be wrong.

Nah... no. That's your perspective, but not mine. I don't think I've been shown to be wrong. Or that I am wrong. I think it's you guys who are and have been shown wrong, so *I'm* wondering why *you* are continuing to argue.

 

[kcl]

Stefan, if you actually believe methods truly have speeds, then answer this:

Is roux faster than CFOP?

 

[Stefan]

Stefan, if you actually believe methods truly have speeds, then answer this:

Is roux faster than CFOP?

Depends on how you measure.

 

[kcl]

Depends on how you measure.

Please provide me with some examples on different ways of measurement in which one of the two methods I mentioned is superior.

 

[Stefan]

Hmm, I might change my mind about objecting to "time" always being involved in speed, but by using a broader definition of "time" than mDiPalma apparently has in mind (as he rejected number of moves as a time measurement). I do agree that there's always something "time-ish" involved. Saw some interesting things here and might think about this tomorrow... (if I don't come to my senses and do more important stuff).

Please provide me with some examples on different ways of measurement in which one of the two methods I mentioned is superior.

http://www.speedsolving.com/forum/s...method-for-3x3&p=967173&viewfull=1#post967173

 

[kcl]

Hmm, I might change my mind about objecting to "time" always being involved in speed, but by using a broader definition of "time" than mDiPalma apparently has in mind (as he rejected number of moves as a time measurement). I do agree that there's always something "time-ish" involved. Saw some interesting things here and might think about this tomorrow... (if I don't come to my senses and do more important stuff).



http://www.speedsolving.com/forum/s...method-for-3x3&p=967173&viewfull=1#post967173

Number of moves isn't a valid form of measurement if a more efficient solution cannot be performed as quickly.

 

[uberCuber]

As just pointed out in my previous post, I consider it a legitimate and much better way to describe a speed. But if you insist, simply prepend the obvious "one cube per" (and btw don't forget to append "seconds").

etc.

Post read and acknowledged. And I learned something new for the day, the fact that an eagle is a coin, interesting.

I want to try to clarify something here, because the definition of "a method's speed" you seem to be working with is very different than the one assumed when I and others make the statement "methods don't have speeds," and I believe this is why this misunderstanding/argument is even happening in the first place.

Generally (including this thread) whenever someone in the cubing community makes the statement "methods don't have speeds", it is in response to "which method is the fastest" or "is CFOP or Roux faster", or other questions of that sort, comparing two or more methods. Do you agree?

Now, I know you can't be positive of the motives and thoughts of other people, but do you agree that the following is a fair assumption? Usually when someone asks a question like "which of these methods is the best/fastest", they are looking/hoping for one answer. Right? Whenever we write "Methods don't have speeds," it is because someone is trying to obtain one end-all answer to their question. If there were in fact one "fastest" method, as they are looking for, then there must be a way of looking at the intrinsic speed of each method to make that one grand comparison that says "Method X is faster than Method Y." Not "X has been used to achieve faster times in competition than Y" or "X has been used to achieve faster times at home than Y," but simply "X is faster than Y."

Making this last statement isn't possible, as you've just agreed. And that's because a method does not have one intrinsic value that defines its speed. This is what is meant by the statement "Methods don't have speeds," not anything referring to the definition of "a method's speed" that you were using, as something that may change regularly. This definition of speed as something changing and dependent on circumstance (though correct) is incompatible with those "what's the fastest method" questions that expect one answer, so we assume the person asking the question has in their mind the former idea, that "a method has a certain intrinsic speed." So the statement "methods don't have speeds" really is correct for exactly the purpose for which it is intended - to knock down this idea of intrinsic speed in a method. Think of the sentence as having the word "intrinsic" implied in it. Since it is a safe assumption that that is how the person asking the question is thinking, the sentence should have its intended meaning to its intended recipient without explicitly spelling this all out.

_________________________________

About the speed-involving-time thing. The particular post of yours that I quoted probably wasn't the right one to choose. Anyway, can you please explain how number of moves is a measurement of time (or speed, whichever you like), considering that it is often the case in speedcubing that a longer alg might be faster to perform than a shorter alg that has an identical effect?


EDIT: I do apologize for this aggressive comment earlier. I was getting very hungry and didn't catch myself slipping into an uncharacteristically annoyed mood; I don't normally get truly annoyed at anything :/

you really are doing a great job of supporting mDiPalma's claim (and a claim I have seen made by others [note: not claiming "a vast number"]) that you post just for the sake of arguing, and continue arguing even after you have been shown to be wrong.

