[SilentNote]

The problem is that there are underlying issues that complicate matters, which have mostly to do with a misunderstanding of what science is and how it works. If you want to address issues in a constructive and scientifically informed manner, then this is not the way to do it.

This is the second time I've seen you say something to this effect, so let's discuss this at a proper place instead of hijacking another thread shall we?

 

[Wyville]

This is the second time I've seen you say something to this effect, so let's discuss this at a proper place instead of hijacking another thread shall we?

Sure, I'd be happy to discuss science in a more general sense (eg methodology, etc). Where would you like to start?

 

[SilentNote]

The problem is that there are underlying issues that complicate matters, which have mostly to do with a misunderstanding of what science is and how it works. If you want to address issues in a constructive and scientifically informed manner, then this is not the way to do it.

Not implying anything of the sort. What I am saying is that what you do here is not science, it is Cargo Cult Science. The arguments presented here lack the depth of understanding, thoroughness, and integrity needed. The problem is that most people here are so convinced that they know how science works that they keep fooling themselves and never actually get to the good stuff (ie. genuine science).

Mind to quote a few examples of Cargo Cult Science? So far the responses I've gotten from this section has been cross verifiable with multiple sources, some research papers, some real world listening tests, and self administered ABX tests. So they seem consistent to me.

I am no scientists so I wouldn't know exactly how scientific inquiry has to be conducted, but the information available here is better than opinions so that's what I'm rolling with.

For example, Hi-Res audio. Sampling rate and bit depth both have nothing to do with audio resolution. While I have not done a scientific inquiry into this (is it even necessary?), the theory, audio demonstrations, listening tests all agree that Hi-Res audio doesn't improve audio resolution and detail. Sampling rate basically limits the highest frequency you can encode and reconstruct, and bit depth has to do with quantization noise. This perhaps may not be scientific to you, so maybe you can start here.

 

[buonassi]

And the more quantization noise and other distortions the lower the resolution

So in very apt terms, resolution is defined by the bit depth. Now, can you hear the increased resolution going from 16 bit to 24 bit? That's another debate.

Highest frequency able to be perfectly reproduced is determined by f*2. Shannon Nyquist theorem.

So, based on my theory, I would expect 24/44 to sound better than 16/176

And if you want to discuss time domain claims of why you may think 176 is better than 44 that's another interesting topic.

 

[SilentNote]

Ok so firstly, this thread is not to discuss bit depth or sampling rate, I'm only using it as an example for proper scientific inquiry, as Wyville has brought it up a couple of times how information in the sound science section is more akin to cult science. To discuss bit depth and sampling rate, you can do that here:

https://www.head-fi.org/threads/sample-rate-bit-depth-and-high-resolution-audio.912949/

Secondly, of course the example I raised is in the context of 16 bit to 24 bit, 44.1 kHz / 48 kHz to 96 kHz / 192 kHz. Going lower than 16 bit / 44.1 kHz will be out of context.

So, based on my theory, I would expect 24/44 to sound better than 16/176

So this is your opinion and so far, no justification, explanation and supporting evidence has been provided.

So unless I'm seriously mistaken, simply agreeing to this opinion without asking for any form of evidence, cannot be "address(ing) issues in a constructive and scientifically informed manner". And if I ask for evidence and he is unable to provide it, then we are back to square one, i.e. the typical debate from sound science section plays out.

 

[Wyville]

Mind to quote a few examples of Cargo Cult Science? So far the responses I've gotten from this section has been cross verifiable with multiple sources, some research papers, some real world listening tests, and self administered ABX tests. So they seem consistent to me.

I am no scientists so I wouldn't know exactly how scientific inquiry has to be conducted, but the information available here is better than opinions so that's what I'm rolling with.

For example, Hi-Res audio. Sampling rate and bit depth both have nothing to do with audio resolution. While I have not done a scientific inquiry into this (is it even necessary?), the theory, audio demonstrations, listening tests all agree that Hi-Res audio doesn't improve audio resolution and detail. Sampling rate basically limits the highest frequency you can encode and reconstruct, and bit depth has to do with quantization noise. This perhaps may not be scientific to you, so maybe you can start here.

