[VNandor]

   
True with acoustic instruments too. Point?

I didn't know acoustic instruments could make sounds with the fundamental being above 22kHz.
 
I also didn't know there are acoustic instruments with more than negligible amount (in terms of power) of frequencies over 22kHz. I assumed the ~-50dBFS because that's what I've seen after plotting the spectrum of some high resolution recordings when I first heard about it years ago.
 

  Oversampling and the ability to produce ultrasonics are two very different things. And you might be assuming a bit much. Have you actually produced a sound and applied spectrum analysis? Might this not be a rather special case confined you that specific combination?

 My soundcard is set to 192kHz. The daw is set to192kHz. I rendered the project in 192kHz. Yes, I did. This is how I know there was a harmonic at 96kHz.

  I used 192kHz in the daw and my souncard operates 192kHz as well. The highest harmonic of the tone I created were at ~96kHz. (Again, using a digital synth).

 
 

  Assuming we are talking about he same thing...dynamic range = total variance between loud and soft. No, I don't agree, because if a synth were programmed for wider dynamic range as a result of velocity than a piano, for example, it would become unplayable. So possible, not practical, not done.

I referred to a piece of software as a "digital synth" (and apparently, I might be wrong on that). Instead of pushing the buttons of a keyboard, I can make an envelope for velocity.
 
 

 
I think it unlikely that the ability of a software synth to output ultrasonics actually tracks DAW sampling rate.  But it wouldn't matter if it did as all of that is inaudible and getting filtered off for release to the world.

The whole point is, what would happen if someone released electronic music 192kHz instead of 44.1kHz? Since there are software snyths that can go well above 22kHz and they are being used by music producers, there could be frequencies that were cut off due to the 44.1kHz distribution format instead of the 192kHz. One could deliberately make music where the waveform of the 44.1kHz vs. 192kHz would noticeably differ. Because the synths' bandwith isn't limited to 22kHz.
 
 
 

  2. Samplers and syths have restricted dynamic ranges and frequency ranges compared to many acoustic instruments or groups of acoustic instruments.

This is what provoked me to join to the discussion. After your post, I'm not entirely sure what counts as a synth but the softwares edm producers use are often not limited in terms of frequency and dynamic range compared to acoustic instruments.
 
Quote:

   
Secondly, with an actual synth, yes, theoretically you can create almost any amount of frequency range and a large dynamic range BUT, I was not responding to what is theoretically possible but the reality of the EDM marketplace. What EDM is designed for (large and loud but limited dynamic and frequency range PA systems), how synths are employed in this genre in practise and how EDM is mixed/produced.
 
G

Okay, to me it appeared as if you were responding to what is theoretically possible to do with synths. I agree that in practice, edm doesn't benefit from 192kHz/24bit because noone deliberately creates ultrasonics and edm is typically being brickwalled. However none of this have to do anything with the synths' capabilities which I just had to point out.

 

[gregorio]

  Okay, to me it appeared as if you were responding to what is theoretically possible to do with synths. I agree that in practice, edm doesn't benefit from 192kHz/24bit because noone deliberately creates ultrasonics and edm is typically being brickwalled. However none of this have to do anything with the synths' capabilities which I just had to point out.

 
Yes but what I posted was in response to this: "Still don't understand why you say EDM benefits the least from Hi-Res." So I was specifically responding to synth use in EDM. Don't worry about being confused about what is called a synth, the once obvious lines between synth, sampler and electronic keyboard starting getting blurred about 25 or so years ago and today the term "synth" is applied to pretty much any electronic instrument, even though most of them can't actually synthesize sound.
 
There's a massive gulf between what is theoretically possible and how many instruments are actually used in practice and this isn't a new thing, it started many decades ago with the electric guitar. 
 
G

 

[Dillan]

I hate the phrase "trust your ears".

 

[pinnahertz]

  I didn't know acoustic instruments could make sounds with the fundamental being above 22kHz.

I never said that. IF an acoustic "isntrument" made a sound with a fundamental above 22kHz it wouldn't be functioning as an instrument because every bit of that sound would be inaudible.

  I also didn't know there are acoustic instruments with more than negligible amount (in terms of power) of frequencies over 22kHz.

Any harmonics -20dB or lower below the fundamental have negligible power, but we should not discuss this in terms of power, it will be confusing.

  I assumed the ~-50dBFS because that's what I've seen after plotting the spectrum of some high resolution recordings when I first heard about it years ago.  

Ah yes, but "plotting the spectrum"...there's a sticky problem. Music is full of transients. The bandwidth of the analysis window affects the analysis response time. Narrower/higher resolution increases response time, and will artificially plot transient signals below their real value. That sort of analysis is not straight forward at all, and is usually very misleading.  If you don't do it right, you get the wrong answer.

   My soundcard is set to 192kHz. The daw is set to192kHz. I rendered the project in 192kHz. Yes, I did. This is how I know there was a harmonic at 96kHz.

Out of curiosity, how do you know this?

  The whole point is, what would happen if someone released electronic music 192kHz instead of 44.1kHz? Since there are software snyths that can go well above 22kHz and they are being used by music producers, there could be frequencies that were cut off due to the 44.1kHz distribution format instead of the 192kHz. One could deliberately make music where the waveform of the 44.1kHz vs. 192kHz would noticeably differ. Because the synths' bandwith isn't limited to 22kHz.

As I see it, the "whole point" is that if it isn't audible, it's not music. Music must be audible. It doesn't matter if an instrument produces sound above 22kHz or not.

