What uncontroversial audible differences cannot be measured?

[audiventory]

On 06.06.2017 at 6:36 PM, wgscott said:

Anyway, if there are well-known, established examples of readily audible but unmeasurable differences (that stand up to various reality checks like blind testing, etc), please list them.

 

Any really audible difference is measurable, because modern measurement tools are more sensitive than ears.

To measurement the difference need take proper tool.

[audiventory]

8 minutes ago, Jud said:

As has been pointed out wrt imaging, even there we have reliable measurements of what produces the phenomenon, as we do of what produces the stereo effect of an instrument or singer in a location that doesn't really exist between two separate speakers.

 

That's why I've gone to examples where humans reliably perceive things machines can't.  This occurs *only* in the perceptual realm, i.e., our heads.  If we want to exclude this - if it's not the sort of thing Bill was looking for - then, as noted before, I think we're done here.

 

Hi Jud,

 

Yes. Recognizing of object location is problematic.

Especially absolute.

If the location changed it can be detected an common difference of original and altered audio stuff.

But detect that it is audible effect, we can't. Without image (in common meaning) recognition.

[audiventory]

4 hours ago, Jud said:

OK, here's another: "Hearing into the noise."  As @Miska has referred to in the context of training sonar operators, and as those of us who tuned in distant AM radio stations as kids that we listened to through all sorts of static and interference can attest, people are able to recognize "signal" - speech, music, the sound of a parent's, child's, or friend's voice in the midst of a group - even when some or all of that signal is below the level of other sounds in the environment.  If you looked at a scope capture of what came through your radio from one of those old so-staticky-it-was-just-barely-audible AM stations, the signal part might well not be apparent amidst the noise.

 

All the examples I've mentioned have something in common: pattern recognition/matching.  At the present moment, humans have greater capability than instruments or computers for at least some types of pattern matching.

 

Before I don't think about the phenomenon in connection with audio.

[audiventory]

2 minutes ago, mmerrill99 said:

You're  a software guy, right? Most newly introduced software has bugs, right. Why aren't all bugs revealed in measurements/testing & are only revealed when the appropriate operational conditions arise?

 

Is it because the testers don't know what these conditions might be before release i.e the software isn't fully characterized?

 

The same applies to audio measurements - yes the tools are very sensitive but are all measurements made which completely characterize an audio device?

 

Does you meant analyzing software? It don't placed at market. Now it for internal using only. So while no sense to discuss it.

I meant "proper" not as "without bug", but as "in some features need spectrum, for other - oscillogramm, for next - spectrum-time diagramm, etc.".

 

If we consider audio devices as tools to store and transmit audio data without distortion, there primary feature:

 

input/output amplitude response by frequency.

 

The feature allow calculate harmonic distortions, amplitude response, dynamic range in any definition and other kinds of distortions.

[audiventory]

3 hours ago, sphinxsix said:

Part of the problem with some (in fact most) objectivists here is that they never say: 'I can't hear the difference' or 'I've never heard such a difference'.

 

My life learned me not once, that improbable things may be probable sometimes

 

Some cases may have other "objective" (measurable and repeatable experiment) reasons, than visible at first sight.

 

[audiventory]

21 minutes ago, mourip said:

What is fidelity?

 

"Fidelity" is "The degree of exactness with which something is copied or reproduced" (Oxford dictionary).

 

If take the definition it is not about sound enchancers and analog sound advantages (specifical nice distortions).

 

Fidelity of playback sine wave fully show fidelity of system (after complex of measurements).

 

Modern digital systems have good "2-D" fidelity (harmonic distortions, frequency response's flatness, linearity, etc.)

 

It open us, that now it is time to pay attention to "3-D" fidelity - to get at home all, that we hear in concert hall.

 

In the concert hall we hear not simple "sine" but we hear interference of acoustic rays by 1 sine - sound wave field.

And this sound field should be reproduced in our ears (individually for each ear) and, desirably, rest body.

Here I considered this issue more detailed https://samplerateconverter.com/content/where-limit-audio-quality

[audiventory]

9 hours ago, semente said:

I wonder what effective measurements would one use to determine the resolution of a loudspeaker driver at low levels...

 

In addition to non-acoustical component features, speakers have dependency of the features by spatial coordinates.
The dependency may be measured in complex with room only.

 

Me seems, deep analysis of features of acoustical/non-acoustical units/systems and comparison between units/systems by measurements are tasks for artifical intellect.

[audiventory]

1 hour ago, fas42 said:

I prefer my music to sound like a "live concert", meaning, you can move around, do other things, not be locked into a tight focus on what's happening, riiight theerre!!

 

Currently we haven't it in single point with speakers.

[audiventory]

16 minutes ago, fas42 said:

 

Not quite sure what you mean by "single point with speakers"  - but if you mean that the sense of part of the sound is coming from a particular place with the soundfield, no matter where you are in the area - that's perfectly possible. If a particular element in the picture is five feet behind the left speaker, and a foot to the left of that speaker, that's where it will always appear to be located, no matter where you happen to move within the room.

 

"Single point" is "single point of listener's position".

 

Sound field (sound hologram) in the point should repeat sound field in some point of concert hall.

The sound field in some point of concert hall is result of interferention of acoustic rays of source there.

 

Speakers is sources that spread acoustic rays in all directions. I.e. captured and played back sound field of concert hall radiated not as single ray directly to ear (for each ear), but the single ray transformed to several rays, that interfere in listening room. It distort captured sound field of concert hall.

 

To correct reproducing concert-hall sound field, need deliver it directly to ears without multiplication to rays.

Easiest way now it is headphones.

 

Otherwise speakers should follow listener and keep creation point of the concert-hall sound field into listener ears, to avoid multiplication to rays in listening room with reflection and re-reflection.

 

Anechoic room can't solve the issue, because speaker is omnidirectional sound source and we have non-concentrated direct waves.

 

[audiventory]

14 minutes ago, fas42 said:

My experience is that the soundfield as captured by the recording can dominate that of the listening room. Think of there being two, competing sound fields in the listening room at any one time while replaying a recording: firstly, that of the recording itself, the sound picture that has been encoded in that; and secondly, the speakers projecting sound into the room, which then interacts with the surfaces and furnishings of the room. Most audio playback has these two disparate "scenes" of sound in conflict with each other, subjectively, and hence the huge efforts to resolve those conflicts. The reason they conflict is because there is no clearly dominant soundfield.

 

However, the dilemma can be resolved by ensuring dominance of the encoded recording soundfield; the ear/brain then rejects the speaker/room soundfield, it becomes a sideshow - in one sense, this is exactly equivalent to using headphones to attenuate, dramatically, the room contribution.

 

So, the goal is to get the subjective isolation of the headphones, without the claustophobic environment that headphones present - this is certainly possible with speakers, and IME is far superior to the headphones experience.

 

I want add information to full clarity. I mean "sound field" in wave theory meaning: sound source radiate infinite number of rays. These rays interfere in all points of concert hall. "Sound field" is result of the interference in a point of the hall.

Listener ears is 2 capturing points. I.e. 2 sound fields are captured.

 

At listener first ear point of listening room, "sound field" should be produce 1:1 like concert hall.

Same thing to second ear.

 

[audiventory]

1 minute ago, semente said:

The brilliance range is composed entirely of harmonics and is responsible for sparkle and air of a sound. Boost around 12 kHz make a recording sound more Hi Fi.

 

Exactly. EQ is one of sound enchancement ways, that I remember each time, when discussed subjective sound quality. It may really sound better. Though, it is not fidelity matter.