It has to do with what I said in the FH7 thread:

"It has more to do with a lack of appreciation for just how difficult it is to construct a genuinely solid scientific experiment and the subsequent interpretation of its results. It is so incredibly easy to be absolutely 100% convinced of something based on a theory, only to find you fail at the first hurdle when you put it to the test in practice. Every scientist worth their degree has been through this and many (unfortunately) tend to forget this later in their career because they get overconfident. Really talented scientists are scary good at avoiding these things and it is something you can only learn by designing your own experiments, failing, trying again and then having your peers rip you another one because you still forgot x, y and z."

What I see is that a lot of people here are 100% convinced of certain things and quite militant when it comes to defending those "facts" despite the fact that I can recognise from a distance that it was never scrutinised carefully. Multiple sources might corroborate a certain opinion, but what about those sources that don't? What about those aspects that you might have ignored? Is the literature used comprehensive? Is it up to date? Is it reliable? Are you using it in the right context? Do you genuinely understand (note: understanding is different from knowledge) what you are talking about? These are not questions one person should ask of another, but questions they should ask of themselves, bending over backwards to interrogate their own opinion. Science is not about arguing right vs wrong, it is about exploring and letting nature do the talking. Once you get that experience you find that you need to be extremely careful of things that sound really consistent to you because nature always has a way of surprising you and your own knowledge is extremely finite.

To put it in a different way. People on online forums are often very confident in their own expertise. People at the bleeding edge of science often (secretly) suffer what is known as "imposter syndrome". Why do you think that is?

My aim for this post is to try and give you a sense of how different the mentality is between what I see here and the day-to-day reality in science at the highest level.

 

[Hifiearspeakers]

It has to do with what I said in the FH7 thread:

"It has more to do with a lack of appreciation for just how difficult it is to construct a genuinely solid scientific experiment and the subsequent interpretation of its results. It is so incredibly easy to be absolutely 100% convinced of something based on a theory, only to find you fail at the first hurdle when you put it to the test in practice. Every scientist worth their degree has been through this and many (unfortunately) tend to forget this later in their career because they get overconfident. Really talented scientists are scary good at avoiding these things and it is something you can only learn by designing your own experiments, failing, trying again and then having your peers rip you another one because you still forgot x, y and z."

What I see is that a lot of people here are 100% convinced of certain things and quite militant when it comes to defending those "facts" despite the fact that I can recognise from a distance that it was never scrutinised carefully. Multiple sources might corroborate a certain opinion, but what about those sources that don't? What about those aspects that you might have ignored? Is the literature used comprehensive? Is it up to date? Is it reliable? Are you using it in the right context? Do you genuinely understand (note: understanding is different from knowledge) what you are talking about? These are not questions one person should ask of another, but questions they should ask of themselves, bending over backwards to interrogate their own opinion. Science is not about arguing right vs wrong, it is about exploring and letting nature do the talking. Once you get that experience you find that you need to be extremely careful of things that sound really consistent to you because nature always has a way of surprising you and your own knowledge is extremely finite.

To put it in a different way. People on online forums are often very confident in their own expertise. People at the bleeding edge of science often (secretly) suffer what is known as "imposter syndrome". Why do you think that is?

My aim for this post is to try and give you a sense of how different the mentality is between what I see here and the day-to-day reality in science at the highest level.

And to add to that: even when the people in sound-science admit to measurable differences, they then immediately disqualify that data by saying its inaudible.

And any time a test is conducted that disputes their ideology, they say that test is flawed or an outlier, or too small of a sample size or insert anything you want here. But you better believe their tests are rock solid! Hilarious.

 

[taffy2207]

So this is your opinion and so far, no justification, explanation and supporting evidence has been provided.

So unless I'm seriously mistaken, simply agreeing to this opinion without asking for any form of evidence, cannot be "address(ing) issues in a constructive and scientifically informed manner". And if I ask for evidence and he is unable to provide it, then we are back to square one, i.e. the typical debate from sound science section plays out.

