THD - It's NOT the % of THD
Well, the computer ate my post...
...attempting grey matter data recovery...
Anyhow, it's not the absolute value of THD.
(got tired of waiting for this to come around in the discussion)
I think Geddes shows this in his work, iirc.
What I read was that amplifiers of wildly different % THD can be adjudged to to sound essentially the same. IF the spectra of distortion was similar.
Seems to me this is likely correct.
I think I read in LA, was it Putzeys (?hope I have the proper attribution) that amps with super-ultra high FB and amps with low, low FB may be quite similar in perceived sonic attributes, and that those in the middle are more of an uncertain lot.
This speaks to nature of (or lack of) spectra of distortion.
Which brings up an interesting question about the super-ultra low distortion (high feedback) amps and the spectra of distortion. Does dropping the spectra down into the noise floor have the same sonic effect as those amps with "good" spectra but much higher distortion levels?
So, in my book, the key factor in distortion is not absolute level.
(this was my conjecture for a long time, but I did not have the means by which to adequately test the hypothesis, Dr. Geddes did... so now I can point to it)
Btw, I wrote this much better the first time through... heh.
Well, the computer ate my post...
...attempting grey matter data recovery...
Anyhow, it's not the absolute value of THD.
(got tired of waiting for this to come around in the discussion)
I think Geddes shows this in his work, iirc.
What I read was that amplifiers of wildly different % THD can be adjudged to to sound essentially the same. IF the spectra of distortion was similar.
Seems to me this is likely correct.
I think I read in LA, was it Putzeys (?hope I have the proper attribution) that amps with super-ultra high FB and amps with low, low FB may be quite similar in perceived sonic attributes, and that those in the middle are more of an uncertain lot.
This speaks to nature of (or lack of) spectra of distortion.
Which brings up an interesting question about the super-ultra low distortion (high feedback) amps and the spectra of distortion. Does dropping the spectra down into the noise floor have the same sonic effect as those amps with "good" spectra but much higher distortion levels?
So, in my book, the key factor in distortion is not absolute level.
(this was my conjecture for a long time, but I did not have the means by which to adequately test the hypothesis, Dr. Geddes did... so now I can point to it)
Btw, I wrote this much better the first time through... heh.
These discussions do edge a bit toward some subtleties, which remind me of the character who wrote about how foolish audio engineers were by not measuring IM distortion, and focusing on harmonic distortion. The guy's name was New and I had a bit of fun about how his observations were not New's.
But he thought that because low-order HD was indistinguishable from frequency response variations, without specifying that this was true for constant amplitude, that any concern about such things was misplaced. Self and I contributed some testy LTEs (Electronics World, Doug's was truncated by the magazine, mine only very slightly altered). But a point I did not make then was that distortion mechanisms are level-dependent, and supply us with aural/perceptual clues that they are not merely a level-invariant change in frequency response. When they are non-monotonic the situation can be even more easily distinguished if the effects are significant enough (vide severe crossover distortion).
This may be a reason why it seems possible to separate the psychoacoustic effects of, say, speaker distortions, from other signal chain distortions, and why it may be legitimate, to a point, to minimize some despite the larger contributions of the latter.
But he thought that because low-order HD was indistinguishable from frequency response variations, without specifying that this was true for constant amplitude, that any concern about such things was misplaced. Self and I contributed some testy LTEs (Electronics World, Doug's was truncated by the magazine, mine only very slightly altered). But a point I did not make then was that distortion mechanisms are level-dependent, and supply us with aural/perceptual clues that they are not merely a level-invariant change in frequency response. When they are non-monotonic the situation can be even more easily distinguished if the effects are significant enough (vide severe crossover distortion).
This may be a reason why it seems possible to separate the psychoacoustic effects of, say, speaker distortions, from other signal chain distortions, and why it may be legitimate, to a point, to minimize some despite the larger contributions of the latter.
There may be some unknown spread in the answer..... but, I'd say .05% to .1% for me..
This might be a good threshold of Just Detectable Difference for myself. You can find out for your self what your JDD is.
If you can get your spkr to not totally mask your minimum JDD level.
THx-RNMarsh
Last edited:
Scott, enough.
you're putting words in my mouth again.
I did NOT say that two amps COULD not sound the same.
Why do I have to restate this, and correct you?
I asked YOU and others to tell me which ones DO????
No. I did not.
NO.
Again I asked YOU and the other readers to say what sounds the same and what sounds different.
I would expect there to be very little difference between two amps of the same model from the same mfr.
The post before this one that I made SAYS WHY. <---
This distortion stuff as any kind of individual metric is stupid, going several orders of magnitude below any estimates derived in reasonable studies (save for Ed's numerology
So, then you say that all amps with low distortion that are "blameless" to use that term again sound identical?? Right?
State your position, Scott.
It is EASY.
stop by... I live in Albany, you can get here in about 2.5hrs.
Bring ur DAC/CD player and your amp. Or anything else, bring a friend or two! You can all tell me whatever you wish about what you hear.
IF you hear no differences between mine and yours, great!
heck, you can hear the diffs between two DACs that I have, and two amps... or three... whatever!!
fwiw, the cues are not always in things that are obvious, but that's another story...
...there are no feats involved.
Just like I know someone who can reliably tell when a record is even a few cents offspeed (I can't) different people can hear differently.