 

[porkynator]

I agree thatit is often hard to tell wether Stefan is being concise or just plain rude. I also think that speed involves time in any case.
But I agree that we can define a method's speed in some way; however, I don't find satisfactory any of the ways Stefan proposed.

Just because you don't see a way and did something useless doesn't mean there is no way.

Some method speed definitions:
- Fastest official average-of-5.
- Fastest average-of-100.
- Have 100 kids learn and practice the method for 10000 hours, then have each kid do 100 solves and define the method speed as an average of those 10000 solves.
- Average solve move count (like I did in my diploma thesis, though ideally using all possible scrambles).
- Average solve time using more sophisticated models (including grips like Gripper and recognition).

- depends on who uses the method
- depends on who uses the method
- This might work, but there is a problem: different methods have different learning curves, so you are just measuring a method's speed after 10000 hours of practice.
- (assuming fixed tps which involve time) doesn't measure speed in practice
- ?

 

[Renslay]

Just a quick note:

The "speed of a (computer) algorithm" is often defined as the number of basic steps (of a Turing machine evaulating the given algorithm, for example). See time complexity:
http://en.wikipedia.org/wiki/Time_complexity
Of course it is closely relevant to the actual run-time (in seconds, for example) of the algorithm, but mathematically, one can say an algorithm is considered "fast" without taking any actual time-measurements. In time complexity, "speed" are usually means functions of the input-size. For example, Quicksort with an average time complexity of O(n log n) is faster than insterting sort, which has O(n^2).
"Time complexity is commonly estimated by counting the number of elementary operations performed by the algorithm, where an elementary operation takes a fixed amount of time to perform. Thus the amount of time taken and the number of elementary operations performed by the algorithm differ by at most a constant factor."

 

[mark49152]

In time complexity, "speed" are usually means functions of the input-size. For example, Quicksort with an average time complexity of O(n log n) is faster than insterting sort, which has O(n^2).

Not necessarily. An O(n log n) algorithm may still take longer per element than an O(n^2) algorithm, meaning that it is "slower" for smaller input sizes. Or, it may have a larger constant time cost.

Which makes this a good example of why it doesn't make sense to think of move count as a measurement of intrinsic speed of a speedcubing method. By that logic, FMC would be the fastest method.

 

[GuRoux]

Post read and acknowledged. And I learned something new for the day, the fact that an eagle is a coin, interesting.

I want to try to clarify something here, because the definition of "a method's speed" you seem to be working with is very different than the one assumed when I and others make the statement "methods don't have speeds," and I believe this is why this misunderstanding/argument is even happening in the first place.

Generally (including this thread) whenever someone in the cubing community makes the statement "methods don't have speeds", it is in response to "which method is the fastest" or "is CFOP or Roux faster", or other questions of that sort, comparing two or more methods. Do you agree?

Now, I know you can't be positive of the motives and thoughts of other people, but do you agree that the following is a fair assumption? Usually when someone asks a question like "which of these methods is the best/fastest", they are looking/hoping for one answer. Right? Whenever we write "Methods don't have speeds," it is because someone is trying to obtain one end-all answer to their question. If there were in fact one "fastest" method, as they are looking for, then there must be a way of looking at the intrinsic speed of each method to make that one grand comparison that says "Method X is faster than Method Y." Not "X has been used to achieve faster times in competition than Y" or "X has been used to achieve faster times at home than Y," but simply "X is faster than Y."

Making this last statement isn't possible, as you've just agreed. And that's because a method does not have one intrinsic value that defines its speed. This is what is meant by the statement "Methods don't have speeds," not anything referring to the definition of "a method's speed" that you were using, as something that may change regularly. This definition of speed as something changing and dependent on circumstance (though correct) is incompatible with those "what's the fastest method" questions that expect one answer, so we assume the person asking the question has in their mind the former idea, that "a method has a certain intrinsic speed." So the statement "methods don't have speeds" really is correct for exactly the purpose for which it is intended - to knock down this idea of intrinsic speed in a method. Think of the sentence as having the word "intrinsic" implied in it. Since it is a safe assumption that that is how the person asking the question is thinking, the sentence should have its intended meaning to its intended recipient without explicitly spelling this all out.

_________________________________

About the speed-involving-time thing. The particular post of yours that I quoted probably wasn't the right one to choose. Anyway, can you please explain how number of moves is a measurement of time (or speed, whichever you like), considering that it is often the case in speedcubing that a longer alg might be faster to perform than a shorter alg that has an identical effect?


