At the same time, with a decently modern transistor-based OPS and some judicious compensation, it's not hard to push distortion in the top of the audio band into the weeds. So, sure, Bruno may prefer flat distortion profile when there's near-audible/audible distortion levels, but it's a point realistically made moot.
And in what PP amplifier does OL H2 distortion not dominate?
And in what PP amplifier does OL H2 distortion not dominate?
Note carefully his words ' but whose distortion is not negligible at the end of the audio range' if you still have 40dB+ feedback at 20kHz distortion IS negligable.
In fact he twice stated the experiments were using non-negligible distortion amplifiers. So if your stuck making an amplifier that will color the sound with its distortion you have a choice to make?
Wow, yes, only $200,000.00 (pair) for the bigger one and $100,000.00 for the smaller one 😀
Your quote starts with the original point OLBW is a meaningless metric. So the claims that a high >20kHz open-loop bandwidth per se (as people like to attribute to Otala) guarantees better sound is based on nothing. So where is the disagreement?
Fair enough .. i see the high ol bw as a necessary (but insufficient) means to get FLAT thd vs frequency. That is an important characteristic of (commercially and critically) successful amplifiers.
That does create a problem for amplifiers like RIAA where the frequency shaping sometimes is partly formed by directly modifying the open-loop gain. You would need to modify that to be referred to input THD to make it a universal principle, which I still think is not totally proven. Bryston comes to mind their plots have the 20dB/dec rise as usual.
Last edited:
commercially successful and critically successful are often mutually exclusive.
For example this NAD M2 Direct Digital integrated amplifier Measurements | Stereophile.com probably sells more than any of the others they review.
For example this NAD M2 Direct Digital integrated amplifier Measurements | Stereophile.com probably sells more than any of the others they review.
@Kasey197: Can't say I disagree with your interpretation of Bruno's presentation and got my memory corrected.
To the last point: Bob Cordell did put this in perspective in his book and showed comparative graphs with low and high feedback. It is clear that indeed, as you say, the Baxandall example was a specific case and with reasonable linear amplifiers it can hardly be seen if at all. I have been making the same case years ago at a presentation I believe it was BAF, but could have been a local NL audio club.
Jan
To the last point: Bob Cordell did put this in perspective in his book and showed comparative graphs with low and high feedback. It is clear that indeed, as you say, the Baxandall example was a specific case and with reasonable linear amplifiers it can hardly be seen if at all. I have been making the same case years ago at a presentation I believe it was BAF, but could have been a local NL audio club.
Jan
What I don't agree with is this. It is here silently assumed that it you have flat OL gain over frequency and thus flat FB factor over frequency, you then automagically have flat THD over frequency. That is an unwarranted conclusion.
It will only be true if the amp has flat OL THD vs frequency which is generally not the case.
Jan
I just picked up a Keithly 2015 6.5 digit meter that also does THD and has a generator. Not something I needed but have always wanted. From the looks of the inside it is and ad for Analog devices.
Will be listening to some stuff with identical distortion curves but done in different ways should be interesting to see if we can tell them apart. I can but only with FFT so far.
Will be listening to some stuff with identical distortion curves but done in different ways should be interesting to see if we can tell them apart. I can but only with FFT so far.
All else being equal, high frequency open loop response is better for audio designs than extremely high feedback with low frequency open loop response because the differential phase distortion will generally be reduced in the higher frequency open loop response case.
wrong - and you have no excuse for mangling fairly basic feedback theory
refereces, sims, hardware measurements all have been linked, posted repeatedly in this thread showing otherwise
Walt Jung and Marshall Leach both changed caps, switching from an initial adoption of Otala's "flat loop gain" prescription to admitting in writing that there wasn't the simple connection with low SID/PID/FM IMD and loop gain shape
refereces, sims, hardware measurements all have been linked, posted repeatedly in this thread showing otherwise
Walt Jung and Marshall Leach both changed caps, switching from an initial adoption of Otala's "flat loop gain" prescription to admitting in writing that there wasn't the simple connection with low SID/PID/FM IMD and loop gain shape
Marsh, hearing noise and hearing the effects of noise are two vastly different things. Yes we all hear "noise" when vinyl plays but who cares? You can listen past it with no effort. However vinyl systems, like all, benefit greatly from reduced noise on the power side and signal management.
The easiest way to think of it is that noise in the power changes the music itself. Really all I can say is, no **** you don't hear the "noise" during playback as a distinct sound. For one a lot of it is beyond the audibly identifiable range (doesn't mean we can't respond to it, we just don't know it's there by ear). For two people take great care to eliminate audible oddities, artifacts, in the music (if you're hearing them it's probably broken). But all the noise going into a system in the power is grafted to the signal ala transistor or whatever. Even more than just that there's dozens of parasitics paths for noise as well. When it's not directly affecting it, noise can a loss of response in the power system from resonations, stored energy.
Another note:
A low distortion amp should sound very clear. Sometimes they do. But the fact of the matter is low voltage distortion doesn't equal proper current response, nor does it equal harmonics our brain can cope with or a well defined linearity (almost one in the same). And you can even go as far as to point out THD+N is a factor, as well approximating where the distortion is occurring since some amps are concentrated in some frequency areas we're more sensitive to, but the overall THD isn't high. It's just not a pure question of THD, proven in the 70's.
The easiest way to think of it is that noise in the power changes the music itself. Really all I can say is, no **** you don't hear the "noise" during playback as a distinct sound. For one a lot of it is beyond the audibly identifiable range (doesn't mean we can't respond to it, we just don't know it's there by ear). For two people take great care to eliminate audible oddities, artifacts, in the music (if you're hearing them it's probably broken). But all the noise going into a system in the power is grafted to the signal ala transistor or whatever. Even more than just that there's dozens of parasitics paths for noise as well. When it's not directly affecting it, noise can a loss of response in the power system from resonations, stored energy.
Another note:
A low distortion amp should sound very clear. Sometimes they do. But the fact of the matter is low voltage distortion doesn't equal proper current response, nor does it equal harmonics our brain can cope with or a well defined linearity (almost one in the same). And you can even go as far as to point out THD+N is a factor, as well approximating where the distortion is occurring since some amps are concentrated in some frequency areas we're more sensitive to, but the overall THD isn't high. It's just not a pure question of THD, proven in the 70's.
To JC: Yes that is correct I believe, in principle. Although I think Cordell and Quan have shown that the difference it is so small that it is unlikely to be audible. And audibility of differential phaseshift has never been show to be audible anyway.
Jan
Jan
I hate to remind you guys, but I design successful audio amps for a living, and have been designing my own amps for the last 50 years.
Now let us take an example of my 'success' in making audio power amps and various price points from a recent issue of 'The Absolute Sound' march 2017 as shown by listing under 'Editors' Choice Awards:
Parasound A23, A21, JC-1, for example. There are more choices as well, listed, and we don't even advertise in the magazine anymore. I rarely ever get less than an 'A' rating for any of my designs in either Stereophile or TAS and WE DON'T EVEN ADVERTISE IN THESE MAGAZINES! If I didn't know what I was doing, how could I be this successful? '-)
Yet, you, who do not design audio products for a living or on a regular basis, think that you do know what you are doing?
Now let us take an example of my 'success' in making audio power amps and various price points from a recent issue of 'The Absolute Sound' march 2017 as shown by listing under 'Editors' Choice Awards:
Parasound A23, A21, JC-1, for example. There are more choices as well, listed, and we don't even advertise in the magazine anymore. I rarely ever get less than an 'A' rating for any of my designs in either Stereophile or TAS and WE DON'T EVEN ADVERTISE IN THESE MAGAZINES! If I didn't know what I was doing, how could I be this successful? '-)
Yet, you, who do not design audio products for a living or on a regular basis, think that you do know what you are doing?
OTOH some here do know when the basic fundamental principles are violated more than a bunch of buddies casually sighted listening is not enough.
- Status
- Not open for further replies.