There is potential in EQ for speakers still, but it requires a lot of DSP power and accurate measurement
https://www.klippel.de/fileadmin/_m...ON_OF_NONLINEAR_LOUDSPEAKER_DISTORTION_02.pdf
https://www.klippel.de/fileadmin/_m...ON_OF_NONLINEAR_LOUDSPEAKER_DISTORTION_02.pdf
Last edited:
I am very suprised that the debate of global feedback still is on today. If we want to get the high fidelity reproduction of music recordings, high global feedback is necessary. It has been proved technically for over 70 years.
But the other problem is appeared , "does the music recordings represent the correct acoustic signal of the performance". In this case, the answer is complex. Many recordings (both modern and from the analog era) are really excellent, and they benefit greatly from high fidelity reproduction. But many other recordings is not up to task, and the high fidelity reproduction will show all the good, the bad and the ugly of the recording, and it is not that pleasant to listen to. Even so many modern recordings are tainted with the excessive compression, auto-tune,... or even some weird taste of sound from music producers and artists. So the high fidelity of music reproduction is not always desirable.
The problem is more serious when you see the demography of potential customer for audio product. There are a large chunk of them are in China, Japan, South Korea, Southeast Asia, India. These people do not only listen to Western Music, they also listen to their own music, in many cases were recorded so poorly when their country is not as developed as today. And they want a good system to listen to both, their own country music and Western Music.
When we have a poor recorded music, the well-controlled distortion (reasonable amount of distortion with appropriate phase, no high order distortion) can be very useful to tame the unpleasant sound. It is a trade-off, of course. It exchanges some clarity/detail for smoother sound, and for many people, it is better. The market segment for "high-end audio" is based on that. And to achieve that well-controlled distortion, the low feedback, simple circuit or tube design is very good choice. That why many people still think low feedback is better than high feedback. And many audio manufactures still spread that to marketing their product. But in fact, low feedback is only the tool to achieve a good compromise for music reproduction because of the inadequate quality of music recordings
But the other problem is appeared , "does the music recordings represent the correct acoustic signal of the performance". In this case, the answer is complex. Many recordings (both modern and from the analog era) are really excellent, and they benefit greatly from high fidelity reproduction. But many other recordings is not up to task, and the high fidelity reproduction will show all the good, the bad and the ugly of the recording, and it is not that pleasant to listen to. Even so many modern recordings are tainted with the excessive compression, auto-tune,... or even some weird taste of sound from music producers and artists. So the high fidelity of music reproduction is not always desirable.
The problem is more serious when you see the demography of potential customer for audio product. There are a large chunk of them are in China, Japan, South Korea, Southeast Asia, India. These people do not only listen to Western Music, they also listen to their own music, in many cases were recorded so poorly when their country is not as developed as today. And they want a good system to listen to both, their own country music and Western Music.
When we have a poor recorded music, the well-controlled distortion (reasonable amount of distortion with appropriate phase, no high order distortion) can be very useful to tame the unpleasant sound. It is a trade-off, of course. It exchanges some clarity/detail for smoother sound, and for many people, it is better. The market segment for "high-end audio" is based on that. And to achieve that well-controlled distortion, the low feedback, simple circuit or tube design is very good choice. That why many people still think low feedback is better than high feedback. And many audio manufactures still spread that to marketing their product. But in fact, low feedback is only the tool to achieve a good compromise for music reproduction because of the inadequate quality of music recordings
Last edited:
That would indeed be nice, if we could move forward there. But then we are barking up the totally wrong three. I don't think there is any possibility to make a step change in reproduction quality by working on power amps or feedback, because this has stopped being the critical link in the reproduction chain for many decades.
Seems to me that the area where steps have to be made are in the acoustics field, speakers, listening rooms, equalization, that sort of things.
Jan
Proved? Necessary? Where is the proof?
The question how technical performance is beneficial for sound reproduction has been raised many times in this thread. Apparently, no one has an answer to this since DIN45500. So, why not stick to that? Why fantasize that if one zero is great, ten are much better? What else is this than a fantasy?
Those 70 years are mostly walking backwards as sound quality of amplifiers is concerned.
Things start well with triodes, but suddenly the WAF rears its ugly had and speakers have to be small and cute and accommodating plants on top 🙂. So, goodbye triodes, hello pentodes with many watts. Unfortunately, these need global NFB to work in a circuit which controls voltage. The nagging doubts begin (mid 40s?) and persist until today. It is mostly a clash between listeners and engineers striving for more zeros and cheaper watts.
This ugliness escalates in the mid 70s when a young generation of untypically "hearing" engineers realise that minimizing thd is a down falling spiral.
