Grey,
It seems we at least agree that some high-feedback amps create a different image than some less linear (in my vocabulary 'less transparent', but YMMV) amplifiers.
The next step is then what is more 'realistic'. If I read you correctly, you are of the opinion that the less feedback, less transparent amp creates a more realistic image. That's not my view; in effect, my view is that neither amp will create a realistic image. It's a waste of time to try to find that elusive amp that re-creates the sound field, the image, of a live concert. I would argue that a reasonable amplifier is much better in re-creating the sound performance of a life concert because it creates a much more stable image, more resolution, better sound stage and what have you. {Actually that is not really correct; I should say that the CD or DVD recording does a better job; the amp, if well designed, just reproduces what's on that CD or DVD, no more, no less}. What that CD/DVD and amp combo fails miserably at is of course the recreation of the ambience, the feeling of *being there* which is mission impossible for any system that merely can create sounds.
So, I am a bit at a loss when you say that you feel that the SE/non-linear amp recreates a live concert better than a more linear, high feedback amp. However you slice the sausage, a perceived difference in sound MUST be caused by a difference of the electrical signal at the amp output terminals. It is true that the SE amp creates spectrum elements, additional overtones if you will, that are absent in the high feedback amp. In what way would additonal overtones, or harmonics, give the impression of more life like reproduction?
Jan Didden
It seems we at least agree that some high-feedback amps create a different image than some less linear (in my vocabulary 'less transparent', but YMMV) amplifiers.
The next step is then what is more 'realistic'. If I read you correctly, you are of the opinion that the less feedback, less transparent amp creates a more realistic image. That's not my view; in effect, my view is that neither amp will create a realistic image. It's a waste of time to try to find that elusive amp that re-creates the sound field, the image, of a live concert. I would argue that a reasonable amplifier is much better in re-creating the sound performance of a life concert because it creates a much more stable image, more resolution, better sound stage and what have you. {Actually that is not really correct; I should say that the CD or DVD recording does a better job; the amp, if well designed, just reproduces what's on that CD or DVD, no more, no less}. What that CD/DVD and amp combo fails miserably at is of course the recreation of the ambience, the feeling of *being there* which is mission impossible for any system that merely can create sounds.
So, I am a bit at a loss when you say that you feel that the SE/non-linear amp recreates a live concert better than a more linear, high feedback amp. However you slice the sausage, a perceived difference in sound MUST be caused by a difference of the electrical signal at the amp output terminals. It is true that the SE amp creates spectrum elements, additional overtones if you will, that are absent in the high feedback amp. In what way would additonal overtones, or harmonics, give the impression of more life like reproduction?
Jan Didden
Jan wrote:
Hi Jan,
What exactly do you mean by "more linear"?
My next question will be: how do you measure that the high feedback circuit is "more linear"?
Hi Jan,
What exactly do you mean by "more linear"?
My next question will be: how do you measure that the high feedback circuit is "more linear"?
traderbam said:
Sigh. I should have said: more transparent. Or whatever. When I say 'more linear' in this context I mean that the output signal of the amp deviates less from the input signal than in a 'less linear' amp. In a less linear amp, the output signal deviates more from the input signal than in a more linear amp. It adds spectral components to the input signal that the linear, high feedback amp, does (much) less. I can measure this difference between a less or more linear amp by measuring the difference between input and output signal in each case. In the more linear feedback amp I will see that the output signal is a more faithfull reproduction of the input signal than in the less linear amp.
I am aware that measurements necessarily are not done with typical musical signals. It is my view that if an amp measures more linear in this context than another one using the usual battery of tests available to the engineer, it will also be more linear when reproducing music. If you don't agree I am looking forward to some proof that this is not the case. And to save you the trouble, anecdotes like "I myself heard that is was different" are not accepted by me.
To save you even more effort, if your next post again tries to pick apart my use of words rather than addressing the factual contents, I shall refrain from addressing THAT. I think I did my best to ask for a factual discussion rather than word splitting; I also think I tried to lead by example. Your choice.
Jan Didden
Jan, a persons observation is a fact. It is the most important fact when it comes to designing a device intended to optimise a persons observation.
janneman said:
You have me at a disadvantage, Jan, as I cannot think of a single post where I have ever mentioned SE (by which I assume you mean single-ended, and by extension single-ended tube?) amps. If, indeed, that is what you mean, then I would like to state for the record that I have never heard an SET amp anywhere, under any circumstances, and can only say that I have no opinion as to their sound, either as individual pieces or as a group.
Nor have I heard any of the current crop of full-range speakers, so likewise I can only say that I have no opinion.
(Which are far more honest statements than most here can make. They condemn without listening. Is it any wonder that they haven't discovered that the Emperor has no clothing?)
Any comments I have made relating to tubed equipment have been in relation to preamps (almost exclusively triodes--the usual suspects, 12AX7, 6922, 6SN7...) or to amps with transformer-coupled pentode output stages (mostly 6550s), some Ultralinear, some not. If anyone can find a post that I have made relating to SETs, I'd be most grateful, as perhaps then I'd be able to get better traction on some of these posts. As it is, I feel like I'm wrestling a greased pig--every time I think I've got a grip on the beast, it slithers from my grasp.
Yes, you read my intention correctly, in that I feel that some semblance of a live image is possible. In fact, I've heard quite a few amps (and preamps) that do a credible job. They create at least as close an approximation of image as they do of instrumental tone, dynamics, etc.--take that statement at face value, nothing more, nothing less. You appear to feel that it is impossible. How to move forward from here, I'm not clear.
I'm not trying to be evil about it or anything, but is there even the ghost of a chance that the reason you aren't hearing a reasonable approximation of an image due to high feedback?
I'm also a little curious about your assertion that I said that a 'non-linear' amp recreates music better than a more 'linear' amp. Do you have some quibble with the specs I quoted above? Given that I've often seen the .1% distortion level bandied about as being the limit of human perception, I would think that anything below that should be tolerable. Surely the .5Mhz bandwidth is acceptable? I haven't measured the damping factor, nor have I bothered to try to calculate it (in part because I haven't decided for sure whether I'm going to use 2 pairs of MOSFETs in the output or 3 and lacking feedback, that will make a big difference in the damping factor), so perhaps that's what bothering you.
Just how non-linear do you think this critter is, anyway, and in what ways?
If you try to argue that the distortion should be lower, then you come up against the "scientifically proven" limits of human perception. (For what it's worth, I fiddle with everything I build until it is below .1% THD--tubes in the past, solid state nowadays.) If you try to argue that the bandwidth should be wider, I'll concede the point only if you advance Spectral, who run on out to something like 1Mhz, albeit by using feedback (I have no idea how much--never seen a schematic of their stuff nor has anyone ever talked about it around me). If you care to grumble about my damping factor, I'll roll over and play dead, except to say that you should hear a Premier One in full cry before trying to claim that a low damping factor is necessary for tight bass.
Back to you.
jcx,
Congratulations! You're well on your way to proving that an elephant's trunk is a snake. Keep up the good work. While you're at it, might I suggest that you prove that a sow's ear is a silk purse?
Grey
GR, SE can also refer to single-ended amps of the silicon variety (bipolar or MOSFET), which I think was what Jan was talking about.
so Grey denies engineering systems analysis or any modern technical/scientific field can tell anyone anything at all - because there is always some "other" domain not captured by our math/knowledge or philosophic viewpoint? We have to give up Math because of Godel’s Incompleteness theorem, can’t use any Classical system such as Electromagnetism, Newtonian Physics, continuous media based Acoustics in “real world”, “human scale” design because of quantum mechanics, General Relativity? – given these huge “Scientific failures/invalidations” obviously more such revolutions will render anything we can articulate today false? – I don’t think so
I also don't understand Traderbam's "questions" – am I expected to present the subject matter of 1/2 a dozen or so college courses with titles like Signals and Systems, System Identification, Nonlinear Systems Analysis, in addition to the 1st 2 yrs of undergrad preparatory material in Calculus, DiffEq?
If people have the background why aren’t they applying that knowledge base and logic skills to advance a discussion of the properties of feedback in amplifiers – which I think is still the thread topic?
Or is everyone going to continue to treat the thread as a Rorschach test and trot out every straw man that has been tossed around in these discussions for the 25 years I’ve been watching audio “subjective” vs “engineering” exchanges – without reference to the progress in practical psychoacoustics , engineering knowledge and tool resolution that has occurred along the way?
I also don't understand Traderbam's "questions" – am I expected to present the subject matter of 1/2 a dozen or so college courses with titles like Signals and Systems, System Identification, Nonlinear Systems Analysis, in addition to the 1st 2 yrs of undergrad preparatory material in Calculus, DiffEq?
If people have the background why aren’t they applying that knowledge base and logic skills to advance a discussion of the properties of feedback in amplifiers – which I think is still the thread topic?
Or is everyone going to continue to treat the thread as a Rorschach test and trot out every straw man that has been tossed around in these discussions for the 25 years I’ve been watching audio “subjective” vs “engineering” exchanges – without reference to the progress in practical psychoacoustics , engineering knowledge and tool resolution that has occurred along the way?
This must b e the 20th thread over th e last 2 years on f/back vs non-f/back and we still have not got to a point where there is any agreement on the issue.
I think the core of the issue for the engineering fraternity that makes use of feedback (or error correction) techniques is:-
Well applied feedback enhances the electrical performance of an amp.
Badly aplied feedback can lead to performance problems (instability etc). Take care of the P + Z's, phase margin etc
The differences in sound between amplifiers cannot be put down to the use/non-use of feedback. There are far too many variables to make this statement and it is far too generic to even begin to have a sensible discussion
Very fine sounding amplifiers, some with feedback and some sans f/back, have been produced - so on this argument alone, no-one can claim feedback 'sucks th e life out of th e music'. (acknowledgments Bob Cordell)
Critically, sound is a personal thing - what Charles Hansen and John curl find great may not be what someone else enjoys. BTW, no one in the feedback camp contests the fact that either of these two gentlemen have created some fine sounding amplifiers. But then so has Dan Krell. Read the Audiophile Awards Issue to get a flavour for how many feedback amps get good reviews from subjective analysis
Finally, I dont think anyone from the feedback camp has said amplifiers without feedback sound crap, or conversely, for an amplifier to sound good, you have to apply feedback.
Our beef: The generic claim that the use of feedback has a deleterious effect on the sound of an amplifier.
Our Generic position: We use f/back sensibly to improve the electrical performance of an amplifier
Our view on f/back and sound: F/back may indeed alter the sonic signature of an amplifer, but it is a leap of faith to state that it makes it sound worse UNLESS it has been misused.
I think the core of the issue for the engineering fraternity that makes use of feedback (or error correction) techniques is:-
Well applied feedback enhances the electrical performance of an amp.
Badly aplied feedback can lead to performance problems (instability etc). Take care of the P + Z's, phase margin etc
The differences in sound between amplifiers cannot be put down to the use/non-use of feedback. There are far too many variables to make this statement and it is far too generic to even begin to have a sensible discussion
Very fine sounding amplifiers, some with feedback and some sans f/back, have been produced - so on this argument alone, no-one can claim feedback 'sucks th e life out of th e music'. (acknowledgments Bob Cordell)
Critically, sound is a personal thing - what Charles Hansen and John curl find great may not be what someone else enjoys. BTW, no one in the feedback camp contests the fact that either of these two gentlemen have created some fine sounding amplifiers. But then so has Dan Krell. Read the Audiophile Awards Issue to get a flavour for how many feedback amps get good reviews from subjective analysis
Finally, I dont think anyone from the feedback camp has said amplifiers without feedback sound crap, or conversely, for an amplifier to sound good, you have to apply feedback.
Our beef: The generic claim that the use of feedback has a deleterious effect on the sound of an amplifier.
Our Generic position: We use f/back sensibly to improve the electrical performance of an amplifier
Our view on f/back and sound: F/back may indeed alter the sonic signature of an amplifer, but it is a leap of faith to state that it makes it sound worse UNLESS it has been misused.
Bonsai, your 'conclusions' have been well known for many decades. It doesn't make them exactly accurate.
Charles and I design amps and preamps in an open marketplace where we HAVE to please the people who want to have an exceptional amplifier at whatever price point we have designed at. WE don't chose whether we have made an exceptional product, others, usually sophisticated listeners do that. We only pursue what works.
We try to impart a little of our design philosophy to people who might want to know our direction in audio design.
Of course, we knew all about negative feedback decades ago, and believed just as strongly as you do, that it was a wonderful cure-all for all amps.
Charles and I design amps and preamps in an open marketplace where we HAVE to please the people who want to have an exceptional amplifier at whatever price point we have designed at. WE don't chose whether we have made an exceptional product, others, usually sophisticated listeners do that. We only pursue what works.
We try to impart a little of our design philosophy to people who might want to know our direction in audio design.
Of course, we knew all about negative feedback decades ago, and believed just as strongly as you do, that it was a wonderful cure-all for all amps.
GRollins said:
Hi Grey,
I didn't mean to say that you said you favor SE amps. I used SE as an example of non-linear amps that demonstably add to the signal, not being 'transparent' as I see it. I might have been cutting too many corners at once; I was only trying to make clear the contrast between a typical non-linear amp like a SE unit (tube or ss) and a typical linear amp like a high-feedback design.
If I understand you correctly now you are comparing a high feedback amp to another, non-feedback amp, where both are linear in the sense that neither adds anything to the signal (or at least both are at the same very low level of adding anything). In this situation, you say that the non-feedback amp gives a more realistic reproduction. Am I with you so far?
Jan Didden
traderbam said:
Absolutely. But is this discussion not about using one's *personal* observations to decide how to optimise *another* persons experience?
If you or Grey say that you really like to listen to amp xyz because it sounds so realistic, fine. Great.
But if you say that I should also use this amp to get realistic reproduction, I'd like *some* warm feeling that indeed I will experience better realism if I would follow your advice. Convince me!
