Master Thread: Are measurements Everything or Nothing?

[virtua]

I think we are addressing that question here about 5 times a day.

IMO the primary origin is that the average listener/reviewer/salesperson doesn't understand that sighted listening creates impressions of sound waves that are dominated by non-sonic factors. They naturally and wrongly assume that the differences they are 'hearing' must be in the sound waves. It all flows from that.

I've been here for a while and for the most part I see people directly addressing other peoples audibility claims, but what I hadn't seen as much is the where, how and why they came to such conclusions and how they were influenced so in the first place other than 'you're experiencing placebo', which of course is absolutely true. These ideas have to come from somewhere (the numerous non-sonic factors you mentioned in brief), and if it was already addressed 5 times a day, then I must not have been browsing the correct forums and for that I apologise for increasing the noise in this forum. If it's been addressed like this elsewhere I would definitely like to have a read of the posts if any come to mind.

 

[Blumlein 88]

I've been here for a while and for the most part I see people directly addressing other peoples audibility claims, but what I hadn't seen as much is the where, how and why they came to such conclusions and how they were influenced so in the first place other than 'you're experiencing placebo', which of course is absolutely true. These ideas have to come from somewhere (the numerous non-sonic factors you mentioned in brief), and if it was already addressed 5 times a day, then I must not have been browsing the correct forums and for that I apologise for increasing the noise in this forum. If it's been addressed like this elsewhere I would definitely like to have a read of the posts if any come to mind.

It is as simple as, "I have these two amplifiers, let us listen to them and see which one is better". That is all that is needed and many people will hear one of them as better. Even more so if there is a volume mismatch. Like one amp is more sensitive at the input and plays just a little louder. It really does sound better, but not for any reason that matters other than level. Add in someone who already believes these things and just hints them to other people. People then convince themselves and then convince others down the road. Once you start down that road it can snowball in a hurry. Not everyone will be much effected and some will balk at big prices on some items. But enough, a sizable minority even is enough to fuel an industry that preys on this. Plus in the early days when some firms were smaller, the guys really believed it themselves.

Does this make sense to you?

 

[312elements]

I agree. I think many objectivist are very concerned about differences that they cannot hear or at least cannot hear without dedicated, critical listening. There is no way to prove this of course, I have formed my opinion on myself and several of real life aquintances over the years. So it's kind of useless to suggest that maybe there is no practical difference between certain very similar studio monitors from two different manufacturers: this just gets the same "you have cloth ears" argument that you get from subjectivists about power cables.

Why should dedicated critical listening be a dismissible requirement? In most things the difference between the best and very good is a thin margin.

 

[312elements]

Just do it directly with the audio files and listen. Something like DiffMaker or Deltawave.

Why not just look at the frequency responses of the two files?

My curiosity has little to do with frequency response. I understand how that works. What I don’t understand or what I’m interested in learning more about is the relationship between multiple tones being played simultaneously.

 

[bodhi]

Why should dedicated critical listening be a dismissible requirement? In most things the difference between the best and very good is a thin margin.

By dedicated critical listening I ment the process of looping short passages over and over again to find out the difference. This is fine if you are interested in perfection for perfections sake, but it's far from the more typical use case of just listening to music/watching a movie.

I just feel it needs to be communicated more clearly when differences in measurements discussed probably are not easily noticeable even though they are very interesting to people that also value objective performance itself.

 

[312elements]

By dedicated critical listening I ment the process of looping short passages over and over again to find out the difference. This is fine if you are interested in perfection for perfections sake, but it's far from the more typical use case of just listening to music/watching a movie.

I just feel it needs to be communicated more clearly when differences in measurements discussed probably are not easily noticeable even though they are very interesting to people that also value objective performance itself.

I appreciate your sharing that. I think this is often where disagreements stem from and I think from the way you crafted your response you might even agree. It’s one thing to say something is the placebo effect and it’s an entirely different thing to say the differences are more or less too small to care about. In one case we’re saying that what you’re hearing doesn’t exist and in the other we’re saying that what you’re hearing is of little importance to the majority of folks and doesn’t really matter.

 

[Blumlein 88]

My curiosity has little to do with frequency response. I understand how that works. What I don’t understand or what I’m interested in learning more about is the relationship between multiple tones being played simultaneously.

Deltawave will derive FR difference from music files without using any test tones. If two files have FR differences then that is an obvious reason they sound different.