It's been beaten to death already :-

https://www.head-fi.org/threads/24bit-vs-16bit-the-myth-exploded.415361/

 

[Joe Bloggs]

"It has more to do with a lack of appreciation for just how difficult it is to construct a genuinely solid scientific experiment and the subsequent interpretation of its results. It is so incredibly easy to be absolutely 100% convinced of something based on a theory, only to find you fail at the first hurdle when you put it to the test in practice. Every scientist worth their degree has been through this and many (unfortunately) tend to forget this later in their career because they get overconfident. Really talented scientists are scary good at avoiding these things and it is something you can only learn by designing your own experiments, failing, trying again and then having your peers rip you another one because you still forgot x, y and z."

That's exactly the process researcher went through... in the early 20th century. Clever Hans the horse was what led researchers to conclude that double blind testing was the only way to go... in 1907.
https://en.wikipedia.org/wiki/Clever_Hans

The fact that such levels of care and rigor in designing and conducting experiments is simply IMPOSSIBLE to ask from your run of the mill audiophile is EXACTLY why we almost categorically reject any and all claimed listening results that fly in the face of that which was established by researchers who DID conduct experiments with such care and rigor--long long ago, when it was still in somebody's vested interest to conduct experiments like this!*

What we ask for is usually very simple... "incredible claim x must be backed by incredible proof" (especially before we start theorizing on why x happens).

*In this sense, you're right, it's almost like cargo cult science, unfortunately the best thing we have today for you guys. "They just don't do science like grandpa used to..." Sad.

 

[buonassi]

[

Sampling rate basically limits the highest frequency you can encode and reconstruct, and bit depth has to do with quantization noise. This perhaps may not be scientific to you, so maybe you can start here.

I'm confused by your opposition to my post. What did you mean by 'start here'? you didn't originally link to any other string.
Regardless, I seem to be off topic so I guess I'll drop out.

Sorry @Wyville , seems I'm not going to be able to contribute anything substantive here.

 

[SilentNote]

At the highest level of scientific practice and inquiry, they would have gotten the basics down. Such as controlling for variables and minimizing bias. They are at the bleeding edge, that's why there's so much more that is unknown to them. The questions they ask are things where solutions have not been found, understanding not developed. That does not mean that the basics are open for re-interpretation. For example, the law of gravity is not open for re-interpretation (in its specific context) even though newer science such as theory of relativity has been developed. Yes, space-time curvature can be used to explain gravity, but in the context of newtonian physics, there's no better model currently in place. You can't just dismiss the law of gravity because we don't fully understand relativity and quantum physics.

The questions and hypothesis posted in Head-Fi is by no means bleeding edge audio science. Most of it is pretty well studied concepts. Things like threshold of hearing, there's very little ambiguity in that concept. We've even developed a scale for audibility and volume of sound. The threshold for hearing, is intuitively designated as 0 dB. Of course this doesn't mean that anything under 0dB is inaudible, in certain circumstances -5 dB and even -10 dB could be potentially audible by the right person. But to assert that one can hear quantization noise at -100 dB from the main signal at 80 dB is denying the scientific studies that previous researches have established.

Paired that with the reluctance to do volume matched ABX test to minimize bias and constant reliance and insistence on subjective opinions and claims, how are those claim more credible than researches conducted years ago by credible scientists, whose results are used by audio engineers world-wide? Open to new ideas is not the same as blindly accepting every assertion that is absurd and inconsistent with past developed knowledge.

Even if the inquiry was cutting edge and knowledge not fully developed, more so significant factual and testing evidence need to be presented. The problem here is that most are not even willing to share their testing methodology and results. Can you imagine a scientist that says he has solved the unified theory of everything but will not share the methodology and results? That's what we are dealing with here, people who asserts conditions that are contradictory to developed knowledge, proceed to justify with subjective claims, and when asked to share their testing methodology and results they outright refuse, citing flaws in established scientific procedures such as ABX in eliminating bias. It's not about right or wrong, but if you want people to take your assertions seriously, it either has to be understood by existing knowledge, or that new evidence has to be presented, and allowing other people to reproduce said results.