Jan Didden
John curl wrote:-
Bonsai, your 'conclusions' have been well known for many decades. It doesn't make them exactly accurate.
Charles and I design amps and preamps in an open marketplace where we HAVE to please the people who want to have an exceptional amplifier at whatever price point we have designed at. WE don't chose whether we have made an exceptional product, others, usually sophisticated listeners do that. We only pursue what works.
We try to impart a little of our design philosophy to people who might want to know our direction in audio design.
Of course, we knew all about negative feedback decades ago, and believed just as strongly as you do, that it was a wonderful cure-all for all amps.
John, where did I state or imply that f/back was a cure all for all amps?
Secondly, yes, you do please sophisticated listeners . . . . . but so do the guys who use feedback! So where does that put the generic proposition that f'eedback is bad'?
Let me give you some feedback: the whole 'feedback is a bad thing' is built on (dirty) sand.
There'snothing wrong with zero feedback amps, but therefoes nothing wrong with feedback amps as well. In both cases, careless engineering will give bad results.
Thank you Jan.
Bonsai, your 'conclusions' have been well known for many decades. It doesn't make them exactly accurate.
Charles and I design amps and preamps in an open marketplace where we HAVE to please the people who want to have an exceptional amplifier at whatever price point we have designed at. WE don't chose whether we have made an exceptional product, others, usually sophisticated listeners do that. We only pursue what works.
We try to impart a little of our design philosophy to people who might want to know our direction in audio design.
Of course, we knew all about negative feedback decades ago, and believed just as strongly as you do, that it was a wonderful cure-all for all amps.
John, where did I state or imply that f/back was a cure all for all amps?
Secondly, yes, you do please sophisticated listeners . . . . . but so do the guys who use feedback! So where does that put the generic proposition that f'eedback is bad'?
Let me give you some feedback: the whole 'feedback is a bad thing' is built on (dirty) sand.
There'snothing wrong with zero feedback amps, but therefoes nothing wrong with feedback amps as well. In both cases, careless engineering will give bad results.
Thank you Jan.
AKSA said:
You can say that again.
Especially wrt what happens just when you turn it on 🙂
cheers
T
Well, for what it is worth, I'm trying to encourage objective thinking. I can't share all my knowledge base because much is proprietary.
However, if you are not so encumbered may I suggest a beneficial direction might be to offer John Curl some theoretical explanations as to why his amplifier sounds more transparent with minimal NFB?
I believe I used that phrase in the other post:
And I stand by it.
I did not say that NFB necessarily sucks the life out of music. But it is my experience and that of others, both amateurs and professionals, that it can. And the circumstances in which it can are common enough that I think it fair and resonable to make this statement, especially in a forum ostensibly used as a resource by novices.
I think this is a circular argument. Which may well be apt given we are discussing feedback loops
To say NFB makes things sound better when properly used is like saying that if two tin cans and a piece of string are designed properly they will perform as a perfect telephone.
True but not helpful.
To be more helpful I suggest addressing the question as to what assumptions about the chosen theory of NFB are being broken when the observation disagrees with it?
Traderbam,
I was not referring to the statement - apologies if I misquoted the source, but acknowleding Bob Cordells point that both feedback and non-feedback amps win awards.
To focus on your specific point using practical examples:-
Lets take the latest Ayre Monoblock with rave reviews all over the place (well done Charles) and Zero Feedback
Now lets look at the Latest Krells. Also got rave reviews despite having feedback (well done Dan).
Another example. Read HiFi+ from a few months ago - Karan, Ayre, Bryston and some other brand I cannot remember (mag is upstairs and I'm troo lazy to go and get it out). All received fantastic write ups. However, it was not the zero feedback amp that was 'towering above them all' in the reviewers final analysis (their words, not mine). It was an amplifier with feedback.
Does this mean that zero feedback amps suck? No - but clearly zero feedback, in the ears of the reviewer, did not deliver the best sonic results.
I was not referring to the statement - apologies if I misquoted the source, but acknowleding Bob Cordells point that both feedback and non-feedback amps win awards.
To focus on your specific point using practical examples:-
Lets take the latest Ayre Monoblock with rave reviews all over the place (well done Charles) and Zero Feedback
Now lets look at the Latest Krells. Also got rave reviews despite having feedback (well done Dan).
Another example. Read HiFi+ from a few months ago - Karan, Ayre, Bryston and some other brand I cannot remember (mag is upstairs and I'm troo lazy to go and get it out). All received fantastic write ups. However, it was not the zero feedback amp that was 'towering above them all' in the reviewers final analysis (their words, not mine). It was an amplifier with feedback.
Does this mean that zero feedback amps suck? No - but clearly zero feedback, in the ears of the reviewer, did not deliver the best sonic results.
- Status
- Not open for further replies.