EDIT: I do apologize for this aggressive comment earlier. I was getting very hungry and didn't catch myself slipping into an uncharacteristically annoyed mood; I don't normally get truly annoyed at anything :/

To an extent one can identify which method is better. Almost 100% of the time, CFOP will be better than beginner's method for a person to achieve better times. It's only when you get to methods that are similar in potential like CFOP, Roux, ZZ, and Petrus that people start to loose certainty. It might be interesting to think about what proportion of people have their best method as CFOP, Roux, ZZ, or Petrus.

 

[Stefan]

I want to try to clarify something here

Great post, much appreciated, thanks. I agree with most of it, including that a method doesn't have one "intrinsic speed" (unless you for example explicitly include turning speed in the method to prevent your robot from locking up) and I know what people mean when they say "methods don't have speeds". It's just that, [post=967188]as stated earlier[/post], I don't like giving up and doing nothing if there is something that we can do. And I believe we *can* measure speeds of methods in useful ways. That's what I'm trying to promote. I think "methods don't have speeds" is correct the way it is meant, but I also think "methods do have speeds" is correct the way it is meant, and more interesting and useful.

Anyway, can you please explain how number of moves is a measurement of time (or speed, whichever you like), considering that it is often the case in speedcubing that a longer alg might be faster to perform than a shorter alg that has an identical effect?

Either you directly consider number of moves a measurement of time (more on that in a moment), or you incorporate a moves-to-time translation in your measurement/model. My fourth definition kinda does the former, my fifth definition does the latter. If you're not familiar with it, please read about [thread=4593]Gripper[/thread], which I referred to in my definition (you can see a version of Stachu in action, note both the code as well as the output on the bottom).

Not going to reply in detail to everyone else and everything individually anymore (takes way too many hours), but I do want to elaborate on considering move counts a measurement of "time". It begins with this question: What is time? From Wikipedia:

Time has long been a major subject of study in religion, philosophy, and science, but defining it in a manner applicable to all fields without circularity has consistently eluded scholars.

An operational definition of time [...] observing a certain number of repetitions of one or another standard cyclical event

I think that the moves in our solutions are such events, and that as such, counting their number is a form of measuring time. And no, this is not me trying to cherry-pick and abuse a definition. I honestly feel that for example "after one episode" is a valid description of how "quickly" (not even my word but the article's) these series got canceled, that "after three beers" is a valid description of how "fast" one gets drunk, that "five generations" is a valid description of how "long" they lived there, and that "in 30 moves" is a valid description of how fast one solved their cube. The episodes/beers/generations/moves are not like "seconds" and similar units, but they nonetheless honestly feel "time-ish" to me, and the statements where they were used honestly feel like describing speeds to me. Apparently the authors writing "quickly" and "fast" also thought so. How about you?

Another thing you're probably familiar with is car routes and their times/speeds. Do routes have times/speeds? This is pretty analogous to cubing methods. So why is there no huge outcry because of Google showing times for routes, how fast you can get from A to B on them? Why is Google showing that, if it's a wrong and bad thing to do?

 

[Phillip1847]

Another thing you're probably familiar with is car routes and their times/speeds. Do routes have times/speeds? This is extremely analogous to cubing methods. So why is there no huge outcry because of Google showing times for routes, how fast you can get from A to B on them? Why is Google showing that, if it's a wrong and bad thing to do?

People don't drive, generally, at drastically different speeds and ways.
People that know full CFOP can average 50 seconds, while faz may average 7.5 seconds. A difference by a factor of 6.67. If we take 67 mph(sorry for not metric) as the general "speed" for driving from point A to point B, there are not people who drive at 10 miles an hour on the same route. This is why routes can have speeds - people drive somewhat similarly. CFOP and Roux cannot have absolute speeds because people differ by a ridiculous amount, and you get faster as you know them more.

 

[CriticalCubing]

Fastest Method?
CFOP, ROUX, ZZ and PETRUS.
Pick which one you like and practice.
Practice makes a man perfect

 

[mark49152]

The route analogy is a good one. Sometimes a longer route on major roads (high move count and TPS) is quicker between the same A and B than a short route on minor roads (low move count but fewer TPS). And that depends of course on driving style.

As interesting as this debate has been, it's pretty obvious to me (and I would guess others) that when someone asks which is the fastest method they are not asking about intrinsic speed. They are asking which method has the characteristics that best lend themselves to that person achieving fast times. Which of course is more complicated than any single metric and also depends on characteristics of the cuber.