Sottomano said: And what are statements that zero global feedback amps are audibly worse because such amps have too high distortion/too high frequency deviation?
As far as I can see, nobody has. But hey, it's a nice sound byte!
Jan
As far as I can see, nobody has. But hey, it's a nice sound byte!
Jan
Huh? What's the faith of someone who takes pride and pleasure from designing amps with as low distortion as possible? Another of your strawmen?
Jan
Prove it 😉 Of course it depends what you mean by good, but nevermind. Toole went a long way to correlate measurements with perception, I don't believe no amplifier designer hasn't done the same, some are reluctant to share though, but this thread is about sharing, so come on all you shy designers out there, what are your secrets?
Just curious: do these amps sound identical to each other? Is it possible to distinguish them from a $200 Sony in a DBT?
How would I know? Surely you don't suppose that global nfb or not is all that determines how an amp sounds??
Jan
Jan
Eh no, you're only reading what you want to read. As jan.didden correctly pointed out, taking pride and pleasure from designing high fidelity amps has nothing to do with faith.
You also don't seem to realize that audiophilia is defined as enthusiasm about high-fidelity sound reproduction. Lowering distortion, increasing output power, and so on and on would be part of the hobby, not a distraction. But I'm not surprised you don't even understand what it is about. (See my previous response to you.)
Please prove the claim that there is no correlation between good measurements and sound. Keep in mind, however, that posting bias-ridden, subjective reviews without any controls and comparing them with random measurements will not support your point. Nor will comparing measurements of devices performing beyond the audibility of whatever metric you're looking at. To the contrary, it supports what I explained in my previous response to you.
I can see what you're trying to do, but all it does is backfire. With your posting you admit to a faith-based position, and at the same time failed to drag others down to the same level ... where you could beat them by pointing out that their position is just as absurd as yours?! (the forum still lacks a facepalm smiley)
Desperate behavior that is drowning in irony.
Given your position, there is no reason to take anything you say seriously anymore, as it is not based in anything tangible, objective or factual but your arbitrary beliefs.
If you want to contradict this then I will gladly present some hypothetical faith-based positions you could not refute regardless of the evidence you presented.
Last edited:
I can provide a reference to a peer-reviewed doctoral thesis that analyzes the reasons why an amplifier with low measured distortion may sound worse than one with what most here would consider horrendous distortion. But please first prove that there IS correlation between good measurements and sound.
Electroacoustics (Science and Engineering) and Psychology (Science) 🙂
More precisely psychoacoustics and physiology...
The participation of the brain in the perception of sounds (signals), speech and music.
Everyone hears differently, based on their own experience (learning)
More precisely psychoacoustics and physiology...
The participation of the brain in the perception of sounds (signals), speech and music.
Everyone hears differently, based on their own experience (learning)
Last edited:
My post should be interpreted with a strong element of humour/sarcasm. Youtube can claim whatever attracts views for more profit out of advertising, but reality will remain as it is. Feedback, as you claim, cannot be completely eliminated. Even a BJT's crystal suffers from natural parasitic negative feedback which cannot be eliminated. Both depletion layes act like insulators in a capacitor with the added complexity that such a capacitance is non linear. Vcb also affects the width of the base emitter depletion layer resulting in an output characteristic with non-horizontal output curves. For high frequency, there is a capacitance between the input/base and output/collector which has to be included for any accurate equivalent circuit representation.
"May" is proof by opposition. If there are observations contrary to the rule, the rule is incorrect. I know Toole, it is just one of many studies in this controversial area.
Agreed. The internal feedback in a triode tube from plate to grid is another example.
Jan
Transistors are non-linear devices. Feedback helps to improve and linearise them. Textbooks about electronics explain that in detail as this is one very important foundation of electronic circuits.
The definition of feedback is: a part of the output from a system reaching the input and adding to or subtracting from it. This is elementary control theory.
A degenerating resistor connected to the emitter has the output current component flowing into it and creating a voltage difference which subtracts from the signal voltage which is usually applied between ground and base. This is negative feedback.
The definition of feedback is: a part of the output from a system reaching the input and adding to or subtracting from it. This is elementary control theory.
A degenerating resistor connected to the emitter has the output current component flowing into it and creating a voltage difference which subtracts from the signal voltage which is usually applied between ground and base. This is negative feedback.
Last edited:
All good points.
It is instructive to read Arto Kolinummi's book. He designs some very linear and transparent amplifiers with no global fb. But the effort is huge, it is very hard and complex; hard enough for him to use it to earn a PhD from it!
Jan
It is instructive to read Arto Kolinummi's book. He designs some very linear and transparent amplifiers with no global fb. But the effort is huge, it is very hard and complex; hard enough for him to use it to earn a PhD from it!
Jan
Also, proving that there is and in fact has to be a correlation between amplification fidelity and fidelity of the reproduced sound is quite simple:
High-fidelity reproduction, in an idealistic sense, is producing sound that would correspond exactly to the amp's input signal. (I chose this starting point deliberately to not invite any even more off-topic discussions.)
