How audible is distortion?

[j_j]

Sorry but I'm still lost, what are the variables you speak of?

At any one point in a sound field, there are 4 variables.
Pressure
X volume velocity
Y volume velocity
Z volume velocity

xyz can be any set of 3d orthonormal axes.

 

[Krunok]

Rather, it's about using compression and other processes to make the average level louder. So soft parts are raised in volume making everything the same, which makes listening tedious and tiring.

Why are they doing that? To avoid clipping and other issues on the lower quality listening equipment? Or for some other reason?

 

[Cosmik]

First, no, a listening test is not a mechanical test, which is to say a test involving only instrumentation and measurement. So let's not confuse the issue, listening tests ARE THE ONLY DEFINITIVE TESTS FOR THINGS LIKE CODECS.

And, yes, the human ear works the same anywhere that human beings can understand human speech. The two sets of properties are inextricably related.



For codecs, that's pure bollox. With speakers you don't have an 'absolute reference'. For a codec, you have an absolute reference, and you can do a distance test, a signal detection test, or any of a variety of blind test methods USING LISTENERS to determine the audibility of a coding system.

And that remains the only valid way to test any codec.

For speakers you are already completely beyond the "transparent" level and can never, EVER get there, so then you can simply test for preference for speakers in a given acoustic.

When you MAKE a speaker, yes, you absolutely must measure a host of things, but no matter what you measure, you'll never, ever get to reproduce the soundfield in the concert hall. At best with stereo, you can reproduce 2/8ths of the actual soundfield at your two ears, even if you forget that heads move around.

And, of course, what point in the concert hall? First row, middle, last row. These three locations have enormously different direct/reverberant ratios, and the speaker required to provide the same experience, coupled with the room, must have different radiation patterns.

And, of course, headphones are another question altogether.

Please do not continue to spread myth about testing of codecs. For codecs you're wrong, full stop.

All right, steady on.

My comment was referring to the general case of the futility of 'development by listening test'. Codec development is no doubt a special case but even so it still needs some "innate structure" or how else does the original box of tricks get programmed in order to be tweaked?

My comment about USA vs. India was a reference to self-driving car development. The parallel for codec development would be that it might be used on sounds and sequences in the 'information space' that have not been tested explicitly by listening test and so might cause the system to fall over. My audacious, dangerous argument was that developing anything in audio purely by listening to stuff would produce 'brittle' results because of the tiny fraction of the information space it is possible to cover in testing; but obviously listening is necessary to check it is basically working - two different things.

I didn't realise codec development was such a hot topic that scurrilous rumours abounded and I was posing an existential threat to it by discussing ideas on this forum.

 

[j_j]

I didn't realise codec development was such a hot topic that scurrilous rumours abounded and I was posing an existential threat to it by discussing ideas on this forum.

There are a lot of people using a couple of insufficient standards that are purely programmed, with no listening effort at all, for codec testing.

This leads to codec designers writing quantization methods that measure well, rather than sound good. Sorry, that's how it's going down in some places, and it's NOT a good thing.

 

[watchnerd]

I'm starting to find myself with a paradox.

Philosophically, with the goal of reproduction, I have no doubt that low distortion in every element of the chain lets me hear more of what is on the original recording. When it comes to hearing details, macro dynamics, maximum frequency range, and low noise, there is no doubt that my digital chain is objectively better at reproduction than my analog chain.

However, when it comes to recreation, the sense that I'm listening to a live music event, my vinyl system, with all its distortions, can often give me more of a simulation of live music. What I *think* is happening is that resonances (distortions) amplify dynamic transients, while at the same time the relatively benign and euphonic 2nd order harmonics add extra timbre above what is actually in the recording, which seems to psycho-acoustically enhance it in a way that makes it seem closer to live.

If correct, all I need is a DSP algorithm to apply the same distortion effects to digital.

 

[Blumlein 88]

I'm starting to find myself with a paradox.

Philosophically, with the goal of reproduction, I have no doubt that low distortion in every element of the chain lets me hear more of what is on the original recording. When it comes to hearing details, macro dynamics, maximum frequency range, and low noise, there is no doubt that my digital chain is objectively better at reproduction than my analog chain.

However, when it comes to recreation, the sense that I'm listening to a live music event, my vinyl system, with all its distortions, can often give me more of a simulation of live music. What I *think* is happening is that resonances (distortions) amplify dynamic transients, while at the same time the relatively benign and euphonic 2nd order harmonics add extra timbre above what is actually in the recording, which seems to psycho-acoustically enhance it in a way that makes it seem closer to live.

If correct, all I need is a DSP algorithm to apply the same distortion effects to digital.

I've had this same idea. A clean transparent medium, and DSP to taste.

I like push-pull triode sound at least for some types of music. I like ribbons over condensers on the recording end for some situations. The long standing problem to me has been audiophiles confusing preference for fidelity.

