A line level op-amp is not the same as a power amp driving a speaker.
But some op-amps can be heard.
But some op-amps can be heard.
Amplification is increasing the frequency and density of electric charge.
In Control Theory, "error" refers to simple, distinctly specifiable deviations (in machine motion). In the case of the tremendously complex audio signal, consisting of myriads of ephemeral changes, the terms "error" and "error correction" become meaningless. What error? Which one?
The ideological discussion on feedback, categorically for or against, can be endowed with subtlety and nuance by putting feedback into the context of topology, device type, biasing, amplification factor, energy level and loading.
1) The emitter resistor (just like the base resistor or any other resistor) reduces the amplitude and velocity of charge vibrations without introducing a loop (feedback).
2) The confusion of output impedance.
Impedance is a property of magnetic field (in common parlance, back EMF), while resistance is a property of electric field. Physically, the difference is like night and day. All physical laws are massive simplifications and idealizations. Ohm’s law has some practical usefulness for engineers and laymen, but frequency, capacitance, inductance not being taken into consideration, it is hopelessly barred by limitation. Not all materials follow Ohm’s law.
All that glitters is not gold. Ohm’s law is not much of a physical law. After having understood, I am sure PMA and Jan will wholeheartedly agree. Stop praising Ohm’s law.
Materials and devices have the tendency to operate at near resonance and are in need of damping. Each amplifying stage should be granted sufficient damping and stability by proper device choice, topology and biasing. Electric feedback provides damping in a rough and rowdy way and introduces distortion that can be difficult to live with. Extrinsic damping does not have the same impact as intrinsic damping.
A low intrinsic resistance gives large signal amplitudes and a high output impedance. Output stages with bipolar transistors, generic industry type MOSFETs, HEXFETs, SiC JFETs, CMOS...fall into this category and exhibit high instability. Output impedance cannot be measured.
Systemic resistance decreases with increasing amplitude and frequency.
More philosophically:
Procedures describing an incomplete whole with just a few simple relations between independent parts produce absurdities. For example, the Fourier theorem.
"Everything simple is false." - Paul Valéry
Beliefs are more powerful than facts. Beliefs are justified by others` beliefs. Anything that is not superficial lacks public interest. Therefore,
"Nothing that is true is popular, and nothing that is popular is true." - Astut folkloric wisdom (courtesy of Jan)
“What people believe prevails over the truth.“ - Sophocles
"The truth cannot be decided by a majority vote." - Democritus
“Wrong does not cease to be wrong because the majority share in it.” - Leo Tolstoy
In Control Theory, "error" refers to simple, distinctly specifiable deviations (in machine motion). In the case of the tremendously complex audio signal, consisting of myriads of ephemeral changes, the terms "error" and "error correction" become meaningless. What error? Which one?
The ideological discussion on feedback, categorically for or against, can be endowed with subtlety and nuance by putting feedback into the context of topology, device type, biasing, amplification factor, energy level and loading.
1) The emitter resistor (just like the base resistor or any other resistor) reduces the amplitude and velocity of charge vibrations without introducing a loop (feedback).
2) The confusion of output impedance.
Impedance is a property of magnetic field (in common parlance, back EMF), while resistance is a property of electric field. Physically, the difference is like night and day. All physical laws are massive simplifications and idealizations. Ohm’s law has some practical usefulness for engineers and laymen, but frequency, capacitance, inductance not being taken into consideration, it is hopelessly barred by limitation. Not all materials follow Ohm’s law.
All that glitters is not gold. Ohm’s law is not much of a physical law. After having understood, I am sure PMA and Jan will wholeheartedly agree. Stop praising Ohm’s law.
Materials and devices have the tendency to operate at near resonance and are in need of damping. Each amplifying stage should be granted sufficient damping and stability by proper device choice, topology and biasing. Electric feedback provides damping in a rough and rowdy way and introduces distortion that can be difficult to live with. Extrinsic damping does not have the same impact as intrinsic damping.
A low intrinsic resistance gives large signal amplitudes and a high output impedance. Output stages with bipolar transistors, generic industry type MOSFETs, HEXFETs, SiC JFETs, CMOS...fall into this category and exhibit high instability. Output impedance cannot be measured.