As soon as you deviate from an ideal amp (and ideal speaker) you necessarily lower fidelity of reproduction by introducing an error. This is simple physics, cause and effect.
For example, an amp with higher 2nd order harmonic distortion will necessarily result in lower fidelity of reproduction than an amp with lower 2nd order harmonic distortion assuming no other variables were changed. In theory this is measurable down to the level where uncertainty becomes too big due to physical tolerances.
But in order to be able to actually perceive this difference the amp's distortion mustn't be masked by the speaker's distortion or the program material. Our hearing brings its own set of distortions and our perception is neither absolute nor objective, hence the need for scientific methods.
Note that I haven't yet said anything about how this distortion is perceived or interpreted. Ironically, if we limit our view to the amp, this only matters if the amplifier in question is low-fidelity enough to produce perceptible distortions. Necessarily, these electrical distortions need to be strong enough to not be masked by the (usually) strong distortions introduced by electromechanical transducers.
So look through these funny audiophile reviews where all kinds of different adjectives are thrown around trying to describe the amp's particular sound. This either means
1) the reviewers are not actually hearing the differences or
2) that they are, but then the reviewed amplifiers are necessarily FX boxes that introduce audible distortions.*
And there is no way around that.
*) Which, either way, the audiophile industry caters to and even aims for to make good money off of audiophiles, as I've explained before. So again, ironically, the arch enemy of the priests of audiophilia is actual high-fidelity and transparency.
High-fidelity reproduction, in an idealistic sense, is producing sound that would correspond exactly to the amp's input signal. (I chose this starting point deliberately to not invite any even more off-topic discussions.)
As soon as you deviate from an ideal amp (and ideal speaker) you necessarily lower fidelity of reproduction by introducing an error. This is simple physics, cause and effect.
For example, an amp with higher 2nd order harmonic distortion will necessarily result in lower fidelity of reproduction than an amp with lower 2nd order harmonic distortion assuming no other variables were changed. In theory this is measurable down to the level where uncertainty becomes too big due to physical tolerances.
But in order to be able to actually perceive this difference the amp's distortion mustn't be masked by the speaker's distortion or the program material. Our hearing brings its own set of distortions and our perception is neither absolute nor objective, hence the need for scientific methods.
Note that I haven't yet said anything about how this distortion is perceived or interpreted. Ironically, if we limit our view to the amp, this only matters if the amplifier in question is low-fidelity enough to produce perceptible distortions. Necessarily, these electrical distortions need to be strong enough to not be masked by the (usually) strong distortions introduced by electromechanical transducers.
So look through these funny audiophile reviews where all kinds of different adjectives are thrown around trying to describe the amp's particular sound. This either means
1) the reviewers are not actually hearing the differences or
2) that they are, but then the reviewed amplifiers are necessarily FX boxes that introduce audible distortions.*
And there is no way around that.
*) Which, either way, the audiophile industry caters to and even aims for to make good money off of audiophiles, as I've explained before. So again, ironically, the arch enemy of the priests of audiophilia is actual high-fidelity and transparency.
Last edited:
Yes, and the other thing is that listeners do not necessarily prefer amps with the lowest distortion.
Lots of amps that demonstrate high distortions have a lot of fans. SE tube amps, for example. The Pass AmpCamp amps.
The problem is trying to link facts and figures about distortion and other amp parameters to personal, infinitely variable listener preference. And that then complicated with the interaction between amps and speakers, and speakers and room. Never works out well.
Jan
Lots of amps that demonstrate high distortions have a lot of fans. SE tube amps, for example. The Pass AmpCamp amps.
The problem is trying to link facts and figures about distortion and other amp parameters to personal, infinitely variable listener preference. And that then complicated with the interaction between amps and speakers, and speakers and room. Never works out well.
Jan
And don't forget the music, there's a reason "girl and guitar" is preferred demo material for hi-end manufacturers.
"If you like to listen to simple music as played by an unaccompanied sackbutt, harmonic distortion figures might be perfectly relevant to you, but with a lot of music you will find that inter-modulation distortion becomes the elephant on the dance floor." No Prizes for guessing who
"If you like to listen to simple music as played by an unaccompanied sackbutt, harmonic distortion figures might be perfectly relevant to you, but with a lot of music you will find that inter-modulation distortion becomes the elephant on the dance floor." No Prizes for guessing who
Last edited:
Amen to both. But fidelity goes beyond measurable performance. There are certain subjective criteria that are straightforward and uncontroversial in listening tests, but are beyond available measurement. For example, stereo imaging, or ambience of a music hall. Faithful reproduction of micro detail. To assume that such subtleties are predicted by relatively crude measurements betrays lack of understanding of what high fidelity sound reproduction is about.
Are these mysterious things hit upon by trial and error or do designers know what they are and just aren't telling us or is it all part of the sales pitch? The OP is looking for ideas.