A good investigator wants to be proven wrong, because it gives him the opportunity to learn and be that much surer that he's not spewing garbage. This is why scientific papers are written in such technical detail and presented publicly to be scrutinized. It's not about being right, like how many poster asserts here and gets mad when they their methodology for inquiry is criticized. While you claim that the Sound Science community are convinced about being 100% right, I see the contrary that the posters here are so held on to their perceived subjective truths that any criticism is akin to gaslighting them. That's not the right way to approach science, or any inquiry.

In short what I want to say is that the uncertainty of the cutting edge science does not invalidate existing studies and concepts that have already been developed. And most inquiry on Head-Fi is by no means cutting edge but well within the domain of developed audio science.

--------------

And to add to that: even when the people in sound-science admit to measurable differences, they then immediately disqualify that data by saying its inaudible.

And any time a test is conducted that disputes their ideology, they say that test is flawed or an outlier, or too small of a sample size or insert anything you want here. But you better believe their tests are rock solid! Hilarious.

It's not so much ideology as established knowledge. And it's often not our tests that we rely on, but the actual scientific research done over the years that has established a baseline knowledge needed to understand what is likely and what is unlikely.

If you are so certain of the audible differences, why are you so resistant to discovering potential bias or flaws in methodology? A good investigator will do everything possible to ensure what they are testing is indeed what they think they are testing, and all criticisms should be accepted and new methodology designed to ensure the results are sound, until all reasonable criticisms are eliminated.

 

[SilentNote]

[


I'm confused by your opposition to my post. What did you mean by 'start here'? you didn't originally link to any other string.
Regardless, I seem to be off topic so I guess I'll drop out.

Sorry @Wyville , seems I'm not going to be able to contribute anything substantive here.

... I'm only using it as an example for proper scientific inquiry, as Wyville has brought it up a couple of times how information in the sound science section is more akin to cult science. ...

Sorry perhaps I wasn't being clear, but I thought the quote in the original post would be sufficient for people to link back to the discussion in the other thread by clicking the up arrow by the poster's name.

----------

It's been beaten to death already :-

https://www.head-fi.org/threads/24bit-vs-16bit-the-myth-exploded.415361/

And perhaps you have not paid attention to the title of the thread or the nature of the thread, or did not comprehend the main point of my post.

 

[Wyville]

That's exactly the process researcher went through... in the early 20th century. Clever Hans the horse was what led researchers to conclude that double blind testing was the only way to go... in 1907.
https://en.wikipedia.org/wiki/Clever_Hans

The fact that such levels of care and rigor in designing and conducting experiments is simply IMPOSSIBLE to ask from your run of the mill audiophile is EXACTLY why we almost categorically reject any and all claimed listening results that fly in the face of that which was established by researchers who DID conduct experiments with such care and rigor--long long ago, when it was still in somebody's vested interest to conduct experiments like this!*

What we ask for is usually very simple... "incredible claim x must be backed by incredible proof" (especially before we start theorizing on why x happens).

*In this sense, you're right, it's almost like cargo cult science, unfortunately the best thing we have today for you guys. "They just don't do science like grandpa used to..." Sad.

It is a great example of the thoroughness needed in scientific enquiry and how unexpected things can have a profound influence. But that is not just the case for experiments conducted by scientists, it also holds for when and where you can use the results of such an experiment. My next response is connected to this...

In short what I want to say is that the uncertainty of the cutting edge science does not invalidate existing studies and concepts that have already been developed. And most inquiry on Head-Fi is by no means cutting edge but well within the domain of developed audio science.

...I understand what you mean, but it is not quite what I meant to point out. At any level there needs to be care that you do not take existing studies out of context and apply their results incorrectly or extrapolate them beyond reason. A lot of that does happen here with studies used in an haphazard manner, the main aim being to achieve a theoretical line of reasoning that seems consistent. My warning is that even if something seems consistent, even if the topic is fairly casual, it might still be incomplete or incorrect.

 

[Joe Bloggs]

It is a great example of the thoroughness needed in scientific enquiry and how unexpected things can have a profound influence. But that is not just the case for experiments conducted by scientists, it also holds for when and where you can use the results of such an experiment. My next response is connected to this...

