Either the bits are all there or they're not?
A digital file is made up of just zeroes and ones, a specific number of them in a specific order. So long as those zeroes and ones arrive at their destination in the same quantity and same order, there is zero difference between the original file and the transferred file. They are perfectly identical. And that sounds difficult, but computing would be impossible if we didn't already solve that problem first. If a computer couldn't perfectly move bits from, say, the hard drive, to the memory, to the processor, then they simply wouldn't work. Even a simple electronic calculator would be impossible if we hadn't figured out how to achieve this. But calculators and computers do work, because we mastered bit-perfect information transmission decades and decades ago, out of necessity. We're so good at it that we can send bit-perfect information between two computers on exact opposite ends of the planet, going through untold miles of wire in all states of integrity, at billions of bits per second with only microseconds of delay. Bit-perfect digital information transmission isn't trivial, so to speak, but it is 100% a solved problem and asking an electronic device to transfer information perfectly is like asking a human to breathe. It's automatic, it's the simplest, most basic task to ask of an electronic device. And this is why the world is so digital today: Because absolutely perfect and nearly-instantaneous information transmission is much easier to achieve with digital technology rather than analog technology. A zero is a zero, a one is a one, and so long as they're all there and in the right order, there is absolutely zero signal degradation whatsoever. Doesn't matter how dirty or noisy or nasty the wires are that got the signal from point A to point B. So long as all the zeroes and ones are where they should be, it's identical, zero loss.
The reason this works is because of error correction. And this is generally the primary bottleneck that dictates how fast our technology is. It's not about how fast we can send information. We can send information at the speed of light, no problem. The bottleneck is, how fast can we verify that the information was transmitted intact. To grossly oversimplify, with error correction, we can send a handful of bits to a machine, and the receiving machine will check to see if those bits make sense or if it just received gibberish, and if it received gibberish, it'll say to the sending machine "Hold up, resend that last chunk" and the transmission will keep retrying until those handful of bits make it intact. And that's why, say, your WiFi gets slower as you get further away from the antenna broadcasting the WiFi signal. The further away you get, the harder it is to transmit bits perfectly, for a variety of reasons. So more transmission errors occur, and error correction steps in and says "Hold up, those last few bits were garbage, resend" more often. The bits get from point A to point B just as quickly whether you're near or far, they just don't arrive intact as frequently the further you get, so error correction has to ask for them to be resent more often. More time spent resending the same little chunk of data until it finally arrives intact means less time sending the rest of the data in line behind it, and so you get slower data transmission overall.
A good example of digital signal transmission without error correction was the bad old days of satellite TV and how shows would look like an absolute mess in bad weather. That's because in the old days of live TV broadcasts, there was no opportunity to request a resend of any data that arrived damaged. Satelittle TV broadcasters just kept sending more and more information regardless of transmission success, and in bad weather, digital information often arrived damaged, and all your TV could do was accept the damaged data, do its best to display the gibberish data on screen (or just go blank altogether), and then move on to the next data it was receiving.
In short, computers and our entire digital world today only work because of bit-perfect communication and error correction, and if you want to learn more about it, here's a good video:
Suffice it to say, error-free digital transmission of data is a solved problem. Even the simplest computer wouldn't work if we couldn't manage that. And really, truly, if we're talking specifically digital audio, yes, you can be 100000000000% assured that whatever digital audio was sent from one machine arrived in an identical, unaltered, perfect state on the second machine.
...PROVIDED ALL THE MACHINES ARE CONFIGURED CORRECTLY.
This is where problems occur when it comes to digital audio transmission. Generally every device in the digital audio chain has a mode of operation where it simply receives digital audio and then passes it on unaltered. But that doesn't mean that every device is operating in that mode. For instance maybe you have a 24-bit/48khz file and it's being passed on in 16-bit/44.1khz mode. Processing and data alteration had to be done to make that conversion, which in all likelihood might be happening for no reason other than, the device doing it was simply configured wrong and that data alteration can easily be prevented, it just
isn't being prevented.
