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?!
-
WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!
[Poll] Is no-distortion really better than any distortion?
- Thread starter lashto
- Start date
-
- Tags
- audibility distortion noise sinad thd
- Joined
- Feb 23, 2016
- Messages
- 20,759
- Likes
- 37,607
Care to give a summary. I don't have AES privileges
- Joined
- Feb 23, 2016
- Messages
- 20,759
- Likes
- 37,607
So 5 headphones used, one was least preferred, and trained listeners seemed not to have a real preference based upon distortion otherwise for the remaining 4 headphones.
Here is a video of the results being presented.
Last edited:
You can safely assume that "I don't get it". And never will
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?!
First, we'd have to define "in what space". Since speakers interact enormously with rooms, that's, well, vague is a very kindly way to put it.
Then we'd have to agree on "distortions". Do linear modifications count (i.e. they can be EQ'ed out)? In what space? Directionality. Nonlinearities only? Noise sources (air movement, voice coil air movement, port movement, buzzing of box, radiation from box, ).
Calling this "ill defined" isn't even strong enough.
That is the sum of this discussion, I think. One can not address questions of preference universally for all people. One could, likely, characterize what a given person likes AT THE MINUTE. That can change, and often does, for instance when one learns to hear impairments that they previously did not notice.
If someone likes distortion, frequency shaping, time-delay aspects (i.e. phase shift), well, that's their preference, and they are entirely welcome to it. It's rude to judge someone else's preference UNTIL they try to make it universal. And there, in fact, is where the apples hit the cider mill, so to speak.
- Thread Starter
- #446
I had no idea about your work and 'discovered' it independently so i'll see you both in court
More seriously, if this was patent-able, ESS would've done it. Seems to be 'just' a math algorithm so maybe patent a device/chip that implements it. Anyway, if anyone can figure out a practical algorithm/device, I'll donate to your kickstarter/etc.
Some previous work (with code examples):
- Euphonic Distortion: Naughty but Nice? DSP-like code from Stereophile of all places.
- Christortion VST (probably there are tens/hundreds of similar ones).
- And of course pkane's really cool Distort.
I also remember reading a paper some years ago which claimed that HD-cancellation is the reason why SET amps sound so good/clean: i.e. the big HD spikes of a SET amp play a cancellation game with the ~same as big HD spikes of a speaker. Far fetched hypothesis and IIRC there were no practical examples/measurements but the math looked alright (for a non audio expert like me). Might be able to google it.
The computation seems daunting but the theory says it's possible, at least to some (usable) extent. The H2 and H3 would be easiest/fastest to remove but those are exactly the ones that you don't want to remove. And the H5+ ones that you really want out, would be much harder/complex (a point for a diff discussion/thread I guess).
I'd say the anti-HD DSP filter should be somewhat similar to a RoomCorrection filter. Just based on the measured HD of dac/amp/speaker (instead of room measurements). Or it may be more similar to noise-cancelling tech. In any case, you'll need to apply the anti-HD-filter during playback. Not sure which would be better: before or after the RC filter.
Good luck!
Last edited:
This is your chance @pkane to grab back all the HD that you put out in the universe
I was waiting for the analysis of the results of this poll to figure out if a clean up was necessary or if I needed to produce more distortion
Reversing non-linear distortion caused by a simple non-linear amplitude transfer function is trivial if you know or can measure the transfer function. Any non-linearity that involves time is much harder to undo (for example where negative or positive feedback is involved).
solderdude
Grand Contributor
With HD comes IMD... the HD is not an issue at all.
It's the IMD
Frequencies are generated that were not there nor are harmonic nor do they 'enhance' anything as they are not related to music.
The IMD is the issue not the HD.
When you build amps with very low distortion (you know the ones that aren't preferred with 100% certainty) for around $ 100.- why build an effect box and try to uneffect it making it the same as the one with high feedback.
removing distortion with speakers/headphones will only work with lower frequencies (partly).
If it were easy to create an algrithm that 'undistorts' music wouldn't you expect it to exist already ?
Why would those buying 'effect boxes' want to remove the HD (and IMD) ?
It's the IMD
Frequencies are generated that were not there nor are harmonic nor do they 'enhance' anything as they are not related to music.
The IMD is the issue not the HD.