Systemic resistance decreases with increasing amplitude and frequency.
More philosophically:
Procedures describing an incomplete whole with just a few simple relations between independent parts produce absurdities. For example, the Fourier theorem.
"Everything simple is false." - Paul Valéry
Beliefs are more powerful than facts. Beliefs are justified by others` beliefs. Anything that is not superficial lacks public interest. Therefore,
"Nothing that is true is popular, and nothing that is popular is true." - Astut folkloric wisdom (courtesy of Jan)
“What people believe prevails over the truth.“ - Sophocles
"The truth cannot be decided by a majority vote." - Democritus
“Wrong does not cease to be wrong because the majority share in it.” - Leo Tolstoy
Good overview, thanks.
I only would want to add that there is always an error remaining, the feedback doesn't drive the error to zero.
Although the rule-of-thumb is that the two inputs of the amp have the same signal, it isn't really true. In your example, there will always be a difference between the two inputs, namely the output voltage/open loop gain. That is the error, and with your numbers, it predominantly consists of the distortion.
In fact, one way to measure the very low distortion of a low distortion amplifier, you could measure the difference between the inputs.
Jan
Audio equipment is mainly made to listen to music which is inherently harmonious and a mathematical construct too and it's not unusual that musicians are good with mathematics too.Yet musicians are not interested in the perfect REPRODUCTION of a mathematical concept, they like variations and that is why they are mainly producing music and harmonics, while some non musicians make enormous efforts buying expensive audio equipment to stick to the fixed recording, just because their imagination can't get used with distortions,noise, variations...Luckily most people are used to listen to crappy music onto crappy equipment thus humanity still has a chance to evolve into a sharp species 🙂
Last edited:
Interesting viewpoint. Classical performances can vary enormously, depending on the interpretation by the conductor. And all these interpretations have their following, and some people prefer one over the other.
But, once on of these seemingly arbitrary interpretations is cast in stone (recorded), we often go to extraordinary lengths to try to reproduce it exactly as it was recorded. Is there no contradiction here?
Jan
But, once on of these seemingly arbitrary interpretations is cast in stone (recorded), we often go to extraordinary lengths to try to reproduce it exactly as it was recorded. Is there no contradiction here?
Jan
"
"
If, indeed, 'good' or 'bad' is not the question, do-you feel all digital modern recordings sound the same than old ones "mixed through several dozen 5534's" ?
I'm part of a very unlucky generation whose main reproduction medium was a cheap chinese, polish or russian cassette player...In ex-communist eastern countries we didn't have regular acces to good reel to reel, cd-players or even good cassette decks until 1995...2000.That might be considered a loss, but it wasn't really because we got to appreciate more the art itself and used our basic brain hearing physiology to fight the lack in the music quality...My views as of an usual representative of such generation might differ a lot from a technical point of view yet i found a few years ago that Ray Dolby was the the real God of audio field in that he unified the noise and distortion theory under one single theory which is embodied in its reduction noise techniques and if global feedback has anything to do with it is the ability to make an amplifier sound good enough .I think that from time to time each of us preffer a different sound for their equipment and having all the flavours ,high NFB,low nfb,no nfb, ultra low distortion, valve distortion, tape distortion, is actually about loving the music itself and all its variations.
I would try to give a more applied example of what i mean: can you say that you don't like these youtube clips on the grounds of being intentionally recorded with aditional background noise, on ad-hoc instruments and even on automated mechanical sound machines?
Would you look into recording white noise with a resolution of -120db THD 🙂 ?
Wintergatan - Marble Machine (music instrument using 2000 marbles) - YouTube
Music Box & Modulin - 2 new music instruments ("All Was Well" by Wintergatan) - YouTube
And here's a live more clean reproduction of the first video which has its own flavors:
Marble Machine - Band Version LIVE - YouTube
And yes...i'm mainly an youtube listener because Youtube is the greatest and cheapest source for new art and good enough quality and actually better than most cassette player could give me in the 90's.I preffer to listen to the music instead of the equipment most of the time.I don't even use the headphones although i have some studio grades ones , i simply use laptop's own tiny speakers and i don't feel that the music lacks quality...maybe the reproduction of it, but not the art itself.
