What use a therory not proved in practice?
But yes, you are right, I am concerned with how something works in hard everyday practice. Or doesn't, as the case may be. I am not into theorizing.
In my case, you would be right to understand that it's MY truth - I am too old and experinced to search for the absolute truth. MY truth simply means that if it works for me as theory has it down to a +/- 5% tolerance, to me, that is the truth.
If my phrasing has caused you to misunderstand, I do apologize for that, it was surely not my intent to make it sound so.
Truth? What is 'truth' when it comes to feedback design? That Cordell has made a prototype or two that he likes? I don't remember hearing his amp, or any INDEPENDENT reviewer having access to one. That is what we mean by 'utility'. Make something, give it out for everybody to try, as wide a selection as possible, and get their listening feedback about what you have done. I can almost NEVER follow Otala's philosophy exactly. I just don't have enough gain-bandwidth, except for my one-and-only open loop design, the CTC Blowtorch. However, Charles Hansen of Ayre has made a number of designs, both preamp and power amp and gives me a real run in listening contests, you know, serious ones, like major reviews, etc. HE is MY competition, not Bob Cordell, and he believes in open loop design even MORE than Matti Otala.
If you think we are kidding ourselves, just look at the number of A rated components that Ayre and I make in 'Stereophile' or 'TAS'.
If you think we are kidding ourselves, just look at the number of A rated components that Ayre and I make in 'Stereophile' or 'TAS'.
Was there ever a Cordell designed Amplifier made available to the public .......? The Ayre product ( as others ) has the same issue that we were discussing earlier, about high distortion in the mili-watt area, distortion starts high and then decreases before increasing again ...
EDIT: This is not to say it is a bad product, just a pointer to what was being discussed previously ..
Ayre MX-R monoblock power amplifier Measurements | Stereophile.com
EDIT: This is not to say it is a bad product, just a pointer to what was being discussed previously ..
Ayre MX-R monoblock power amplifier Measurements | Stereophile.com
Attachments
Last edited:
For the record , Bob is an amplifier Guru ....
The Amplifier Guru speaks
http://jipihorn.free.fr/Documents/Divers Blog/Cordellinterview.pdf
The Amplifier Guru speaks
http://jipihorn.free.fr/Documents/Divers Blog/Cordellinterview.pdf
Yes, we had originally thought such and there was a discussion on this earlier John and it was shown that the distortion on a lot of amps in the mill-watt area was very high, even full class-a amplfiers.
I will try and link this ...!!!
As to the graph posted on the Ayre, Distortion starts out higher then lower before rising again, are you saying below one Watt, what is being measured is Measurement noise residual ..?
This is that all important 1st watt ...
Although if an amplifier has a stable and smooth set of nonlinearities in the vicinity of zero signal, the net distortion should fall off predictably with decreasing level, it's not inconceivable that the concatenation of stages of some sort might have an overall signature, a curve-of-growth, that actually produces net distortion that is non-monotonic with level.
Also, another gotcha is to suppose that the measurement equipment itself has no problems at lower levels. If you've ever tried to make a really good true-rms converter, or even just a good low-level rectifier, you quickly come to appreciate the difficulties.
If given an amplifier with a claimed rise in low-level distortion, independent of the residual noise, it might be a good idea to use the traditional analyzers in conjunction with a careful nulling approach, and see if the results are in agreement.
Yeah, I calculate a little under 0.2mV of noise would do it, and that's a pretty reasonable number for a power amp.
It is impossible for a class A region to increase distortion at low levels.
To reassure yourself, look at the Ayre graph more carefully
It is SCALED to be read out in Watts. The voltage sensitivity therefore changes as to the rated load resistance. Note the noise increase with the scale magnification increases. In fact it should be approx. 3 dB per measurement, because it takes less volts out (and therefore in as well) to achieve 0.1W out, for example, with a lower resistance load.
To reassure yourself, look at the Ayre graph more carefully
It is SCALED to be read out in Watts. The voltage sensitivity therefore changes as to the rated load resistance. Note the noise increase with the scale magnification increases. In fact it should be approx. 3 dB per measurement, because it takes less volts out (and therefore in as well) to achieve 0.1W out, for example, with a lower resistance load.
Noise .
This might be general knowledge , however I have never seen discussion about this . A friend of mine was taught by Martin Hawksford . If I am correct Martin offered an explanation for people liking tube amplifiers . He said that tubes have a pink noise spectrum and transistors at best white ( blue ) . The suggestion was that reducing noise to the lowest possible levels produced a feeling that the warmth of the sound is better . If this is true it might explain why tube amps and transistor amps with similar distortion and general qualities still sound different .
How do people feel about null tests where a " blameless amplifier " acts as a buffer and mimics a preferred amp ? I was also told that the other form of null test useful where the distrotion residuals are listed to . Seems reasonable to me that it should relate to the sonic signature of the amp . Quad always maintained that a 405 would always sound identical to the amp it was coupled to with the proviso that phase was correct and load was sensible .
I don't know if this is Urban Myth . I was told Bob Carver met a challenge to modify one of his amps to sound exactly like any another by null cloning it ( feedback loop and bandwidth limiting ) . The test done in a hotel room with the tools and pieces he brought with him . The story continues that he won his test . If the next part is true he said what it might prove is that some had better ears than him .
This might be general knowledge , however I have never seen discussion about this . A friend of mine was taught by Martin Hawksford . If I am correct Martin offered an explanation for people liking tube amplifiers . He said that tubes have a pink noise spectrum and transistors at best white ( blue ) . The suggestion was that reducing noise to the lowest possible levels produced a feeling that the warmth of the sound is better . If this is true it might explain why tube amps and transistor amps with similar distortion and general qualities still sound different .
How do people feel about null tests where a " blameless amplifier " acts as a buffer and mimics a preferred amp ? I was also told that the other form of null test useful where the distrotion residuals are listed to . Seems reasonable to me that it should relate to the sonic signature of the amp . Quad always maintained that a 405 would always sound identical to the amp it was coupled to with the proviso that phase was correct and load was sensible .
I don't know if this is Urban Myth . I was told Bob Carver met a challenge to modify one of his amps to sound exactly like any another by null cloning it ( feedback loop and bandwidth limiting ) . The test done in a hotel room with the tools and pieces he brought with him . The story continues that he won his test . If the next part is true he said what it might prove is that some had better ears than him .
I doubt it. I would expect that a cascade of stages, each with a smooth gain curve around zero-crossing, has a total gain curve which is also smooth although perhaps slightly less smooth. This assumes that there are no mathematicians' pathological functions lurking around.bcarso said:
I would not want to do a proof, but I expect that the product of a few rapidly-convergent smooth infinite series is itself a rapidly-convergent smooth infinite series. Distortion will then be monotonic with signal level, for sufficiently small signals.
I assume you mean Malcolm Hawksford? I thought both valves and transistors had essentially white noise, apart from 1/f at lower frequencies. The circuits might have different noise spectra, due to different architecture.nigel pearson said:
BTW why the space before punctuation? It makes reading your posts more difficult.
Some of us had a suspicion that Carver actually didn't do anything, or at most added a single resistor to the output of his amp to get the frequency responses equal.
Despite the amazing performance of many people in ears-only testing of things like frequency response, data compression, phase, level, and localization, when the tests don't support the marketing or a priori assumptions, they may safely be dismissed with a sneer.
Despite the amazing performance of many people in ears-only testing of things like frequency response, data compression, phase, level, and localization, when the tests don't support the marketing or a priori assumptions, they may safely be dismissed with a sneer.
- Status
- Not open for further replies.