[Poll] Is no-distortion really better than any distortion?

[pozz]

Somebody remind me how to post 3 wave files please. I have 3 files, one of which is noise (to avoid any copyright issues), one of which is filtered noise, and the third of which is inverse filtered, filtered noise.

This clearly demonstrates the reversibility of linear processing.

I will also note that the inverse filtered, filtered noise minus the original results in errors around 1e-15, well under 16 bit level.

Do you have a Soundcloud or Dropbox account?

Or you can use WeTransfer to produce a temporary link.

 

[lashto]

Take note @Robin L , this right here is utter nonsense. We're talking about whether all transducers audibly distort @lashto , and then you suddenly start talking about blind ABXing transducers that differ in other attributes like frequency response etc. I'm baffled why you think that's relevant here. I've literally just linked to measurements of a transducer that has THD about 0.05% around our ear's most sensitive frequencies. This is ~-65 dB, well below your -50 dB value.

Spoiler: only for booo

You can safely assume that "I don't get it". And never will . It's a 'simple' matter of psychology & philosophy: I just don't get anyone who thinks that _anything_ in this universe is known with 100% certainty. Or anyone who thinks that his opinion is worth more than others (in an opinion poll).
Anyway, just forget the above. Also forget that ABX stuff which seems to be so troublesome for you.

Just give us a link to a test which proves that a transducer does not have audible distortion.

Transducer as in "any apparatus that produces an audible sound wave". Could be any speaker, any head/earphone, any whatever. And you may define distortion as HD-only if you wish. Or even less, link an example of a test methodology which, as of AD2020, can (theoretically and practically) prove that a transducer does not have audible (H)D. I have a pretty hard time imagining how would that test go but then I am not an EE .. or any sort of audio expert.

@j_j many thanks for joining the discussion. Maybe you are aware of such a test?!

 

[lashto]

I wonder, since the term "distortion" is often colloquially used to refer to any alteration of the sound, whether, when the OP first started this thread and the poll, he meant this term broadly in its colloquial sense or narrowly in its technical sense.

The context of the question seems to raise the issue whether some alterations in the sound can be euphonic, whether or not these alterations are linear. If so, that suggests the question may use the term "distortion" in its general informal sense. For example, in some cases, adding the right amount of the right kind of noise can subjectively improve sound quality (indeed, it can objectively increase resolution, but that's a different subject). Strictly speaking, that (adding noise) is not distortion. But I speculate that and potentially other examples may be within the purview of the OP's question.

This thread/poll is only about people's perceptions and opinions of Distortion, and it's NOT supposed to be about the definition of D. The OP clearly says "feel free" (to use your own interpretation of distortion).

But since it seems to be the subject of the day, I am also curious: does an universally accepted and/or standardized definition of (audio) distortion exist !?
Wiki defines Distortion generally as "any alteration", @j_j has a good point about "only non linear alterations" (and IIRC @solderdude and others did post ~same), AverageJoe will probably tell you something like "sounds bad", etc... A pretty gray-ish mess AFAICS.

 

[solderdude]

When you change the frequency response you can undo this without problems.
You change the waveform but do not distort it. It can be undone.

When you add non-linearities you add frequencies that aren't supposed to be there and cannot be undone.

When you change the frequency response (deliberately) to get a more pleasing sound (tone control) then that's fine.
It sounds 'better', is reversible and does not introduce frequencies that aren't supposed to be there.. It's a taste thing that changes the electrical waveform.

When you change the electrical waveform so that the acoustical result is closer to the original waveform then the goal is to improve tonality (fidelity) of the entire reproduction system. In order to do so electrical waveform is changed but not by adding frequencies that aren't in the original waveform.
These 'changes' do not degrade fidelity.

When you change the waveform, purposely, to add frequencies that aren't there and do so that it reaches audible levels it is another thing.
You cannot undo that. And while harmonics are present in all instruments and it's levels may change somewhat giving 'color' changes of instruments (and voices) you also add frequencies that are not present in the original waveform and cannot be undone.
As long as those extra frequencies remain below objectionable levels or add some kind of noise and you love it that's fine but irreversible.

The claim is that 'non-linearities' are not detrimental to the sound and would be beneficial (more warmth, more 'musical', more 'being there', more 'analog') but that's really all a belief. Someone may like it, or prefer it... that's fine. It actually IS distortion just not reaching objectionable levels.
Don't say it is better.

 

[lashto]

When you change the frequency response you can undo this without problems.
You change the waveform but do not distort it. It can be undone.

When you add non-linearities you add frequencies that aren't supposed to be there and cannot be undone.

When you change the frequency response (deliberately) to get a more pleasing sound (tone control) then that's fine.
It sounds 'better', is reversible and does not introduce frequencies that aren't supposed to be ther.. It's a taste thing that changes the electrical waveform.