...I understand what you mean, but it is not quite what I meant to point out. At any level there needs to be care that you do not take existing studies out of context and apply their results incorrectly or extrapolate them beyond reason. A lot of that does happen here with studies used in an haphazard manner, the main aim being to achieve a theoretical line of reasoning that seems consistent. My warning is that even if something seems consistent, even if the topic is fairly casual, it might still be incomplete or incorrect.

Can you give an example, of *what* research you think is being taken / stretched beyond its original context (presumably to dismiss a claim)?

 

[castleofargh]

I'm with @Wyville on this(I'm going to explain when and why don't panic^_^), but my main problem is that I do not know how we could change anything. most threads don't even reach the point where people can agree on a few axioms or known facts and use that as building blocks for a given topic. how do we discuss the significance of a sound variation when one side of the argument rejects both measurements and controlled tests? how do we diagnose the cause of an event when that event is not going to be properly documented and will usually stop at "trust me I know what happened", or "I know what I heard"? how do we even begin to define the audible impact of something to a person who doesn't believe in the very concept of hearing threshold and has no clear understanding of the magnitudes involved? or maybe even believes that the human ear is much more accurate and able to sense things than dedicated recording tools?
we are constantly confronted to people who are completely unqualified to even discuss a topic, let alone come bragging about a conclusion they reached based on a vague idea and some sighted anecdote. what are we expected to do beside obviously reject the conclusion and point out the logical fallacy, or total lack of supporting evidence, or how if the guy is right, he's not a human being? we cannot discuss science or behave following the scientific method when most posters don't know what it is, don't understand why it is a necessity to try and get closer to facts and accurate models, and don't understand that their subjective impressions of the world are not showing them accurate objective reality. one big problem here is that people come making empty claims. it's already bad enough, but when confronted, most have no intention or ability to provide data about what happened to them. they won't have documented anything and don't plan to ever do it. they only want us to take anything they claim at face value...
any of those issues would be a conversation stopper for any actual scientist trying to get somewhere on a given subject. so of course there isn't much science in this section. instead we see a lot of preaching and people who get mad because today, yet another guy saying he was a chess master, but he doesn't know the rules, doesn't have a chess board, doesn't plan to ever bring one, and ultimately never cared about chess at all. all he wanted was for people to believe that he was indeed a chess master because he said so.

but no matter what reason I find to explain what's going on, I agree about all the shortcomings of most regulars on this section(myself included, obviously), like how easily we accept empty statements so long as they agree with us or our general line of thought. or how we will not bother to properly demonstrate our views or wonder if we got there using some baseless belief we got at one point and never questioned. and all the nasty territorial wars where we do try to push anyone that isn't like us out of the section on the motive that within Head-fi, that section is the only place where accountability and controlled tests aren't banned or strongly frowned upon. that of course is a clear reason as to why both are systematically brought up. because in this sub section, we can, so we do. the end result is messy, angry, extremist because after discussing the same BS 500 times for years, we all end up taking more and more shortcuts that actual science and proper reasoning wouldn't allow. it's bad, but it's tiring not to take shortcuts. having to properly assess the level of the person we're talking to on that specific topic, so we can try to make some explanation he might understand(because that's where we are most of the time, forget an explanation that he might accept, just getting understood tends to be a long term project nowadays). some here clearly just go for the kill as a way to assess who they're talking to. that's not good, obviously. if we're not ready to have order and method on our side, we shouldn't participate in a section about the science of stuff.
but again, knowing and being able to stop ourselves, 2 vastly different matters. what I'm saying is that beside trying to change Headfi, and change people in it, I don't know how this section could possibly become more like its name.


as for science in general, we're not doing that. we're not researchers, we almost never have statistically significant samples, or experiments documented well enough that we can use them, interpret them correctly, maybe replicate them, find the potential flaws, try to fix them, etc. most people worry about being right instead of worrying about how to find out the truth. that's going to happen everywhere sadly. what we can do is take the existing data and try our best to stop people from misinterpreting it and jumping to conclusion that go way beyond the experiment itself. if we just did that well, we'd be pretty badass.... for the amateur audiophile microcosm.