My example of 24/48 audio accidentally being converted to 16/44, there's still debate about whether or not that conversion introduces any actual humanly perceptible artifacts. But there's other ways signal processing is being automatically performed that are 100% audible. For instance, most EQ'ing these days is digital signal processing that is usually done intentionally by people to achieve a desired effect. But Windows, for another example, introduces all sorts of signal processing options, many of which are enabled by default. "Enhancements" and "spatial sound," etc etc. So, if your goal is to transmit digital audio from one point to another with zero alteration or loss, you have to make sure every digital device in the chain is configured to work that way. And the more digital devices you introduce into the chain, the greater the likelihood that some device in the way is configured to be doing something you don't want it to do. When it's just a Windows PC transmitting to a USB-connected DAC, easy peasy, go into Sound Settings, uncheck a couple boxes, job's done, you get bit-perfect audio. When you have all these other devices in the way doing god-knows-what to the audio, it's not so simple. So the best way to have perfect, unaltered digital audio transmission is by having as few devices handling the audio as possible.
So anything is possible after it passes the DAC, including an alteration of the sound?
So far I've been talking specifically about digital audio signals. Once the digital signal hits the DAC and gets converted to analog, all bets are off, this is where all kinds of entirely uncontrollable and uncorrectable signal alteration can creep in. For instance, electromagnetic interference. Digital audio transmission is susceptible to things like EMI just like analog audio, it's just that digital audio has that error correction step to make sure it, at some point, receives information unaltered by EMI or anything else. There is no error correction in analog signal transmission, because while digital signals are either zeroes or ones, analog signals are fluctuations in voltages that can vary by as small an amount you can measure. Plus, the simple act of measuring the signal for error correction would itself alter the signal anyway. You can never be truly sure an analog signal arrives intact, because you can always just keep taking finer and finer measurements and finding finer and finer discrepancies. And that's how you get things like amps with a supposed 0.00000001% THD. It's a pretty meaningless figure at that magnitude. All it's really telling you is how precise an instrument they use to measure these things. All the competitors need to do to improve on that figure is invent a device that can measure even finer. Nothing to do with the audio equipment itself.
But suffice it to say, once a digital audio signal is converted to analog, that analog signal is influenced and altered by every single component and electrical connection and millimeter of wire it passes through. Whether or not these differences and alterations are humanly perceptible is the great debate in audiophilia that will likely go on forever. It's definitely true that different amps sound different. Just compare something like the most tube-sounding tube amp to anything from, say, Topping. They're night and day. The difference isn't even slightly subtle. And the very real and very audible alteration that a tube amp can impart on an audio signal is the entire reason tube enthusiasts buy them. Because it's a perceptible and desireable alteration. But then you get into debates along the lines of copper wire versus silver wire for analog audio transmission, or "audiophile power cables" or "audiophile fuses," etc etc. Sure, a difference might be measurable, but it doesn't mean it's humanly perceptible, and even then there are so many variables that just comparing one sample of each doesn't tell you the whole story.
Because analog signals are so susceptible to alteration, that's why it's generally regarded as best practice to keep the signal in your chain digital as long as possible: Because transmitting the digital signal with zero degradation or alteration and verifiying the transmission was perfect is, to the user, trivial. And having as few digital devices as possible in the chain means it's easier to make sure they're all configured properly and easier to be aware of exactly what's happening inside every device. But once you convert the signal to analog, anything can happen, electrically and/or environmentally, and at this point I believe you just have to follow your ears. Every analog device and every component inside every device influences the sound (be it perceptibly or otherwise), and every analog signal goes through dozens of components before it reaches your headphone/speaker drivers, and then finally your ears. It's up to you to determine if you can hear those influences, and if you think you can, whether or not the influences are desireable.
. If it were galvanically isoalted, it would not be able to carry power.