 

[Thomas savage]

I've had this same idea. A clean transparent medium, and DSP to taste.

I like push-pull triode sound at least for some types of music. I like ribbons over condensers on the recording end for some situations. The long standing problem to me has been audiophiles confusing preference for fidelity.

The end lol...

 

[Cosmik]

I'm starting to find myself with a paradox.

Philosophically, with the goal of reproduction, I have no doubt that low distortion in every element of the chain lets me hear more of what is on the original recording. When it comes to hearing details, macro dynamics, maximum frequency range, and low noise, there is no doubt that my digital chain is objectively better at reproduction than my analog chain.

However, when it comes to recreation, the sense that I'm listening to a live music event, my vinyl system, with all its distortions, can often give me more of a simulation of live music. What I *think* is happening is that resonances (distortions) amplify dynamic transients, while at the same time the relatively benign and euphonic 2nd order harmonics add extra timbre above what is actually in the recording, which seems to psycho-acoustically enhance it in a way that makes it seem closer to live.

If correct, all I need is a DSP algorithm to apply the same distortion effects to digital.

I have the notion that there might be an 'uncanny valley' in audio where if you get close to neutral, but not quite close enough, it sounds 'weird'. In that case you need to introduce enough distortions to kick it firmly away from neutral and into the artificial.

You can have two aims, it seems to me: a system so clean and neutral that it sounds like a window onto the music and has no sound of its own, or a system whose sound you like.

Are passive speakers (low in harmonic distortion maybe, but high in other types of distortion including phase and timing) a barrier to getting a sufficiently neutral system?

 

[j_j]

The long standing problem to me has been audiophiles confusing preference for fidelity.

AAAAAAaaaaaaaaaaaaaaaamen!

 

[oivavoi]

I'm starting to find myself with a paradox.

Philosophically, with the goal of reproduction, I have no doubt that low distortion in every element of the chain lets me hear more of what is on the original recording. When it comes to hearing details, macro dynamics, maximum frequency range, and low noise, there is no doubt that my digital chain is objectively better at reproduction than my analog chain.

However, when it comes to recreation, the sense that I'm listening to a live music event, my vinyl system, with all its distortions, can often give me more of a simulation of live music. What I *think* is happening is that resonances (distortions) amplify dynamic transients, while at the same time the relatively benign and euphonic 2nd order harmonics add extra timbre above what is actually in the recording, which seems to psycho-acoustically enhance it in a way that makes it seem closer to live.

If correct, all I need is a DSP algorithm to apply the same distortion effects to digital.

Since it's the day for such confessions: I recently heard a tube based system (driving horns), and to my astonishment I perceived it as very "musical" - the very audiophile descriptor of tubes that has made me smirk. It was definitely not fidelitous, but on acoustic music it sounded very "real". I suspect it has to do with the somewhat lush and dark tonality, which is in line with how acoustic music can feel in real venues (high frequencies get attenuated over distance). Perhaps also how the distortion of tubes can vary with volume (that euphonic distortion increases when volume goes up, and this mimics how real instruments can behave).

But would I exchange that system for my D&D 8Cs? Nope

 

[Cosmik]

Since it's the day for such confessions: I recently heard a tube based system (driving horns), and to my astonishment I perceived it as very "musical" - the very audiophile descriptor of tubes that has made me smirk. It was definitely not fidelitous, but on acoustic music it sounded very "real". I suspect it has to do with the somewhat lush and dark tonality, which is in line with how acoustic music can feel in real venues (high frequencies get attenuated over distance). Perhaps also how the distortion of tubes can vary with volume (that euphonic distortion increases when volume goes up, and this mimics how real instruments can behave).

But would I exchange that system for my D&D 8Cs? Nope

Was the music 'low complexity' (e.g. solo voice, duet, small number of instruments) or 'complex' (e.g. symphony orchestra)? The effect must surely be dependent on the material and, I often think, may influence the music that vinyl/valve audiophiles listen to...

 

[Wombat]

Was the music 'low complexity' (e.g. solo voice, duet, small number of instruments) or 'complex' (e.g. symphony orchestra)? The effect must surely be dependent on the material and, I often think, may influence the music that vinyl/valve audiophiles listen to...

Demonstrations of lower power SETs invariably use low complexity music from hand -picked vinyl sources.

 

[oivavoi]

Was the music 'low complexity' (e.g. solo voice, duet, small number of instruments) or 'complex' (e.g. symphony orchestra)? The effect must surely be dependent on the material and, I often think, may influence the music that vinyl/valve audiophiles listen to...

Listened to some variety of material. Yes, some things didn't sound "right". So it becomes an effect box, which can be pleasing on some material, and not on other things.

 

[watchnerd]

I have the notion that there might be an 'uncanny valley' in audio where if you get close to neutral, but not quite close enough, it sounds 'weird'. In that case you need to introduce enough distortions to kick it firmly away from neutral and into the artificial.

You can have two aims, it seems to me: a system so clean and neutral that it sounds like a window onto the music and has no sound of its own, or a system whose sound you like.

I think you're on to something with the uncanny valley concept.

Which, again, argues for neutral system + DSP to taste.

On the DSP side, I find it annoying that the numerous 'vintage gear' effects and techniques that exist as DSP plugins in the DAW world don't exist in consumer audio. The technology is there, we're not starting from zero.

 

[Sal1950]

Which, again, argues for neutral system + DSP to taste.

Isn't that pretty much what goes on here with the majority of our members?
Most of us do chose fairly neutral electronics and then minimally turn the tone controls to taste when we pick our speakers. Beyond that then, many are using DSP or active speakers to "tune" the system even further.
We all have our preferences and chart our own path.

 

[watchnerd]

Isn't that pretty much what goes on here with the majority of our members?
Most of us do chose fairly neutral electronics and then minimally turn the tone controls to taste when we pick our speakers. Beyond that then, many are using DSP or active speakers to "tune" the system even further.
We all have our preferences and chart our own path.

In terms of normal EQ via DSP, sure, lots doing it.

But I don't anyone who is running a 'make it sound like a SET' or other effects overlays.

 

[tomelex]

Audibility of distortion and preferences can merge. For one, only by comparing two things (ie you are at the sweetspot and amps or whatever are switched out in what I would say are comparable quality gear) can you say with certainty that "i can hear a difference" and then so what, how do you know which one is more distorting than the other, since you don't know what the original sound was played back at the studio over the speakers.

What I am getting at is sometimes one hears a system (in the above case a tube system post #90) and well it sounded good, OK, but there are so many things there besides the human variability of your "feelings and emotions and God knows what else is going on in your brain" sensor apparatus that you can like something that is clearly not as "fidel" to the source.

Let me give an example, most SET amps distort the lowest bass notes, and this distortion harmonics rise up into the midrange, and say you are comparing one SET against another, the one that distorts more of the low end notes will sound louder and more "bassy" and better to some folks, while the one with a much more "fidel" replication of the bass notes will sound more "solid state" as a term we might all understand. I have found this out with my designs and its interesting to see folks choose based on the more distorting bass notes as a "better" sound when comparing two SETS.

Now, any SET is a tone control but when you come down to it, everything is a tone control but by degrees. I wonder, if we had digital first and then came along vinyl with its interbreeding channel distortions, higher even harmonics and worst inter channel phase and micro noise and dynamic splatter would people have perceived it as more better? I have always said that a perfect two channel replication system might not be the best way to generate the simple and constrained illusion that stereo is trying to pull off.

 

[Sal1950]

But I don't anyone who is running a 'make it sound like a SET' or other effects overlays.

Not consciously no, as in turn the switch to SET or zero feedback SS imaging. But picking a "sounds good to me" preference my have taken him down the same sonic path.

 

[Blumlein 88]

Not consciously no, as in turn the switch to SET or zero feedback SS imaging. But picking a "sounds good to me" preference my have taken him down the same sonic path.

I had hoped such things as SET emulation or triode emulation or zero feedback class A emulation would have been available to consumers by now. All of that is possible, but I don't know of anyone offering it. Probably because so many audiophiles think SET's offer a superiority beyond the abilities of solid state or class D. It is not true, but that doesn't prevent the bulk of the potential market form believing it.

You do have such emulations in pro gear from some companies. One of my interfaces offers reasonably good emulations of various tube and old school solid state EQ and compressors. Not just a general one in all cases. Several of them give you the exact piece of gear it is supposed to sound like.

Universal audio is the best known purveyor of plug ins though not the only one. Here is a list of plug ins they offer listing the gear it is supposed to sound like.

https://www.uaudio.com/uad-plugins/all-plugins.html

 

[watchnerd]

I had hoped such things as SET emulation or triode emulation or zero feedback class A emulation would have been available to consumers by now. All of that is possible, but I don't know of anyone offering it. Probably because so many audiophiles think SET's offer a superiority beyond the abilities of solid state or class D. It is not true, but that doesn't prevent the bulk of the potential market form believing it.

You do have such emulations in pro gear from some companies. One of my interfaces offers reasonably good emulations of various tube and old school solid state EQ and compressors. Not just a general one in all cases. Several of them give you the exact piece of gear it is supposed to sound like.

Universal audio is the best known purveyor of plug ins though not the only one. Here is a list of plug ins they offer listing the gear it is supposed to sound like.

https://www.uaudio.com/uad-plugins/all-plugins.html

I love the UA plugins, and have good results with the RTR emulators, in particular.