Join Dave Schroeder for an in-depth discussion in this video Bit depth, part of Digital Audio Principles.
The other thing we need to think about when thinking about analog to digital conversion is bit depth. But before we go into that, I want to go over a couple of definitions that'll help us understand bit depth. The first, there is Dynamic Range, or the available range of volumes between the loudest and softest audible sounds. Now this isn't like your home stereo, like you can turn it from zero up. It's not how loud something can be, or how loud it can go. It has more to do with the difference between the loudest sound and the quietest sound. An example would be if you're watching a movie, and someone sneezes, and then there is a four car pileup and a giant explosion.
Chances are that the volume difference between those two sounds is pretty good. The sneeze is a little quiet cache, and the explosion is a big thing that almost knocks you off your chair. That's a good way to think of Dynamic Range. It's the difference between those two sounds and what difference is available. The Dynamic Range is that range, it's how much of a difference you can work with, how much louder can the explosion be than that sneeze. Dynamic Range is important, because the greater your Dynamic Range is the greater the Signal-To-Noise Ratio is.
Now the Signal-To-Noise Ratio refers to the difference in volume between what you want to hear, the signal, and what you don't want to hear, the noise. The greater a Signal-To-Noise Ratio is the better. So when you turn the volume of something up, you can hear quite a bit of the sound or the music without also having to hear some of the noise. So by signal, we mean the things we want to hear, like the music, or the voice-over. Whatever we're working on, and recording, and producing. By the noise, we mean things like room hiss or background air. Actually, here I'll be quiet for a minute, if you turn up the volume, count to three.
Turn up the volume a little bit. See if you can hear just the general background hiss or noise of this recording. Ready, I'll be quiet for three seconds. You can turn it up, and then turn it back down, one, two, three quiet. So if you actually did turn up your speakers there at that point, you probably would've heard some of the hiss or background noise that's taking place even in this recording. Finally, let's talk about the Digital Dynamic Range. This refers to the volume range of a digital audio file.
It goes from negative infinity or silence up to zero, or the loudest the file can be. Now we're talking about a file, and we're talking about the volume range that's fixed in that file. You can take a CD home from the store, put it in your stereo, and turn it way up, or turn it way down. That's not what we're talking about. But on that CD itself, there are digital audio files. But those audio files on that CD have a Dynamic Range. They can only go from negative infinity up to zero. In a digital audio file, any sound that goes above zero is actually where we get distortion or clipping. So let's take a look at bit depth.
Bit depth itself dictates the number of discrete volume levels that can create the dynamic range available in digital audio. In other words, it's how many different volume levels a sample can choose from when it takes a look or makes a note. So we have our sampling rate, and it's taking samples. It takes a look at an audio file, or takes that sample. Part of the thing it wants to record, or make a note of, is how loud that sound is. The bit depth of a digital audio file determines how many different volumes it can pick from to say it's this loud.
A simplified version would be to say, we have four different intervals like on your volume knob it goes zero, one, two, three, four different intervals of sound. When we take a sample, it says, okay, this one is two. On the next sample it says, oh, this one is as loud as three. This one is as loud as zero. So that would be one bit depth. But now if we had maybe eight intervals, so from zero to one, to two, to three, to four, and so on, up to eight, then it can go and say, well, this one is two. This one is four. It's not necessarily that four is twice as loud.
It's that there is a greater degree of discrete intervals there that it can choose from, so I can create a better representation of what the actual volume is. So the greater the bit depth, the greater the available dynamic range. So let's take a look at a few different common bit depths that you'll see in digital audio, and the amount of dynamic range that they'll provide. There is this 8-bit, which only provides 48dB, 16-bit, which provides 96, 24 at 144, and 32 at 192, and so on. A couple of real world examples here, when you buy a CD a music CD from the store, it's at 16-bit. So you can get 96dB of range out of there.
But if you get a DVD, usually that audio sound there at 24-bit. So that in the movie you can have sound effects and music that can have a greater range of volumes. It's important not to confuse bit depth with bit rate, which you probably know bit rate from MP3s. It's one of the settings you can pick. If you've used iTunes or compressed some of your CDs into compressed formats, you've probably selected something called the bit rate. This is entirely different from bit depth. We'll talk about it later in another movie when we talk about compressed file formats. But for now we're talking about sample rate and bit depth.