As far as multiple tones go the same fellow has written software called Multitone. Guess what it tests?

And btw, do you understand according to Fourier any waveform can be broken down into a combination of sine waves no matter how complex the waveform?

 

[312elements]

Deltawave will derive FR difference from music files without using any test tones. If two files have FR differences then that is an obvious reason they sound different.

As far as multiple tones go the same fellow has written software called Multitone. Guess what it tests?

And btw, do you understand according to Fourier any waveform can be broken down into a combination of sine waves no matter how complex the waveform?

I’m slowly soaking that in. It makes sense but when we factor in things like Doppler distortion I can also imagine that some speakers with a similar frequency response could potentially sound different. Doppler distortion just being one example of what could happen. The more I think about it the less I’m inclined to believe that Fourier will satisfy my curiosity, but it’s been an excellent resource in getting the neurons firing. Thanks!

 

[Blumlein 88]

I’m slowly soaking that in. It makes sense but when we factor in things like Doppler distortion I can also imagine that some speakers with a similar frequency response could potentially sound different. Doppler distortion just being one example of what could happen. The more I think about it the less I’m inclined to believe that Fourier will satisfy my curiosity, but it’s been an excellent resource in getting the neurons firing. Thanks!

Fourier is accurate regarding how this works. For instance one wave with Doppler distortion is not equivalent to one without.

 

[IPunchCholla]

I’m slowly soaking that in. It makes sense but when we factor in things like Doppler distortion I can also imagine that some speakers with a similar frequency response could potentially sound different. Doppler distortion just being one example of what could happen. The more I think about it the less I’m inclined to believe that Fourier will satisfy my curiosity, but it’s been an excellent resource in getting the neurons firing. Thanks!

https://pubs.aip.org/asa/jasa/article/68/6/1561/777588/The-audibility-of-Doppler-distortion-in

 

[312elements]

https://pubs.aip.org/asa/jasa/article/68/6/1561/777588/The-audibility-of-Doppler-distortion-in

thanks for sharing. That was interesting.

 

[krabapple]

I agree. I think many objectivist are very concerned about differences that they cannot hear or at least cannot hear without dedicated, critical listening.

Again: believing that any numerical difference , no matter how imperceptible, makes A 'better' than B, is not 'objective' thinking. It is totally subjective 'feeling'. And that's fine; just acknowledge it for what it is. And don't try to build any claims about A or B's 'sound' from it.

 

[danadam]

For what it's worth, I'll never subject myself to a blind test ever again. It's just not fun to try to listen to minute differences. And it's wasted time when there are blatant differences.

That's how we end up with "night and day" differences between digital cables

 

[krabapple]

You are completely wrong.

I have stated very often that to hear a repeatable difference between lossless and a high quality MP3 you need (a) a track you are familiar with (b) familiarity with the patterns to listen to and (c) a track that ideally features something like castanets. The MP3 psychoacoustic model will often try to seduce you with a slight focus on hearable high and low notes, and many people doing blind testing fall into that trap and declared the compressed track their preference. It's been shown in hundreds of blind testing tests on the internet. And measurements support that every time.

I don't entirely disagree, though I wouldn't phrase it as 'familiarity with the patterns to listen to' . I think you mean familiarity with the artefacts that are characteristic of lossy encoding.

My point about emotions is completely separate... if you don't listen to music to gain some emotional input, just throw your awesome measuring $100K gear into the trash bin because it is not doing what it is supposed to do. In no way ever did I imply emotions triumph over measurements. I clearly stated (for those with reading comprehension) that emotion is the goal when listening to music. If you dispute that, I pity your engagement in this hobby. Personally, linearity and good measurements are a gateway to that.

Whatever the role of the emotions in the listener, they are a property of the listener. Not a property of the sound source. The divide between 'objective' and 'subjective' approach to audio is simply the willingness to acknowledge that. The default stance of a 'subjectivist' is to assume that their perceived or felt difference -- in preference, in emotional response, in 'warmth', etc -- must be due to a difference in the source. An 'objectivist' knows that it needn't be, at all.

It is a confusion of effects and causes that drives the 'debate'.

 

[312elements]

Fourier analysis is exactly the answer to your curiosity.

Please allow me to make a comment. You seem to be a person who has anxiety in regards to science. You're waffling, with doubts in your mind. On one hand, you seem to respect scientific principles, but on the other hand, you seem to retain residual ties to the Subjectivist ways of thinking.