When you build amps with very low distortion (you know the ones that aren't preferred with 100% certainty) for around $ 100.- why build an effect box and try to uneffect it making it the same as the one with high feedback.
removing distortion with speakers/headphones will only work with lower frequencies (partly).
If it were easy to create an algrithm that 'undistorts' music wouldn't you expect it to exist already ?
Why would those buying 'effect boxes' want to remove the HD (and IMD) ?
- Thread Starter
- #449
It is surely a faaaar from trivial task. And I still have serious doubts that one can do a 100%, chirurgical removal of all HDs. However, it should be possible to do a good-enough/audible removal. IIUC, the latest ESS dacs do some of that for H2 & H3.I was waiting for the analysis of the results of this poll to figure out if a clean up was necessary or if I needed to produce more distortion
Reversing non-linear distortion caused by a simple non-linear amplitude transfer function is trivial if you know or can measure the transfer function. Any non-linearity that involves time is much harder to undo (for example where negative or positive feedback is involved).
And I bet people will pay for the anti-HD filter. E.g. in this thread alone there are 94 voters who can hardly wait!
Here's a deal @pkane: I would pay $2 for the H2, $3 for the H3 and so on. Sounds like a 100% fair price to me
- Joined
- Oct 25, 2019
- Messages
- 11,111
- Likes
- 14,774
It is surely a faaaar from trivial task. And I still have serious doubts that one can do a 100%, chirurgical removal of all HDs. However, it should be possible to do a good-enough/audible removal. IIUC, the latest ESS dacs do some of that for H2 & H3.
And I bet people will pay for the anti-HD filter. E.g. in this thread alone there are 94 voters who can hardly wait!
Here's a deal @pkane: I would pay $2 for the H2, $3 for the H3 and so on. Sounds like a 100% fair price to me
There are 164 still waiting for your conclusions. Priorities.
- Thread Starter
- #451
Judging by the amount of open questions, I guess it is safe to assume that you do not know of any such test/methodology either.
Your questions are all good but I do not have answers. Only what I said already: you can define D as 'simply' non-linear D, or even HD-only if easier. Otherwise, I am just another guy with the ~same amount of open questions. And also said already that I don't know how to do such a test.
The source of the whole discussion was this:
I just asked mister booo to tell us how did he test that those HPs "don't distort audibly". Looks to me that he is the one who already knows all the missing answers.
Last edited:
- Thread Starter
- #452
that looks like just another proof for the good old "the THD value does not correlate with good/likable sound". Or am I missing something?!
Here's a deal @pkane: I would pay $2 for the H2, $3 for the H3 and so on. Sounds like a 100% fair price to me
That's price per individual music track, right?
I'm on it!
- Thread Starter
- #455
that is a tiny bit too greedy for my taste but Elon Musk might be able to afford one album per day.That's price per individual music track, right?
Looks like this will be a hard negotiation, we'll need lawyers and stuff
P.S.
Found an interesting paper while looking for D simulators. Too long & math dense for me but maybe you can use it: DIGITAL IMPLEMENTATION OF MUSICAL DISTORTION CIRCUITS BY ANALYSIS AND SIMULATION
Last edited:
- Thread Starter
- #456
Surely not all 164 still care about it.
And after all the 'abuse' and noise I had to put up with in this thread, I might need a week of vacation. With 3 shrinks. I hear they make excellent beer buddies.
May find some time tomorrow. Or during the weekend. Hopefully...
solderdude
Grand Contributor
- Thread Starter
- #458
from those I really had to take a 1 day vacationYep.
I do try to read everything in this thread. Can't though. If it was that important, do you have a link, a helpful hand, a beer...?!
And if you mean the IMD stuff, you're probably right AFAIK, what else should anyone say?
Presumably, if you cancel the HD soon enough, the IMD will not happen. Otherwise, IMD-cancelling might be a job for alien civilizations with Dyson-sphere amounts of energy & computation. Or maybe not...
Last edited:
- Thread Starter
- #459
It's an opinion poll and people post well, opinions. That is pretty messy stuff. By definition I would say.
After all the 'work', I am not the biggest fan of this thread either...
And generally, you should not expect to find answers in this thread. Just questions. Wide open ones...
This threads gone beyond being useful and now seems to be creating animosity with no upside as far as I can see so I'm closing it.
thread closed
thread closed
Similar threads
- Poll
- Poll
