The Simfile Quality Debate

[stargroup100]

But it doesn't change the fact that an octave is going to be inherently more consonant than a minor second. This is fact and it is objective. You can build guidelines on top of that that get more and more subjective as they get more and more advanced and specific, which deals with a issue of preference, but it doesn't change the basic principles. In this case, the example of an octave being more consonant than a minor second.

[Reincarnate]

An octave may have all sorts of nuances and differences when compared to a minor second, but that has no bearing on an objective quality standard with respect to human perception. That's like trying to argue something like "This type of apple X will have more sugar than apple Y and therefore, based on this objective fact, I can extrapolate to an objective standard saying that everyone will like apple X." That's just a fallacy in itself.

[Reincarnate]

I mean, okay, using your own example:

Consider a comparison between Jaws (minor second) and Somewhere Over the Rainbow (perfect octave). Using your logic, EVERYONE will prefer Somewhere Over the Rainbow to Jaws. Correct?

You'll probably go "No, because there are all sorts of subjective overlays on top that change one's perception of the songs!"

My response to this is "that's the point." Even if one particular combination of sounds has a particular appeal over another in a vacuum, that doesn't mean they are objectively better. Trying to change, for instance, Jaws into a song that exemplifies the perfect octave over the minor second would result in a totally different song that lacks the same appeal.

[kaiten123]

Originally Posted by stargroup100

Never said fun factor was an inherent property of a chart.

yea, but the fact that it isnt means its not objective

Originally Posted by stargroup100

I could walk a mile and say that it felt really short, and another person could walk that same mile and say that it felt long. That doesn't change the length of the mile, it doesn't mean that our opinions are wrong, it just means that the length is one mile but might feel like something different.

this is exactly my point. all of your arguments about music/simfile quality involve how it is perceived (fun, etc.).

whether the mile is "long" or "short" is subjective because it is based on one's perception or experience, the fact that it is "one mile in length" is an objective fact as it is an inherent quality of the path walked.

in the same way, whether music/simfiles are "good", "bad", "fun", etc. is subjective since its base on one's perception or experience, though we can say objectively that "this simfle has its notes placed closer to amplitude peaks in the song" or any number of similar statements as they are inherent properties of the chart.

Originally Posted by stargroup100

In what universe does the notion of common sense fall into a subjectivity?

Subjectivity means that different people have different views on the subject, but the definition of common sense is a set a propositions and beliefs that most people agree is of sound judgment. How is this subjective?

common sense is subjective because different cultures and different time periods have different "common sense" and none have any objective claim over all the others to being "the 1 true common sense".

also, i was talking about trying to argue that something is true because it is common sense which is silly because common sense is based on nothing more than the views of a large number of people. there was a time when it was "common sense" that the earth was flat.

Originally Posted by stargroup100

The problem is, you still don't have a good idea of where to draw the line between subjective and objective.

if you had bothered to read any of my posts you'd know that objectivity is based on whether it is a property of the object being observed or the subject observing it.

example:
saying "file A has a BPM closer to that of the song than file B" is objectively true/false since it is about a property of the files/songs.
saying "file A is more enjoyable than file B" is subjectively true/false since it is dependent on the person playing.

[Reincarnate]

At any rate, my overall point here is that the best you can do is figure out the "clusters." What are the 4-5 different types of files that people like to play? Based on that, we judge quality based on the standards of each cluster.

In other words, it's like asking "What makes a quality song?" The answer will depend. A person who listens to only techno music will have a very answer from someone who likes classical and will have even yet a different answer from someone who likes rock or pop or death metal.

Across all clusters, though, I'd say the steps need to make sense. If you're throwing out weird patterns to a song as if you hadn't even used the music as a guide, I think it's safe to say most people wouldn't like it. In other words, I agree that structure is pretty important, but to me it's almost a self-evident tautology. The whole point of a simfile is that it's a bundle of both a song and stepchart. The two obviously need to coincide with each other in at least having the steps FIT to something from the music. It's like asking what makes a good sandwich by saying "Well, I think whatever is between the two pieces of bread should be edible and go with the bread." Similarly, the file that goes with the music needs to be playable/sensible.

Past that base level of having a playable file with steps that are fit to the music in some way, it all depends on which cluster you're making your file for.

[MarioNintendo]

Wow. I think you guys are trying to dive deeper than the subject can let you go! I've tried to follow the conversation, but lost track around page 2.

You seem to be looking for the constraints that would make a universally good chart. To me, that's like asking yourself what you need to implement in a song in order to make it succesful. How can anyone know the answer to those questions if it's all based on subjectivity rather than pure reason?

[Reincarnate]

Originally Posted by MarioNintendo

Wow. I think you guys are trying to dive deeper than the subject can let you go! I've tried to follow the conversation, but lost track around page 2.

