False. Once again you don't know the matter about you're speaking. And you, plainly, don't know how feedback really work. Your knowlege of fact is purely "audio ideological", not really based on engineering experience.
Piercarlo
PS - I imagine that you think about electronic engineers (especially those designing audio appliances) as stupid monkeys severely jammed in brain for not being able to grasp your "subjective poor quality and high (subjective?) distortion, even if it´s beyond measurement boundaries, no matter how many chapters are written in support of the opposite".
But pheraps the real question is that "subjective quality" is not a measurable entity (even if it would exist out of your mind and your imagination), otherwise it immediately ceased to be "subjective" for becoming "objective", hence searchable, investigable and measurable.
Because this DON'T happen, what should think anyone about this "mistery?" That's playng an awful "conjure of silence" as someone thoughed many years ago about UFOs? Or maybe phreraps that "audio subjective quality" (poor or not poor as you prefer it) is, out of one's personal beliefs, a technical "non-entity" which cannot be measured because, physically, there's nothing to be meaured?
That you have observed at least is not "a subjective poor quality..." etc. etc. but a known fact for which exist (from long time ago) a simple explain: harshing of amplifier sound start exactly from the point where, falling of NFB loop gain, drive the intrinsic distortion of circuit above noise floor that, in the same time, is worsened by increasing high frequency contributes from power supply due to the lowering of SVR ratio. On this topic may be written a well thick book with plenty of measures, data, conjectures that are anything but not subjective.
In fact the reason for worsening of sound of the many amplifiers are quite more earth bound than usually subjectivist think about. There are no problem of "timing", switching and so (or at least NOT in audible frequency band; in the region of Long and Middle Waves may be... if you can "naturally" heard those frequencies!); the most is cause by plain nonlinearities that become inefficiently neutralized by falling NFB (unavoidable if stability of circuits is to be preserverd).
The subject is a really vast field for inquiries (especially detailed ones) but none of the topics involved required any "alchemical" reasoning about Ft, switching times and so on but rather a plain technical and possibly CALM reasoning!
Hi
Piercarlo
Try to design speakers with the same level of perfectionism...
What is the Ft of a paper cone? And how about a wooden enclosure?
When I think about amplifier optimization I never forget that speakers typically exhibit many +/-3dB peaks and dips across their passband (or worse) and speaker THD is hardly below 1% at moderate to high listening levels. You have to know when you should give up while trying to improve amplifiers.
Amplifiers may be fun but speakers and acoustics are really interesting challenges.
What is the Ft of a paper cone? And how about a wooden enclosure?
When I think about amplifier optimization I never forget that speakers typically exhibit many +/-3dB peaks and dips across their passband (or worse) and speaker THD is hardly below 1% at moderate to high listening levels. You have to know when you should give up while trying to improve amplifiers.
Amplifiers may be fun but speakers and acoustics are really interesting challenges.
This is all very true (distortion figures can be much lower but also much higher and THD is NOT the only kind of distortion that matters for the loudspeakers). Anyway even the best speaekrs are still the weak link of the chain.
But, for reasons that i cannot fully understand little changes in the amplifier make a huge impact on the sound.
The most exciting thing is to build together both the amplifier and the speaker that it will drive, making optimal synergistic choices
I've written something. In Italian of course.
Elettronica audio - (a cura di Piercarlo Boletti): LA CONTROREAZIONE - 1A
Take a friend that knows italian and read it. Or wait for when i'll found the time for translating it (not in the near future however).
Piercarlo
PS - In the blog there are, in italian again, one two parts article about QUAD 405 that explain some thing also about feedforward (that was invented by same man that invent feedback, Harold. S. Black)
Last edited:
100% correct, welcome to the club.
BTW, you are wasting your time.
With Lumba Ogir Surely. But my aim is to provide some piece of knowledge to interested and really passionate dyers. Lumba Ogir by itself count for nothing.
Hi
Piercarlo
Question: if really NFB don't better anything respect to open loop transfer function of amplifiers, why it's so largely used? Just for the dubious pleasure of complicating uselessly our lives?
Hi
Piercarlo
Piercarlo,
thank you, the translation is done, seems to be well-written, I`m still reading it. Let's see if there is something of value.
thank you, the translation is done, seems to be well-written, I`m still reading it. Let's see if there is something of value.
Subjective judgment is not a measurable or quantifiable, taste is a matter of feeling, like everything in aesthetics, plodding away with that is naive and absurd, just a waste of time.
In other words they CORRELATE data and impressions. A thing often misunderstood is that for doing a good work it need a good knowledge about EITHER sides of problems: earing AND measuring, not earing OR measuring. A "third side" may be a strong will in searching the most simple explanation of data and phenomena without falling in simplicism often masked as "evident" things that really are not evident at all - An art by itself...
I agree totally with your post.
Hi
Piercarlo
My hope is that you may be stimulated to start a search for a deeper knowledge of the matter. Problems (serious problems) about NFB exist, especially those derived from its improper use; but are not those usually claimed about by many audiophiles.
Hi
Piercarlo
I agree. It has long been fashionable in some parts of the audiophile community to dis negative feedback. There are often many intuitive theories put forth as to how it creates more distortion at higher orders, but when these theories are tested and measured/simulated they are found to be lacking. There is widespread mis-understanding of Baxendal's results, for example. Much of this dis-ing of NFB came about as a result of the TIM discussions in the 1970's. Many designers were naive back then and did not understand slew rate issues. Many solid state amplifiers just sounded bad because the technology was young and output transistors were poor. Feedback to some extent received guilt by association.
When NFB is properly used by good designers, the results are very good, both measurement-wise and sonically.
Having a clean open loop amplifier with good dynamic range is important to achieving good results with NFB. Those who paper over problems in the open loop with NFB will still have a poor amplifier after they close the loop. Interestingly, no-NFB amplifier designs virtually force the designer to do the right thing in the open loop, else they have a really terrible design.
Happy New Year,
Bob