You can't have it both ways. There is no middle ground. You don't pick and choose which one you wish to believe for one issue and which to believe for another issue. It doesn't work that way.

You trust science for medicine, for cell phones, for cars, for weather radar ........ relax and trust it for audio, too.

Jim

It's not that I don't trust the science as much as I question the interpretation of the science from time to time. That's all. For example @Blumlein 88 just shared that such distortion would be measurable and @IPunchCholla just shared that it wouldn't be audible in an average listening environment. I have no reason to believe that either of them are wrong but it creates a grey area. I personally put a lot more weight into measurements than I do double blind tests. I can think of scenarios where double blind tests are more valuable than others, but just because the majority of people do or don't notice something or because the majority of people agree on one thing but not another doesn't necessarily make it right. The majority of people get things wrong all of the time.

As far as relaxing... I'm not stressed about this. This is a hobby. It's a distraction. Dare I say it's even fun. As far as not being able to have my cake and eat it too. I don't know that it's necessarily fair to say that. Some of the science being shared often feels over simplified to make a point. As in someone has a point to make so they create or find a test to prove that point forsaking other potentially pertinent information. That doesn't necessarily mean that they're wrong or that they're trying to be deceptive, but it's also not the whole story. That creates skepticism. I'm just enjoying the journey. Not trying to ruffle any feathers.

 

[voodooless]

Please enlighten me where I was wrong in stating that people that go through ABX still dispute the outcome after the fact? It happens very often.

So what?

I am NOT stating that makes sense. I am just stating that the parrot call around "ABX exposes the TRUTH" clearly is NOT an established fact (and has seldom ever convinced a "subjectist").

I don’t see how this follows? Somebody disputing something doesn’t make that thing more or less a fact.

 

[krabapple]

Why should dedicated critical listening be a dismissible requirement? In most things the difference between the best and very good is a thin margin.

Truly 'dedicated' critical listening happens when only inherent properties of the sound are being compared.

That very rarely is what is happening when hobbyists do 'critical listening'. It requires what scientists call controls for confounding factors to be in place. It's not normal listening, but it can come close, relative to.....

....the thing where someone trying to hear, for example, the small diff between 320 mp3 and lossless, compares a tiny snippet of the audio featuring cymbals or castanets, over and over and over -- I call that 'forensic listening' or maybe 'microscopic listening'. It has nothing to do with normal listening. It it doesn't tell me anything about whether the listener hears 320 vs lossless during normal listening. Once the listener can pass a 320 vs lossless ABX without doing microscopic listening, then they've convinced me (they've probably trained themselves sufficiently to hear the artifact routinely).

 

[krabapple]

It's not that I don't trust the science as much as I question the interpretation of the science from time to time.

And yet you have already seen that it's possible for your 'questions' to amount simply to something you don't know, not something 'science' -- or its interpreter -- doesn't know.

Why don't you take that to heart?

 

[voodooless]

With “critical listening”, the critical part is done separate from the listening and usually involves eyes and other senses. They are two different things. That’s why it doesn’t work. You need only “listening”.

 

[312elements]

Truly 'dedicated' critical listening happens when only inherent properties of the sound are being compared.

That very rarely is what is happening when hobbyists do 'critical listening'. It requires what scientists call controls for confounding factors to be in place. It's not normal listening, but it can come close, relative to.....

....the thing where someone trying to hear, for example, the small diff between 320 mp3 and lossless, compares a tiny snippet of the audio featuring cymbals or castanets, over and over and over -- I call that 'forensic listening' or maybe 'microscopic listening'. It has nothing to do with normal listening. It it doesn't tell me anything about whether the listener hears 320 vs lossless during normal listening. Once the listener can pass a 320 vs lossless ABX without doing that, then they've convinced me (they've probably trained themselves sufficiently to hear the artifact routinely).

Your particular example isn't one that I find terribly interesting. That came out wrong. I agree with your assessment of the lossless debate. If I'm doing dedicated critical listening (sitting in the MLP, no laptop or phone to distract me. Just sitting and enjoying the music. I'm paying attention to the low level details in a way that I'm not regularly doing with music streaming to the Sonos in the kitchen while I'm cooking dinner. I like music as I assume most of us here do. I've almost always got music on, but the last hour or two of every night I spend listening in a way that's different than how I listen the rest of the day.