You seem to be looking for the constraints that would make a universally good chart. To me, that's like asking yourself what you need to implement in a song in order to make it succesful. How can anyone know the answer to those questions if it's all based on subjectivity rather than pure reason?

Because there are typically a finite number of clustered groups when it comes to segmentation analysis. That makes it much easier to cater to their tastes.

The question can't really proceed until we identify those clusters, because then it's all speculation. Unless of course, we wish to speculate. I think the main clusters would be split between pad and keyboard files, but the subclusters would be harder to define.

Please watch this video so you get a rough idea where I am coming from, because this is how the food industry was transformed and this same logic applies here. I know it's 18 minutes long, but there are a lot of interesting insights in here (mainly in the second half, as the first half is pretty obvious to most of us).

[stargroup100]

Subclusters are not difficult to define among keyboard charts. If I wanted to take the time I could do a very heavy breakdown. But I won't because I don't see a need for it.

Another misconception is that people keep thinking that we're looking for a universally fun chart. No such chart will probably ever exist, but that doesn't mean you can't come close. The points that I have mentioned simply describe what makes certain works in different mediums popular. The concept of subjectivity and objectivity are simply used to help describe this. In no way am I saying that you can use objective facts about making charts to make the perfect chart that everyone likes.

[SKG_Scintill]

I love how the commercial at the end is completely against what had just been said

[Reincarnate]

Originally Posted by stargroup100

Subclusters are not difficult to define among keyboard charts. If I wanted to take the time I could do a very heavy breakdown. But I won't because I don't see a need for it.

Another misconception is that people keep thinking that we're looking for a universally fun chart. No such chart will probably ever exist, but that doesn't mean you can't come close. The points that I have mentioned simply describe what makes certain works in different mediums popular. The concept of subjectivity and objectivity are simply used to help describe this. In no way am I saying that you can use objective facts about making charts to make the perfect chart that everyone likes.

The point though is that you probably won't be able to "come close." Preference variance is simply too wide. You can try to please the most amount of people with one chart, but it won't be nearly as optimal as what you could do by making a few charts that cater to completely different tastes.

To this extent, subcluster breakdowns are vital and there is most definitely a need for it. Again, please watch the video if you still disagree.

[stargroup100]

-___-

I'm not trying to get close to the perfect chart. My point was the quality and subjectivity both exist in chart making, it's not 100% subjective.

[Reincarnate]

Originally Posted by stargroup100

No such chart will probably ever exist, but that doesn't mean you can't come close.

I was responding to this. But more broadly, I'm arguing that you *do* need to identify your clusters if you want a better shot at creating the best set of simfiles.

[Arch0wl]

You watched that TED Talk too? I've seen Gladwell's talk like, four times. I even bugged my girlfriend into watching it. It's ****ing great.

[Reincarnate]

Yeah, TED talks are pretty sweet.

Completely random aside about Gladwell irrelevant to this discussion:

I don't know if you've ever read Outliers (cool book by Gladwell, even if broad in generalizations and not very rigorous/academic), but at one point he discusses the effects of luck + hard work etc, and mentions the "Smartest Man In America" -- a man with an IQ over 200 who was held back because of lack of support.

If that alone doesn't ring warning bells, I should add the finishing touch, here: I've actually *debated* him before. He's a religious nutcase who's very ignorant of science, abusive of mathematics, untrained in jargon, unnecessarily verbose, and downright condescending.

[~kitty~]

Well if the simfile is on sync, then I think it's mostly based on opinion and a person's playing style. I've recently tried changing to spread, and I enjoy different simfiles now due to that change. I don't have a lot to say because I barely even understand the subject at hand, I just wanted to toss in my opinion.

[stargroup100]

Originally Posted by ~kitty~

Well if the simfile is on sync, then I think it's mostly based on opinion and a person's playing style. I've recently tried changing to spread, and I enjoy different simfiles now due to that change. I don't have a lot to say because I barely even understand the subject at hand, I just wanted to toss in my opinion.

Good point. This is a completely different issue though. When you change your playing style, you are essentially playing a different game. A pattern feels different between two playstyles, whether it's something minor like switching from index to spread or something major like DDR to Guitar Hero. In the case of different playstyles, it's less of an issue of personal taste and more about how the two playstyles themselves differ from each other.

[Reincarnate]

Some files are easier to play with certain playstyles over others (FOTBB is loads easier on index, for example, whereas I have no idea how you'd play, say, Pants on index very effectively). I agree that if a file is very biased towards one playstyle, then someone who plays with a different style is going to view your file very differently.

[Kilroy_x]

At this point I am addicted to most anything which has a high level of dissonance while still having an internal logic.