Suddenly, there are a whole lot of new issues, all brought up by GNFB, that need addressing. Tims, slewing distortions, memory, etc. Yes, this research was invaluable but quickly forgotten by the main stream engineers who never cared one bit there was human consumption involved at the end of the chain. The research gave birth to a number of still respected transistor amps with very average distortion measurements.
The one conclusion i can draw from this statement is that you have a very limited view on what is available in today's high end. A large number of systems have speed, clarity, detail and dynamics which appear almost unbelievable. No attempt at smoothing things over at all, on the opposite. I can give dozens of examples but the first coming to mind is a full Aries Cerat system.
Those 100 years are steadily walking backwards as sound quality of amplifiers is concerned. As a consequence of the triumphant technological "development", what is obtainable today is cheaply produced junk.
@atmasphere
Setting up DBT tests to ascertain what speaker sound consumers want is an objective assessment of preference. Based on tnelutput of the tests, design specifications can be tweaked.
Claiming feedback amplifiers don’t sound as good as non- feedback amplifiers without taking the trouble to conduct a similar test or have corroborating data is the thing I am referring to.
Setting up DBT tests to ascertain what speaker sound consumers want is an objective assessment of preference. Based on tnelutput of the tests, design specifications can be tweaked.
Claiming feedback amplifiers don’t sound as good as non- feedback amplifiers without taking the trouble to conduct a similar test or have corroborating data is the thing I am referring to.
I’ve heard 1940’s vintage SET and push pull amplifier designs.
No thank you.
We are living in the goLden age of audio. Let’s at least recognize that.
No thank you.
We are living in the goLden age of audio. Let’s at least recognize that.
Have you heard them done with good or modern components? Compared to the "improved" pentode/ultralinear version with nfb. And on easy speakers? Such a comparison is not so easy to do right as it may appear.
As living in the Golden Age of audio - i cannot agree more strongly. The world, being a much richer place than even 10 years ago, lets all kind of nearly impossible audio fantasies take a material expression. It is truly fantastic and inspiring. It is therefore even more surprising that many participants on this forum defend so stubbornly old engineering dogma.
Yeah right. It's about time we replace Ohms law with something 21-st century. I mean, V=I*R is sooo old school ...
BTW, I know you are just making this up, but on the chance you come through: pray tell, which dogma is being defended?
Jan
Yes and no, it seems for the most part audio quality is overlooked in the media, as an extreme example prevalent at the moment, I find it unbearable trying to listen to people on the TV speaking into their lap top microphones, I bet most people couldn't care less.
Thank you for asking an easy question.
The dogma that the maximum amount of global NFB should be applied in the pursuit of ultimate quality.
The dogma that moderate amount of nfb is worse than none.
The dogma that NFB is a perfect solution and not only another tool which to needs to be used carefully and sparingly and only when required.
The dogma that global nfb amps cannot achieve SOA sound quality in conjunction with sympathetic speakers.
Gee, didn't realise those were so many, should have numbered them 🙂
I also propose a useful technical term "a dogma boomer" 😀
The dogma that the maximum amount of global NFB should be applied in the pursuit of ultimate quality.
The dogma that moderate amount of nfb is worse than none.
The dogma that NFB is a perfect solution and not only another tool which to needs to be used carefully and sparingly and only when required.
The dogma that global nfb amps cannot achieve SOA sound quality in conjunction with sympathetic speakers.
Gee, didn't realise those were so many, should have numbered them 🙂
I also propose a useful technical term "a dogma boomer" 😀
Last edited:
I suppose it's corona virus time, end nothing useful to do but useless discussion with entrenched anti GNFB cohorts.
Or the dogma "every single part contributes to sound" proclaimed by all these parts rollers.
No dogma. If you define quality as the most transparent and linear amplifier you can, it is simply a technical fact.
This can be and has been technically demonstrated as fact. See the various writings of Self, Cordell, and earlier Baxandall.
This thread has discussed the fact that this is NOT the case, and I didn't see anybody disagree.
You just made that up, right? If you think there's anything true here, where did you get that then?
So, you may think you were replying to an easy questions, but instead it shows a mix of misunderstanding and what looks like made-up stuff.
Jan
I know it is impossible to reason someone out of a position they have not arrived at by reasoning. But there's a lot lurkers here, and they should get the facts and figures.
Jan
I guess you mean the conversion of second-order distortion to higher-order distortion by negative feedback; the total distortion goes down, but the harmonic content also changes. That was already shown in L. I. Farren's 1938 paper, https://linearaudio.nl/sites/linearaudio.net/files/Farren feedback1938.pdf , and he referred to an even older article by Richard Feldtkeller in Telegraphen- und Fernsprechtechnik, vol. 25, page 217.