Sample rate and bit depth work together. So it's a combination of sample rate and bit-depth that determine the overall quality of the sound. An analogy I like to use for sample rate and bit depth is that of camera. If you think about how many pictures your camera can take in a second like a speed shot, that's the equivalent of the sample rate. It's the number of shots per second. In terms of bit depth, it's kind of like how many colors a camera can discern. Your camera might take a black and white photo, and see four different shades of gray, or a nicer camera might be able to see eight different shades of gray.
Well, obviously, the more shades of gray you can see, or discern, when you take a picture, and make that recording, the more accurate your photograph is going to be of what's actually happening. So here we kind of have a rough mockup of bit depth and the dynamic range. The greater the number of bits, the bigger the range between infinity and zero, the more different volume levels you can have access to. So here is a visual that shows bit depth as it pertains to digital audio. The greater the bit depth, the greater the number of volume intervals between infinity and -0, which is the loudest.
So it's clear that a higher sample rate and a higher bit depth make for better audio. But it's not quite that simple. There are a few things you have to take into account. The main concern is that the higher the sample rate and the higher the bit depth, the bigger your files are going to be. We all know that file size is something to keep in mind. When you're working with digital audio, you'll be dealing with very large file sizes. So here I kind of have a list of the combination of sample rates and bit depths, and these different file sizes they have. So we're looking at for One minute of mono digital audio (uncompressed) at 44.1 kHz and 16-bit depth.
A bit depth of 16, we'll be looking at 5.1 Megabytes for that single minute of mono audio. You can see as we go up, it gets bigger. 96 kilohertz set a bit depth of 24, 16 Megabytes. 192 kilohertz at 24-bit, 33, and the highest available I think right now is 192 at 32-bit. That's 44 Megabytes per minute of mono audio. So as you can see, this can really add up, especially if you're doing a lot of multitrack recording.
So the reason we're pointing this out is that you have to make some decisions based on the quality you want in relationship to how big a file you want to have, or how much disk space you have, or how much you want to have to work with. So working with sample rate and bit depth, and making decisions about which one is to be used is more than just always go for the highest best quality, because you have to kind of figure out what kind of file sizes you'll have to dealing with. Make just a few choices on what's appropriate for the scenario. Now I'm not going to tell you which bit depths to use, but I will give you a little bit of information here.
As I mentioned, store-bought CDs come at 44.1 kilohertz, 16-bit. You've been listening to those for 15 or 20 years. You can tell they sound pretty good. So that's not a lousy quality by any means. It's pretty good quality. A lot of times we'll work in our digital audio software at higher sample rates and bit rates, but when we actually export it, or put that on to a CD, or use it to create an MP3, we have to go back down to 44.1 and 16-bit anyway. A lot of newer devices are capable of playback at higher sample rates and higher bit depths.
So a lot of these higher sample rates and bit-depths are very relevant. I always say it's good to figure out what you think your final delivery mediums is going to be, or who you think is going to be listening to it, and in what capacity, and from there make some decisions about what you're going to use. I would say that using 24-bit to get a greater dynamic range when you're doing music and voice-over recording at 44.1 is usually worth it. It takes up a few more megabytes, but usually there is a payoff there. It's very common now if you're going into a professional recording studio to have your music recorded at 96K at 24-bit.
Today, I'll just go out and would say that, if you go to a professional recording facility, chances are they're recording at a sample rate of 96K and a bit depth of 24. So that does it for sample rate and bit depth. As you can see, sample rate and bit depth are the two main factors that go into analog-to-digital conversion. There are two things that you want to keep in mind, you'll find yourself making decisions about these settings when you start new recording sessions, or when you export or import files. So it's good to have an understanding of what these things are. We'll see sample rate and bit depth in a lot of different areas when you're working with digital audio in terms of importing things, or setting a session bit rate or sample rate.
So it's good to be aware of what sample rate and bit depth are, and how the different settings you can choose will affect the quality of what you're working on.