It took more than reading, it took a lot of experimenting and an attitude of "if I believe X and the data say not-X, then I'm wrong and have to correct my understanding."
As far as suggested basic reading, I would strongly recommend starting with The Feynman Lectures, basic physics texts to understand what electrons and electricity are. Tough sledding and it will exercise your calculus skills, but these three volumes will give you a solid grounding in basic physics.
Harry Olson's "Music, Physics, and Engineering" is a very accessible text on sound. Beranek's "Acoustics" is a bit more advanced, but if you're diligent, you can tackle it.
There are a lot of great books on signal processing which will give you a good foundation for understanding digital methods- Steven Smith's nice book, "Digital Signal Processing: A Practical Guide for Engineers and Scientists" is a good start (I was lucky enough to learn this stuff the hard way 35-40 years ago, apprenticing for the Chief Scientist at Nicolet, the pioneers in this area).
This should keep you busy for a while.
Yes I agree with that completely. I thought just this afternoon about superconductors and the way that electrons interact with passive objects in the audio chain, this was also when I was thinking about ways to reduce the noise levels in lm317/337 voltage regulators which are a component of my dac project in another thread...
Anyway, if we can reduce the background noise of a circuit on the molecular level by removing heat/energy then we should be able to reduce the audible interference of both recording and reproduction circuits.
I wouldn't need to reduce the temperature that far, just to about freezing.
I wonder if this would work with the DAC IC itself? Its obvious to me now that the most critical component of a low level circuit is its power supply but suppose someone wanted to pour liquid nitrogen over the semiconductors of a DAC circuit and then measure the result...
Anyway, time for bed! Thanks for the entertaining discussion.
Last edited:
It's much simpler than that- no need to get esoteric. See Feynman's elegant derivations of Johnson noise, for example.
If you spend a little time learning some basics, you can calculate in a few seconds that the best case noise reduction for that is slightly less than 0.4dB.
Yes that does tend to happen with me. What brought that on was my reading about research on fabricating nano-scale electron ''tube'' devices. At that nano scale the device does not have to operate in any vacuum. The electrons can travel from a cathode through grid to anode, and there are so few gas molecules occupying that space, it works like a vacuum tube, which is not really operating in absolute vacuum anyway. The average amount of electrons making it to the anode unimpeded in the nano structure without vacuum is ''comparable'' to a normal vacuum tube.. but sort of working in a semi solid state. Not the same voltages and currents but fascinating nonetheless. Electrons do pass through gas molecules quite normally, some get ''bounced off'' the nucleii...here I go again
One of my oldest buddies (oh, lord, I just realized we've known each other for more than 50 years!) used to fabricate those. They were done as arrays with the idea of using them in sensing. They actually do work, they rely on tunneling for operation, but the operating characteristics make them more of an interesting curiosity than a practical component for analog electronics. He once set up an array with a ring-like structure in each cell to act as a control grid and just for fun, we used it in a little preamp circuit.
Nothing magical happened, alas.
Nothing magical happened, alas.
Are we not recording the voltage level, the number of electrons that pass a certain point, lots of electrons. So how can the quantum state of the electron matter! Analogue and digital recording both have limitations in resolution, such as film and digital sensors have. I am rather puzzled with your theory.
I propose that distortion and noise of an audio signal won't reach perfection until we are capable of also reproducing the quantum state of each electron as it passes through the audio recording and reproduction equipment.
To ignore this state would be to produce errernous results on the reproduction end which generate noise and distortion.
I also propose that to do this no digital system could ever attain perfect reproductive results in an audio system, it could come razor-sharp to the edge but never attain the perfection that an analog audio system would attain.
Lets assume for a moment that this is possible in an audio recording system which captures the electrons and leaves them in an unchanged state, assuming that this would someday be possible then we will have the ability to perfectly recreate an audio signal and reproduce it.
Digital cannot do this, the universe is too mathmatically complex and the interactions that an electron had with the audio cable, with the microphone circuitry, with the plug that plugs into the back of the mixer board, are all unrecorded.