But the post before this explains what is going on... <sigh>
(at least to a significant degree)
Did I say that? You said yesterday or maybe today or maybe both, that YOU have never heard two DAC's or amplifiers that sounded the same. How can you read my comment to mean you stated two amps COULD not sound the same?
Jeez again what did I say distortion alone is at this time a useless metric. Enough, time to listen to Popul Vuh's "Nosfetatu" soundtrack. Now there's music without a soul.
Last edited:
It begins with _everyone_ stopping the single-number description of distortion as "X%", as if all amplifiers were perfectly monotonic. That's a linear model fiction that we all accept without the appropriate level of yelling "Hey! Wait a minute!"
I'm tired of yelling; others need to step up.
All good fortune,
Chris
it is true that I can not recall a situation where I thought that any two amps or DACs of different model or manufacture did sound the same to me. But, that could always happen, it is possible.
This has been my personal experience.
I'm not the only one...
I am willing to check out and try "blind" testing too.
Happy to change my views to fit reality.
No problem saying that I was wrong, or have changed my mind.
You?
Let's review what you wrote:
"Why should we accept your constant toss off comments about NEVER hearing any components sounding the same when you discount any stories of the opposite happening."
I don't ask you to accept my "comments".
I have not discounted any "stories".
I did ask questions and ask for your position, and a list of parameters or gear that you are confident sounds all the same. To keep it brief. Same for all readers WRT this question. No takers thus far - if it was so simple and obvious the list would be hundreds of amps long... eh?
As of now the list is ZERO.
none.
not even two (?) amps cited in someone's "story".
This has been my personal experience.
I'm not the only one...
I am willing to check out and try "blind" testing too.
Happy to change my views to fit reality.
No problem saying that I was wrong, or have changed my mind.
You?
Let's review what you wrote:
"Why should we accept your constant toss off comments about NEVER hearing any components sounding the same when you discount any stories of the opposite happening."
I don't ask you to accept my "comments".
I have not discounted any "stories".
I did ask questions and ask for your position, and a list of parameters or gear that you are confident sounds all the same. To keep it brief. Same for all readers WRT this question. No takers thus far - if it was so simple and obvious the list would be hundreds of amps long... eh?
As of now the list is ZERO.
none.
not even two (?) amps cited in someone's "story".
Or focusing on this single parameter, one where we have fairly limited hearing sensitivity.
That s quite possible, with a higher room temp this will change the sound velocity and energy radiation capability (less matter/m3..) as well as our external ears characteristics, to summarize amps that heat the room will sound hotter...
Also as i pointed it loss of sensitivity (for a man..) is almost 40db at 8KHz at an age of 65, so whoever pretend that his old system sound as good as 30 years ago is just aknowledging that he doesnt hear the difference despite dozens of dB lower higher frequency levels.
You're brain is filling in the lost information because you have an expectation bias. Come back after a proper ABX to prove us otherwise.
It s quite possible that there s some compensation made by the brain, although i personaly dont know about this being wrong or right.
https://courses.physics.illinois.edu/phys406/lecture_notes/p406pom_lecture_notes/p406pom_lect5.pdf
Page 29 of the doc, we can see that women lose only 25dB at the same age for this frequency but lose more at lower ones, basically women have a more linear (in respect of frequency) loss than men.
40dB is not that much since the threshold of audibility of a sound can vary by 20db from a person to another, so if your sensitivity (at 8KHz) at 20 year is 10db better than average your actual loss will be only 30db (when you ll be 65..) in respect of the global average.
Anyway these stats are a prove that golden ears do not exist, or at least that they dont last and that variability over time is too high for someone to pretend that an amp sound not as good as an amp he listened to one year before...
Last edited:
I would like to clarify what happens to me in an ABX double-blind test. First, I do really lousy with very marginal sources and speakers. I just hear compromised sound, and small differences between amps just don't seem significant.
Second, I tend to lose track when listening to level matched electronics when both preamps (for example) are operating class A and the music is constantly changing in timbre. I can hear differences when I can keep track of which preamp I am listening to, but I lose track when an unknown (x) is presented for a short time as to which it is. That is all that I have ever said about ABX testing relative to myself, alone. I have been with others who DID keep track better than me. Golden ears? Probably, two of them were audio reviewers on occasion. So let's keep it straight, OK?
As far as audio quality differentiation, I doubt that it is really necessary to have extended hearing, if you have enough experience with listening to audio differences. That is why people over 60 can still discriminate between audio equipment. This frequency response thing is just something that somebody can measure, so they make out that it is more important than it is.
Second, I tend to lose track when listening to level matched electronics when both preamps (for example) are operating class A and the music is constantly changing in timbre. I can hear differences when I can keep track of which preamp I am listening to, but I lose track when an unknown (x) is presented for a short time as to which it is. That is all that I have ever said about ABX testing relative to myself, alone. I have been with others who DID keep track better than me. Golden ears? Probably, two of them were audio reviewers on occasion. So let's keep it straight, OK?
As far as audio quality differentiation, I doubt that it is really necessary to have extended hearing, if you have enough experience with listening to audio differences. That is why people over 60 can still discriminate between audio equipment. This frequency response thing is just something that somebody can measure, so they make out that it is more important than it is.
- Status
- Not open for further replies.