When you change the electrical waveform so that the acoustical result is closer to the original waveform then the goal is to improve tonality (fidelity) of the entire reproduction system. In order to do so electrical waveform is changed but not by adding frequencies that aren't in the original waveform.
These 'changes' do not degrade fidelity.

When you change the waveform, purposely, to add frequencies that aren't there and do so that it reaches audible levels it is another thing.
You cannot undo that. And while harmonics are present in all instruments and it's levels may change somewhat giving 'color' changes of instruments (and voices) you also add frequencies that are not present in the original waveform and cannot be undone.
As long as those extra frequencies remain below objectionable levels or add some kind of noise and you love it that's fine but irreversible.

All good and very interesting points but still: is there a clear and universal/standardized definition of (audio) distortion?!

The claim is that 'non-linearities' are not detrimental to the sound and would be beneficial (more warmth, more 'musical', more 'being there', more 'analog') but that's really all a belief. Someone may like it, or prefer it... that's fine. It actually IS distortion just not reaching objectionable levels.Don't say it is better

That is a DBT-proven belief. With "yes it is better" included (as in ">50% of subjects do prefer it"). Sounds more like a fact to me.

 

[solderdude]

Sounds more like a fact to me.

That's your opinion.
It is fine to have an opinion.
Doesn't make it factual because it sounds like a fact to you.
The fact is that non-linear distortion adds things that aren't there and have changed the waveform.

is there a clear and universal/standardized definition of (audio) distortion?!

There is no single definition of (audio distortion). There are several types of distortion. One is more detrimental to (perceived) sound quality than others. They are all measurable and verifiable and there are audibility thresholds depending on the applied stimulus.

When the goal is to reproduce a waveform exactly no alterations are preferred. (Is what ASR is about)
In practical sense this means low distortion (linear and non-linear) or for most at least inaudible (thresholds) levels.

When the goal is to reproduce soundwaves at a certain listening position that as closely as possible resemble the original waveform then the electrical waveform has to be altered in such a way that the final result is least changed. This may be done by changes in phase and amplitude but does not involve adding frequencies that aren't in the waveform. (Also is what ASR is about)

When the goal is to listen to preferred sound and one has a preference to do it in this or that way... fine, great for you. All bets are off and is not what ASR is really about but some still prefer. Add whatever you like. Don't say it is 'better' for music reproduction. It is preferred/like by some.

Then there are audibility levels.

When you realize the implications from above it should be clear that there is no single definition in audio. The only reasonable definition is that the applied signal is as close as possible to the original (fidelity). There is an electrical path and an acoustical path here and the third factor (the least quantifiable one) is perception where also preference plays a role.

That last bit is where your poll is 'fuzzy' and not ASR alike. The first 2 are. This makes your poll utterly inconclusive and you should have worded it better and give better options. That is if you want to find out how ASR members 'think/feel/opine' about distortion levels and types.

 

[lashto]

That's your opinion.
It is fine to have an opinion.
Doesn't make it factual because it sounds like a fact to you.
The fact is that non-linear distortion adds things that aren't there and have changed the waveform.
.

"sounds better" is of course a matter of personal preference and opinion. 100%
However, a DBTested preference is a proven Fact not just some lalala Belief. Or am I missing something in this 'beautiful' nitpicking contest!?
.

There is no single definition of (audio) distortion.

Thanks, that's all I asked for.

 

[solderdude]

However, a DBTested preference is a proven Fact not just some lalala Belief. Or am I missing something in this 'beautiful' nitpicking contest!?

A DBT proven preference is merely proving one or more people have a preference for a sound. You cannot conclude anything other than that.
You can DBT a 'flat' response versus a colored response where the coloration is strictly due to linear distortion, or due to solely non-linear distortion or versus devices that add both non-linear and linear distortion in certain amounts where is isn't clear what triggers the preference. Tonal balance or added frequencies in not sound degrading amounts.
The latter is what is often done. Compare tube amps which not only add frequencies but also change tonal balance and the real question is WHAT your brain tells you it likes.

But this is about the poll and consequences. Asked in a not very scientific way which is prone to be misunderstood with equally flawed choices.
You say you got your answers but wait to see conclusions of others instead of clearly speaking your mind on the results.

That's what this is about. Your 'distortion' means something entirely different than what most ASR members consider distortion to be.
Distortion is usually seen as a negative thing not something to 'improve' sound quality.

Just create another poll with the following options. ( I am not going to do this, you or someone else can )

When it concerns a complete audio system what would be your preference regarding the aspect 'distortion'.

• I want as little as possible ... or ... I want at least 0.00x %
• I don't mind distortion as long as it stays below audible levels
• I don't even mind if it is reaching audible levels but does not ruin the music experience for me
• When it sounds fine to me it's O.K, I couldn't care less/have no idea about this 'distortion' thing.

Then you will better know how ASR members feel about what I think you wanted to find out.

 

[lashto]

A DBT proven preference is merely proving one or more people have a preference for a sound. You cannot conclude anything other than that.

Never planned to.
Just wanted to 'nitpick' that the preference for some HD spectra is a well documented and well proven Fact. And since it is preferred by most people, it's a very important Fact. It's not just some audiophile belief that people lalalala about endlessly and uselessly, it's something that everyone should acknowledge, accept and consider "as is". Same as for any other Fact. (fans of "alternate facts" please look elsewhere

Your 'distortion' means something entirely different than what most ASR members consider distortion to be.

What exactly is "your distortion"?!?!
I never even tried to define D, neither in this thread nor anywhere else. Also specifically said a few times "feel free to use your own definition".
I don't even know how to define D. And it looks like nobody else does (yet).

... the real question is WHAT your brain tells you it likes.

exactly.
Technically, any HD is (undesirable) distortion but "who cares" is also a very important question. For many people it's actually the more/most important Q.

 

[solderdude]

that the preference for some HD spectra is a well documented and well proven Fact. And since it is preferred by most people, it's a very important Fact.

Can you point to these proven facts ?

How do you know whether or not people prefer a certain amount of linear distortion and its profile and what part is actual changed FR depending on other factors. You may think/believe/feel it is the distortion profile but may well be a change in FR that is the real audible issue.

 

[lashto]

Can you point to these proven facts ?

How do you know whether or not people prefer a certain amount of linear distortion and its profile and what part is actual changed FR depending on other factors. You may think/believe/feel it is the distortion profile but may well be a change in FR that is the real audible issue.

use @pkane's Distort or a similar tool to add any HD spectra you wish. And only that. ABX it, DBT it, do whatever test/comparison you consider relevant. It's easy and you don't need to take anyone's word or test for it

Or, since you are an expert, do it the hard(ware) way. Like this guy. Or this other guy

 

[solderdude]

Yes, you can do that with Paul's software, but where is the objective proof that people prefer certain distortion profiles over having no distortion.
You mentioned it is a well proven FACT.
And which distortion profile would that be and is that profile the same for all frequencies.

 

[lashto]

Yes, you can do that with Paul's software, but where is the objective proof that people prefer certain distortion profiles over having no distortion.
You mentioned it is a well proven FACT.
And which distortion profile would that be and is that profile the same for all frequencies.

You have a PM. There is already too much talk about Distortion itself here. And it's quite offtopic. This poll/thread is not supposed to be about D itself but about people's opinions of it.

Sorry for 'encouraging' that offtopic discussion, hope we can all go back to talking about people's Opinions/Votes.

 

[BDWoody]

Just wanted to 'nitpick' that the preference for some HD spectra is a well documented and well proven Fact. And since it is preferred by most people, it's a very important Fact.

Can you support that statement?

You throw that out as if it is generally accepted fact. I don't believe it is.

 

[solderdude]

Is no-distortion (or below audible limits) always better for music playback than any distortion ?!

So it's not about distortion ?

 

[solderdude]

The PM'ed 'proof' turned out to show the majority of the folks taking the test preferred low mounts of distortion to inaudible amounts of distortion.
Not surprisingly ...
@lashto needs to find other DBT evidence that shows the majority of people enjoys audible amounts of distortion.

 

[Plcamp]

I see distortion as a figure of merit (among others like efficiency) that one should strive to improve. And to me that means speaker design, because it dominates.

Then again, I would not treat it as the absolute at expense of everything else. I would, for example, probably prefer a highly efficient speaker with higher distortion, than a far less efficient one that doesn’t achieve the dynamic range that sound realistic.

 

[BDWoody]

The PM'ed 'proof' turned out to show the majority of the folks taking the test preferred low mounts of distortion to inaudible amounts of distortion.
Not surprisingly ...
@lashto needs to find other DBT evidence that shows the majority of people enjoys audible amounts of distortion.

Yeah...I got that too.

From the 'proof':

Looking at the results here, I believe the data supports the idea that in a blind test, of those who reported hearing sonic differences, lower harmonic distortion did correlated to "better" sound.

I'm missing something...

 

[tomtoo]

We talk about the reproducing chain thats very importend. In the producing chain on many places intentionaly distortion gets added. Thats why imo it makes not much sense adding things up in the reproducing chain. If we like to talk about hifi leave things how they where intendet. If you talk about personal preverence, hell do what you enjoy.

 

[Jimbob54]

Yeah...I got that too.

From the 'proof':


I'm missing something...

I have no wish to see the "proof"- just to clarify though, is it suggesting the folks that heard a difference preferred inaudible distortion, or low level (but audible) harmonic distortion ?