The noise which you hear in uncompressed analog recordings (or the lack of noise that you hear) is a result of rejecting the quantum states of these electrons.
I propose that that noise is a critical component of the audio signal but we discard it like its no big thing. It is a part of the reality of the soundstage but we discard it as if its unwanted.
We never ask what the air surrounding the performers is doing because all a microphone can record is the air coming out of the mouth of a performer or the air surrounding the instrument, only when does a microphone sit inside the same room of the performers does it begin to paint a decent picture of what things sound like in that room. And only when does a loudspeaker become a part of the room that the music is being reproduced in does it decided to paint a decent picture of what the air was doing in that room at the time of recording.
We haven't even stopped using AC powered devices and still have hum in recording sessions (which I would gladly filter out) so how could we ever propose to evolve from this ecosystem of thinking that noise is a bad thing?
Yes, duplication errors in analog systems is a critical problem which will haunt analog for the rest of its days. But who here has listened to an uncompressed analog audio signal recorded live and not gone "wow"?
The next state of analog audio reproduction isn't digital systems, its a perfect (non-destructive, no-contact)copy of a tape using an electron microscope with perfect duplication of the quantum state of each electron as it entered the recording system. So duplicating the physical tape even on a molecular level may not be enough. If we can crack that then maybe analog isn't so dead after all.
The only system which would ever come close to making a perfect copy of an audio stream would be the first master tape which records at a concert. Every other tape after that cannot ever reproduce the exact state of a molecule of magnetic iron ferrite unless it was duplicated on a molecular level.
THINK ABOUT FILM.
Film buffs love 35mm or higher grades, why is that? Its because of the noise which is a part of the films chemistry and reaction with the reflection of light sources as it records the scene. We're ignoring this "noise" by filtering it out. But its a critical component of the audio soundstage. Its a critical component of the rooms acoustic "image".
This is probably a long winded way of saying that we need more resolution, that analog could only ever reproduce this higher resolution, and so therefore digital is equivalent to a transporter or replicator on star trek where no one likes the taste of the beverage despite it being molecularly exact to the original.
But analog is like taking the shuttlecraft. And now I'm gonna go have my breakfast!
Last edited:
This makes absolutely no sense. Not from the standpoint of the nature of signals, the nature of electrons, or basic noise theory. I know it bothers you to hear this, but it is indeed absolute gibberish.
If I were to start throwing out hypotheses about the nature of the development of Greek grammar, I might first want to learn Greek, and I would not be offended if people who DID understand Greek told me that I was making no sense.
Agreed, its wrongly worded. But if you read the rest of the post instead of shooting like a texan then you might get an idea of what I'm trying to convey.
I edit a lot.
We developed digital audio standards first without a full understanding of the states of an electronic audio signal, how do you think it feels being chopped up into a million little pieces and then redefined as something else?
If a person was an electron how do you think it feels being treated like that? "That's not what I meant!" it would say...
Last edited:
haha
Thats a terrible example and only serves to strengthen the notion that analog/free range-natural-straight-from-the-tree is better.A better analogy would be that I am simply the avatar that you see on the screen infront of you, and the text that you see on the screen infront of you, yet you need to reproduce what you read in your head inorder to understand what I'm trying to convey.
At least thats what digital thinks of me.
If this were analog I would have all the electrons that could fit down a pair of copper wires to be able to reproduce what they see on a CCD video camera and what they hear through a microphone. Instead I'm left with this archaic text based digital system.
Last edited:
That's nonsense, as well as a classic pathetic fallacy. Shannon and Nyquist were well-established theorems (not theories!) 50 years before anyone was playing with digital audio. The rest of the post was similar gibberish which doesn't actual deal with the realities of devices or signals.
Learn Greek before you start expounding on the history of Greek grammar. At your request, I've given you some good references- you might want to read them and work through the problems to gain an understanding of the basics.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.