  This is what provoked me to join to the discussion. After your post, I'm not entirely sure what counts as a synth but the softwares edm producers use are often not limited in terms of frequency and dynamic range compared to acoustic instruments.

I did a quick survey of 14 commercial sample libraries.  I found that, when stated at all, the sampling frequency is 44.1.  I found one at 48, and one at 96.  Most never state sampling rate, but if it were anything other than 44.1 or 48 it would be a marketing "advantage" to say so.  Since they don't, I have to assume they are at 44.1 or 48, with 44.1 being far and away the most common.  That places a hard limit on ultrasonic content that no acoustic instrument has.  While that would technically differentiate a sample/synth from an acoustic instrument, in practice it doesn't matter.  For acoustic sample libraries, the ones that spec it at all are at 24 bit depth.  One was 16 bit.  Most aren't stated, but again, if it's more than 16 bit it becomes a marketing point and would be stated so.  However, no sample would ever use much more 12 bits because these samples are acoustic recordings in acoustic spaces with acoustic noise floors, etc. 
 
Apart from soft synths (and I question even that), it doesn't seem that the typical hardware digital synth runs over 48kHz.  Analog synths of course likely have no HF limit.  The theoretical dynamic range of a synth is limited by practical playability.  So, again, we're arguing theory vs practice.  The piano is one of the more dynamic acoustic instruments, but typically sits in a DR of 30-40dB, from barely striking a not to full out bashing with all your might.  Average piano dynamics within music is more like 15dB.  So, again, you may be able to build a synth patch with 70dB of DR, but you couldn't actually play it and make music with it, so what's the point?  It'll never show up in a recording of music of any kind.

Okay, to me it appeared as if you were responding to what is theoretically possible to do with synths. I agree that in practice, edm doesn't benefit from 192kHz/24bit because noone deliberately creates ultrasonics and edm is typically being brickwalled. However none of this have to do anything with the synths' capabilities which I just had to point out.

I agree you did point out that some synths could theoretically have greater dynamic and frequency capabilities than most acoustic instruments. Please understand that, while theoretically true, it has no bearing on music creation or recording.

 

[Yuri Korzunov]

I hate the phrase "trust your ears".

If no other instruments or skills, why not?
 
However, it is not reason for claims about that it is right way.

 

[CarlosUnchained]

I hate the phrase "trust your ears".

 
 
Don't trust your ears, but trust others ears even less.

 

[Yuri Korzunov]

   
 
Don't trust your ears, but trust others ears even less.

I don't see any difference between ear's owners

 

[Argyris]

Trusting one's "ears" doesn't actually mean trusting one's hearing, but rather one's perception of sound, which is inextricably clouded by emotion and conscious and subconscious bias. Most commonly, if someone believes they will hear something (e.g. a difference between a cheap piece of gear and an expensive one), they will most likely "hear" it. Where people then get into trouble is they attribute what they "heard" to the acuity of their hearing, or perhaps the perceptiveness of their mind in general. They then conclude that they somehow have better ("golden") ears than those who can't hear what they (the golden eared audiophile) can, and that anybody who claims there is no difference just isn't blessed with their hearing or powers of perception. In reality, they may have simply gotten caught in a feedback loop wherein the result is always what they expect it to be because it's now important to them, on an emotional and identity basis, to be able to "hear" a "difference".
 
It's this emotional investment, not only in the stable of expensive gear many audiophiles acquire, but also in the notion that owning it and being able to detect minute differences between functionally identical items places them in elite company, that drives subjectivist audiophiles' general loathing of blind testing and sound science. The alternative would be to face the possibility that, no, they don't necessarily have better hearing or perception than others. And, even if they did, absent visual confirmation they might not be able to tell the difference between a $10,000 cable and a coat hanger any better than the average listener or objectivist could. A simple blind test would of course clear up the matter. However, since nobody can force them to take a blind test and prove that they can indeed hear the difference, the subjectivist is free to disparage the process and the scientific basis behind it, choosing instead to, as the phrase goes, trust their ears.
 
When it comes to minute differences, or audio gear or software of dubious value, I am wary of any variation of the "trust your ears" mantra. I don't trust my, or anyone else's, ears, because I know that they are easily fooled by a myriad of factors.

 

[Yuri Korzunov]

A simple blind test would of course clear up the matter.

 
Unfortunatelly, double blind test is not simple as seems at first sight.
It is big laboratory test with methodics, that demands for development accounting many infinitesimal details.

 

[gregorio]

  Unfortunatelly, double blind test is not simple as seems at first sight.
It is big laboratory test with methodics, that demands for development accounting many infinitesimal details.

 
That all depends on what is being tested and what level of confidence is required from the result. In many, if not most cases a simple blind test is sufficient. If we take as an example the McGurk Effect, what does a big, methodical, double blind laboratory test teach us that the simplest of blind tests doesn't?
 
G

 

[Yuri Korzunov]

what does a big, methodical, double blind laboratory test teach us that the simplest of blind tests doesn't?

 
Better "marketing" objective proofs

 

[gregorio]

  Better "marketing" objective proofs

 
? The McGurk Effect doesn't have any marketing.
 
G

 

[Yuri Korzunov]

? The McGurk Effect doesn't have any marketing.

Do I mentioned the effect?

 

[gregorio]

  Do I mentioned the effect?

 
No you didn't but that's because you mis-quoted me!!

 

[Yuri Korzunov]

We discussed pure audio double blind test before your post. Not audio-visual effect.