Last edited:
Not really. Impedance consists of resistance and reactance. Resistance is a result of energy dissipation. Reactance is a result of energy storage. Both magnetic fields (e.g. inductors) and electrostatic fields (e.g. capacitors) store energy.
Wikipedia: impedance
Oh yes. I'd add "What people want to believe..."
Last edited:
I don't know if its the right place to discuss this here, but I heard a lot this "NE5534 used in all mixing consoles" theory although even worse op-amps were used in countless mixing equipment .Well many people ignore the fact that NE5534 is used mainly in balanced or inverted mode in professional mixers, hardly and with great care in noninverting topology, but there's also a small secret in the bag of every sound engineer or musician to get past the inacuracies of any electronic recording equipment and that is easily explained if you look at the next video:
How does THE MODULIN work? - DIY Music Instrument - YouTube
Using an adsr and a compressor you can make very high quality recordings on the crappiest mixing console...
Why was the electric guitar miked? Well,the speaker-microphone system was a slew rate killer in the first place smoothing out any sharp note.
You may be using a condenser microphone to record a string piano, but that won't pass through without being fed into some sort of compressor first.Not even talking about any compressor circuit distortions here .For that you better ask Barrie Gilbert or DBX...
What was a tape machine doing to the sound until 2000 when almost any tape recording ceased being used ? Guitar players used for a long time the echoplex and the vocalist the Roland re-201...501
You can't figure out what audio electronic equipment should look like without considering the musician and sound mastering engineering behind any recording.
So may i add too to the theory that global feedback can't be used on crappy amplifier circuits 🙂 ? Global feedback won't make a poor design to become well behaved just like that...You need local feedback at every stage too in order to get a global feedback working right.There's no gobal feedback working by himself without local feedback .Then you need compensations if the design is not very fast, you also need compensations if the design is too fast...
How does THE MODULIN work? - DIY Music Instrument - YouTube
Using an adsr and a compressor you can make very high quality recordings on the crappiest mixing console...
Why was the electric guitar miked? Well,the speaker-microphone system was a slew rate killer in the first place smoothing out any sharp note.
You may be using a condenser microphone to record a string piano, but that won't pass through without being fed into some sort of compressor first.Not even talking about any compressor circuit distortions here .For that you better ask Barrie Gilbert or DBX...
What was a tape machine doing to the sound until 2000 when almost any tape recording ceased being used ? Guitar players used for a long time the echoplex and the vocalist the Roland re-201...501
You can't figure out what audio electronic equipment should look like without considering the musician and sound mastering engineering behind any recording.
So may i add too to the theory that global feedback can't be used on crappy amplifier circuits 🙂 ? Global feedback won't make a poor design to become well behaved just like that...You need local feedback at every stage too in order to get a global feedback working right.There's no gobal feedback working by himself without local feedback .Then you need compensations if the design is not very fast, you also need compensations if the design is too fast...
Last edited:
A strong upvote?
traderbam,
1) Some Wikipedia articles should be completely rewritten.
2) There is no need to add anything to that sentence.
Jan,
linguistic expressions do not contain meaning. Meaning is assigned by the interpreter.
Modern digital recordings don't go through a mixing board? What do you feel? Why not stick to the subject at hand does GNF "suck the life out of the music" and how do you determine this?
Someone should let Bruno Putzeys know that it sucks the life out. I bet Kii would get better reviews 😀.
Good to know that nothing has changed here, I've been banging my head against other walls lately.
Talking of the horrors associated with many recordings it is perhaps a good time to recall the amazing concept of the Kinergetics amps. They employed "frozen" magnetic pickups in the NFB in order to correct for magnetic hysteresis distortion in recordings. Allegedly JC preceded them by incorporating a tape head in the NFB loop.
I like the concept: the reproduction chain correcting for some of the sins during recording. This may explain why some less linear devices sometimes sound better.
Now how do we compensate for all those dreadful 5534s? 🙂
I like the concept: the reproduction chain correcting for some of the sins during recording. This may explain why some less linear devices sometimes sound better.
Now how do we compensate for all those dreadful 5534s? 🙂
I don't think the corruption can be undone. It would be like trying to unbreak an egg. Strangely, the more musical recordings I have were made either before transistors or after about 2000 when digital encoding got fixed.