I hear what you're saying and partially agree
I think 16/88 would be my minimum spec for final product
Indeed. However designing by measurements only, without accompanying listening tests, may prove to be insufficient.
I agree Joshua. I first started my audio engineering career by mostly relying on measurements. Then, after a few 'sonic failures' even though I thought that I had done just about everything right, I became more open to new approaches to audio design, beyond OP amps, either discrete or IC based, and found that with some effort, I could design reasonably low distortion discrete designs that people agreed sounded pretty good too. This started with the Levinson JC-2 line stage, 40 years ago, now a staple in my arsenal, in a number of variations.
THD just for a low number, does not parallel subjective acceptance of a design, at least with the people that I design for.
THD just for a low number, does not parallel subjective acceptance of a design, at least with the people that I design for.
To my view, the spectrum of harmonics distortion is more important than the mere number of THD. See what Earl Geddes (Gedlee) wrote about it here http://www.diyaudio.com/forums/multi-way/100392-beyond-ariel-961.html in post #9610:
BTW, the fact that we, humans, are far more sensitive to higher harmonics distortion may also explain, at least in part, why some people enjoy LP's more than CD's, or SET amps. Possibly, it isn't because those people like the euphonic effect of high percentage of lower harmonics distortion, but because of the low percentage of higher harmonics distortion.
The FACT that THD without weighting the harmonics has been known for the last 75 years. Yes, in the 1941 'Radiotron Designer's Handbook' you will find a detailed summary of the orders of harmonic distortion and how THD, without harmonic separation is next to useless.
On the other hand, what is the problem when the THD is so low, it is almost impossible to measure it, let aside hear it. This is where we must realize that STATIC test signals measured by AM detection are not enough. In fact, the THD figures might be misleading as to the actual sound quality of a piece of audio electronics.
On the other hand, what is the problem when the THD is so low, it is almost impossible to measure it, let aside hear it. This is where we must realize that STATIC test signals measured by AM detection are not enough. In fact, the THD figures might be misleading as to the actual sound quality of a piece of audio electronics.
It's a temporal dislocation issue more than anything else. So one can call it tenth order harmonics, but more than anything, the key to understand is that is is a temporal/phase issue. Thus, rarely do I consider distortion in any form of engineering terms, but look at all of them as temporal displacement issues, for that is all that matters. It's one of those points where engineering speak obfuscates the reality from being seen, addressed, or worked properly with -by all but the highly observant.
Last edited:
Demian, I was talking about Human Hearing! 
Tinitus or ringing in the ears is a bad sign, perhaps indicating hearing damage, at the least temporary threshold shift.
The idea of turning up the speakers until they distort has for many been the desired volume level. With the loudspeakers I often use they really don't distort until well after levels that cause hearing damage if you are too close. (Under 10M.)
ES
Tinitus or ringing in the ears is a bad sign, perhaps indicating hearing damage, at the least temporary threshold shift.
The idea of turning up the speakers until they distort has for many been the desired volume level. With the loudspeakers I often use they really don't distort until well after levels that cause hearing damage if you are too close. (Under 10M.)
ES
Only a matter of phrasing to elucidate, or frame the point in mind. To do so, in that way, so it can be connected to the understanding that we hear, in harmonics, in the positive peak, in temporal-relational terms. That in the ear, we build a signal and decipher it, in the complex sense, over time.
To understand that harmonics of distortion may be relevant with respect to nomenclature in engineering terms, so that things may not be misunderstood, when expressed or communicated.
But to apply this back to what we hear, and how we hear - it is temporal displacement of these reflections and complex field originated (as relating to lattice structures and field integration with said polarized lattices) LCR complements/distortions, in level and placement, that are critical.
Just my way of looking at it.
To understand that harmonics of distortion may be relevant with respect to nomenclature in engineering terms, so that things may not be misunderstood, when expressed or communicated.
But to apply this back to what we hear, and how we hear - it is temporal displacement of these reflections and complex field originated (as relating to lattice structures and field integration with said polarized lattices) LCR complements/distortions, in level and placement, that are critical.
Just my way of looking at it.
Last edited:
Engineering speak may be aimed at clarity, but it is hobbled by having only one dance pattern, a lack of extra limbs and a severely myopic point of expression which tends to entrap the dogmatic mind in a ever shrinking circle.
I'm not baiting you, but you seem to be asking me to. I've no problems with engineering speak, but it does have very real limits. Moving in and out and through those limits is one of the ways to transcend said limits.
Volume is what separates the men from the boys, as regards hearing systems producing distortion. A system should be capable of running at elevated, realistic SPLs, because then one's ears can easily pick the duds - our hearing systems very rapidly adjust to momentary peaks of volume, in exactly the same way they do for live acoustic sounds, and in the inbetween, 'quiet' moments of the track all the disturbing, spurious distortion is exposed, in all its unpleasantness. If a system performs well in this test then the volume can be dropped right down, and subjectively the tonality does not alter ...
Come on, man. First get to Human Speak. Then get to Less-Limited Engineering Speak, if that is what is required. Then maybe even I will be able to understand what you might have to say. What you wrote seems quite obfuscatory, assuming that you did have something real to try to say.
I was with you on the timing-displacement variation versus frequency, if that is what you meant. And intuitively, it seems possible that some very small percentage of transient response inaccuracy (phase varying vs frequency, or as higher slew rates become less accurate?) could wreck imaging and maybe timbre accuracy and maybe some other stuff. I am not an expert at any of this; just trying to understand.
Maybe small variations in phase angle versus frequency could be seen with square wave testing, where they could appear to look like ringing, since the Fourier components wouldn't quite sum the way they should and thus the edges would be improperly constructed. Anyway, it seems reasonable to expect that the phase errors could also show themselves in THD components.
Please further-explain the lattice ideas.
maybe related to monoatomic gold??
Ah, for the olden days. The big slap upside everyone's head is still just around the corner.
- Status
- Not open for further replies.