Because you can't quantify that goal, it is very fuzzy and if you ask three people what the goal means you get four different answers. ;-)
You have to have some way to decide whether your goal has been reached. That is not possible for your goal.
A possible design goal could be: I want to design an amp that to me does not sound different from the input signal. It can be verified and you can know when it has been reached.
Jan
The power paradigm was the means that was used to obtain flat frequency response in the absence of GNFB. As far as what it is that is being disputed, I would say that those that are disputing me are claiming that I don't understand how the voltage rules work when I use them every day. SY, as you know, in moderation of a site, the rule is 'attack the argument not the poster'. I've been attacked personally because I presented a bit of history. It got conflated somehow that I personally had some sort of 'profound' misunderstanding and such and despite his remonstration and backpedalling, DF96 was putting words in my mouth. I've used the word 'paradigm' (as I have now explained about 4 times on this thread) because its a different way of looking at things and the expression (not the math) is different enough that it appears to be blasphemy if you stay within the dominant paradigm (in this case, the voltage rules). I'm having trouble understanding how people can understand physics and have really poor reading comprehension at the same time.
In a nutshell, Fast Eddy gets it.
I accept your apology. However (text in bold), you still seem to be confused: I'm not making the claim. You are still trying to shoot the messenger.
Just like Fast Eddy was with Jan earlier. Dude, I know the science. I spent 4 years getting my degree. I was simply presenting the viewpoint that seems to exist within the power paradigm. You keep trying to claim that I don't understand how stuff works. Do you think I could have gotten as many awards in the press as I have by being an idiot?
Uh, FWIW, I answered SY's questions directly without avoiding anything. If you can't see that, seriously: Take some reading comprehension classes.
Actually classic trolling behaviour is any post in which the goal of the post is to intentionally alarm, antagonize, or provoke other users. Latching on to a particular word does not fall into the definition, and when SY used it, it bothered me because I knew it was not new and it was not something I was claiming. So its not trolling, it was correction.
I thought since this is a thread about GNFB, people might be interested in knowing how flat frequency response was done in the old days, and further that the concepts in use then are still used (in both cases, consciously and unconsciously) so I presented it. But the result is that I got personally attacked for being an newbie (which I am not, been in the business over 40 years), accused of trying to violate the laws of physics, etc.
I'm currently of the opinion that you still owe me some apology.
Perhaps the issue is with the term "paradigm." I think of it in a Kuhn sense.
In reality (and feel free to correct or elaborate), you're talking about speakers, not amps. An amp will simply be a Thevenin source with some output impedance. Period. A speaker can be designed for zero source impedance or some finite source impedance. A speaker designed for the former will not work well with an amp designed for the latter and vice versa. There is no such thing as "constant power" unless the speaker has a flat impedance curve, in which case power will be constant regardless of the amplifier's source impedance.
Do you have any disagreement here?
In reality (and feel free to correct or elaborate), you're talking about speakers, not amps. An amp will simply be a Thevenin source with some output impedance. Period. A speaker can be designed for zero source impedance or some finite source impedance. A speaker designed for the former will not work well with an amp designed for the latter and vice versa. There is no such thing as "constant power" unless the speaker has a flat impedance curve, in which case power will be constant regardless of the amplifier's source impedance.
Do you have any disagreement here?
OK that is what I meant, but as I don't know how the input signal sounds without connecting an amp, better to say as close as possible to the live (classical) concert.
Damir
Somewhat. If you play around with a tube amp with zero feedback, you will find that they do indeed exhibit something very near constant power within a reasonable range that is met by many speakers. IOW you don't need a flat impedance curve for this concept to work- ESLs are a good example of that. They vary by about 10:1 over their range, yet you can still get good response over that range with the right amp.
In the old days, all speakers were designed for amps that had an output impedance in the range of 2-10 ohms or even more.
What we are talking about here is a speaker and amplifier matching issue; its not a speaker thing exclusively or an amplifier thing exclusively. Of course, the problem is still around- one must still be careful matching amps and speakers for best results. IOW the voltage rules did not solve the problem, although plug and play (if your only concern is flat FR) is much better.
Part of the problem is that tubes failed to go away after being declared obsolete. Some speakers are really designed to work better with tubes than SS and vice versa. Universally a speaker designed to work under the power rules will operate fine with any tube amp. However a speaker designed under the voltage rules may not, despite the tube amp in question being a voltage source. And you certainly can't put a transistor amp on all speakers and expect flat FR (again, ESLs are a good example). So the matching issue is still with us.
A further concern is coloration. We have known for some time that the ear/brain system translates distortion into tonality. Adding GNFB to many amps often results in a coloration called 'brightness' (and also harshness) which many audiophiles find to be anathema. This is why we are having this conversation of this thread BTW; if GNFB worked universally (more to the point: if designers designed proper feedback networks that actually did the job without introducing problems) the point would be moot; this thread would not exist.
I'm out of town this weekend (in a bicycle race)- you all have a pleasant weekend! 🙂
doesn't happen - commercial music recording/production simply isn't about reproducing the closest measurement to a audience member's experience in a live event
multi mics are always used with directivity, separation, locations chosen to emphasize certain sound response, minimize others like audience noise
Classical Symphony recordings may have "realism" a higher priority than individual close miced or even separate studio booths for each player in more popular music production - but it still is manipulated for an "enhanced" result over having a binaural dummy seated in which ever seat you think is the "best" in the house (which differs by performance style, orchestra arrangement, featured instrument or performer and listener preference too)
Last edited:
(My inserted square brackets) To the first point, no. An amplifier can have an output impedance of 8 ohms (say) and not waste half the power. You merely need to sense output current (entailing a small voltage drop) and adjust the output drive via feedback to effect the desired impedance. You don't need an 8 ohm resistor in series with a voltage source.
On 2: Of course loudspeakers may have resistors as part of the voice-coil-level crossovers, although usually not in series with a woofer but rather part of the adjustment of the response of the other drivers.
On the other hand, with amplifiers:
If I recall correctly, Bob Carver, in typical Carverspeak, had a second output of one of his power amps that he called a "current source output". It featured a big fat series resistor, I think of an ohm or two (some current source!). Needless to say it did waste power, but using that output was advertized as sounding more tube-like. The same effect could have been realized with current-sensing feedback as described above.
Dadod, dam good one, pretty sure jan meant test tones... I like your way of evaluation.
Funny if your angle is listening to music then you realize that data does not reveal a lot about how it sounds.
Just the fact that we simplify music to test tones and real life speaker loads to RCL combinations in order to get a somewhat credible extract of how the unit performs in the real world tells a lot.
It's like trying to make live tv with animations. A very coarse picture.
Measurements can only be a control of that the unit is operating as expected. Only listening can say something about the performance quality of the unit.
It's like looking at car data, horsepower, weight, size ect ect, and then use that to judge how the car is handling on the road. Data will give you pointers, but the real test is the feel of driving. That you can't quantify and put in numbers.
Funny if your angle is listening to music then you realize that data does not reveal a lot about how it sounds.
Just the fact that we simplify music to test tones and real life speaker loads to RCL combinations in order to get a somewhat credible extract of how the unit performs in the real world tells a lot.
It's like trying to make live tv with animations. A very coarse picture.
Measurements can only be a control of that the unit is operating as expected. Only listening can say something about the performance quality of the unit.
It's like looking at car data, horsepower, weight, size ect ect, and then use that to judge how the car is handling on the road. Data will give you pointers, but the real test is the feel of driving. That you can't quantify and put in numbers.
Last edited:
That won't work - it will never even sound close to live. That's a common misconception. Think of all the stuff round a live concert, the acoustics, the PA system, the recording mike setup, the mixing, etc etc. No chance in hell that it can be reproduced in your room even half way the way it sounds live.
Another goal could be: as good as an amp that you know and you think sounds good.
Or, another goal: reproducing the input signal with less than 0.01% distortion, all harmonics at most 1/3 of each previous one, flat frequency response, enough power for your speakers to avoid clipping 98% of the time.
Use your imagination ;-)
Edit jcx x-posted.
Jan
Last edited:
Here's his rationale:
"Unnatural harmonic BALANCE = unnatural sound. Cheever explains it quite well in his thesis. It is not just even or odd harmonics but like what Jean Hiraga said long ago a monotonic pattern (so even and odd where the next higher odd is always lower than the previous even in an exponential decay). This means a pattern with only even or only odd is wrong and anything with high order harmonics above say 9th (and then only at high volume levels) is also not following a natural pattern.
Like all things in human perception it is about PATTERN recognition and not absolute levels. You need a 2nd followed by a 3rd that is 20db lower followed by a 4th, 5th etc. each with at least 10db lower than the previous. Very few amps get close to following this ideal.
Putzey's, like a lot of engineers, think that if they can supress everything enough it will be effectively invisible but there is always enough residual that it is not the case and since the residual doesn't follow the pattern it is audible even if it is very low in level. Putzey's and a lot of other engineers are trying to work against nature rather than working with nature to get the best sound.
Geddes reached similar conclusions but without all the ear analysis but with blind listening tests nonetheless.
So, in the end, it is about the application of correct SCIENCE to produce technology that most people would find to be natural sounding. Of course self-deception, marketing, convenience (weight, heat, size) and drivers other than pure sound quality will always exist and give a reason for unnatural sounding devices to exist (mostly driven by price because to truly follow the recommendations is invariably expensive).
Putzey's surely has impressive ENGINEERING but he is not following what has been determined by science and so his goals are misplaced. Audio is one of the few fields of technology where scientific understanding has been largely dismissed. It is frankly a bit bizarre.
It is only so subjective because so few are following good science when thinking about audio engineering decisions. You then get all kinds of flavors of wrong. I have said it before and will say it again...MOST amps are wrong...some a bit wrong some very wrong. Note I didn't specify a type. The ones I have found to be the least wrong are Class A tubes or hybrids. This also fits with the measurements from these types of devices. "
It seems to me that hearing, like the rest of our senses, might work the same way but that those inputs are interpreted differently, and thus we don't all agree on qualitative differences.
But this person claims to be a "scientist" and research has "proven" what sounds best.....I guess someone forgot to tell all those happy ncore owners that....
Kirchoff's laws date from 1880's and still describe the low frequency limit of Maxwell's equations now as they did in the 1920's. Your concept of "voltage rules" makes no sense.
EDIT - Sorry I think I get what you mean, don't you think "before they invented feedback" makes more sense than "voltage rules"? A friend's grandfather swore by push-pull 45's with inter-stage transformers into one of those stiff speakers mounted in his closet door.
Last edited:
Wouldn't that depend on the nature of each distortion? I mean, there are an infinite number of ways to distort (signals to add, iow). And for that matter, levels to measure distortion at. Do you reference the 1% (power? Voltage?) to the amps rated output power, to some nominal power, to the average program power, peak program power, program spl peak or average, threshold of hearing? Lots of ways to replicate erroneously, just one to replicate exactly.
There is a wide continuum from pure voltage drive with Zout = 0 (which you can never reach) and pure current drive with Zout = infinite (which you can never reach). But hey, you can't put that in a sales brochure 😀
Jan
Shorter rationale: I want distortion.
Again, that's fine if you start out with that as your goal. If your goal is to have the amp output replicate the amp input and do your signal processing elsewhere (like digitally!), it's not so fine. He's criticizing Bruno because Bruno isn't interested in having the amplifier be an effects box, he's only interested in having the amplifier amplify.
This is why I keep harping on setting the target first before deciding where to shoot.
I've seen this comment a lot, and I don't quite get it. Say, for an example, I had two voltage amplifying stages, the second of which has high gain but which also has a tiny input peak voltage capability. I put global feedback around both stages and drive to full output, the voltage signal level present at stage 2's input is small, it all works great. But if done with separate feedback around each stage, now the input to stage 2 needs to be a higher level to try to reach full output (the stage's gain is now lower from the NFB), but can't because of the tiny handling range, distorts badly. Clearly not the same (or am I missing something?). Doesn't it depend on what kind of distortion is generated where? Or is my example too pathological?
Last edited: