|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
32.210.164.98
In Reply to: RE: Audio Science Review posted by E-Stat on October 19, 2019 at 06:29:16
Yeah, but...
In a simple harmonic distortion test, there is one tone and the harmonics are by nature correlated. For a bunch of other reasons, a simple harmonic test doesn't tell you much.
In an IMD test, it is possible to correlate the tones, especially if you have enough of them.
Whether the tones are correlated to each other or not is only a detail that isn't the crux of the situation.
The critical point, I think anyway, is that in actual systems that try to reproduce many tones, the power in all the intermodulation distortion products just completely swamps any simple harmonics. That is true in a home audio system as much as it is in a modern CATV or wireless transmission system. This was also described in great detail in the middle of the 20th century or thereabouts. Guys at Bell Labs and the BBC churned out papers on the subject.
Pretending that this doesn't matter is only fooling oneself.
In addition, the audio world tends to be a too blasé about interfering signals that are running around the system through the power line connections and so on. The general idea is that "if it's out of the audio band, you can't hear it", etc. Whether you can hear the stuff outside the band isn't the important part. These signals can and will mix (mix in the mathematical sense, not the usual audio engineering term) with other signals to create yet more distortion products within the audio band.
This problem has gotten worse over time with digital audio becoming a big part of home systems plus the garbage spread everywhere by various microprocessor based computing systems like you might find in your alarm clock and by poorly filtered switching power supplies.
The usual response to a comment like this is that the feedback system found in most amplifying devices cleans that right up. Not so! In order for the feedback to work you need adequate open loop gain at the frequencies you're trying to reject for the feedback to work. Guess what! The open loop gain drops dramatically in most amplifying devices aimed for audio use outside the audio band. So, this rejection mechanism doesn't work. This isn't some secret mystery either - people just choose to ignore it.
Anyway, I've gotten way off topic and have probably pissed off most people who had the bad fortune to stumble into this thread. So, fini.
Follow Ups:
Pretending that this doesn't matter is only fooling oneself.
Your welcome to your opinion, but I find zero correlation with THD or IMD telling you anything valuable as to how an audio component sounds. You need not present any more logical fallacies.
The usual response to a comment like this is that the feedback system found in most amplifying devices cleans that right up. Not so! In order for the feedback to work you need adequate open loop gain at the frequencies you're trying to reject for the feedback to work. Guess what! The open loop gain drops dramatically in most amplifying devices aimed for audio use outside the audio band. So, this rejection mechanism doesn't work.
Agree 100%. As does Nelson Pass . Which is why I'm not a fan of op amps which require ~40 db of corrective feedback due to their large open loop gain. Great numbers. Mediocre sound.
First, please point to where I suggested in any way that the existing tests have been shown to correlate to musical enjoyment, realism, or whatever floats your boat. In fact, I specifically said that was not the case.
Second, if you don't believe that corruption of the original musical waveform - which was in fact variations in air pressure - through noise and distortion in the electrical transduction of same is the cause for inaccurate or unsatisfying musical reproduction, just what is the cause? Evil leprechauns casting spells over the listener?
And, although I am not a fan of opamps in audio, I think you misunderstand just how feedback works. Or, you didn't spell it out right. Open loop gain is not what needs to be corrected - that can be done with a resistor or two. Large open loop gain is what allows feedback to work in amplifiers. It is part of the correction mechanism.
Maybe feedback has been used as a magical cure-all where it's not. People who design and build open loop amplifiers might specifically address problems that designers who've used feedback have chosen to ignore. Some pretty smart people believe that feedback has not been applied suitably in audio products. They could be right - I try to keep an open mind for the possibilities.
I wasn't going to continue here, and will not after this post, but since you chose to throw derogatory terms in my direction, I felt I should reply so that everybody is clear on what I am saying. You can agree or disagree with me if you like, even in a public forum such as this one. You can call me stupid, ignorant, and whatever else you want, too, but please don't twist around what I said or did not say.
despite what appears to be now a tongue in cheek response which began this exchange.
but since you chose to throw derogatory terms in my direction...
The logical fallacy you employed is Argument from Authority . Unnecessary.
Large open loop gain is what allows feedback to work in amplifiers.
Conversely, inherently linear designs that don't require corrective feedback need not possess such.
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: