Setting Recording Levels
When recording to analog tape, most people try to push the recording level to the highest point they can without distortion. This is done to maximize the signal-to-noise ratio - in this case, the ratio of audio signal recorded to the amount of inherent noise on the analog tape medium. Recording at high levels on analog tape increases the clarity of the recording and keeps the inherent noise as low as possible. And in fact, many people even like the sound of slight analog distortion. This occurs when the tape is overloaded with audio signal (sometimes called "tape saturation"), and can create a pleasingly "warm" sound. The ability to do this is one reason that some people are still partial to analog recordings over digital ones.
However, recordings you make in your Pro Tools home studio can rival high-end analog recording made in professional studios! In contrast to analog tape, there's very little noise associated with digital recording, so the signal-to-noise ratio is higher, providing very accurate representations of your music. As with analog, most people push digital recording levels, but in this case to allow for the use of greater bit depth. As you will see in the next section, using the entire bit depth provides the most accurate "picture" of an analog waveform. Unlike analog distortion, however, overloading a digital track causes digital clipping - the audio signal is literally cut off at 0db, often creating a decidedly nasty sound. "Digital distortion" isn't pleasing like analog distortion can be, so digital levels should always remain below 0db (the clipping point).
Yet, even if you see the red peak indicator on one or several of your tracks, that doesn't necessarily mean your signal has been digitally clipped. Pro Tools has a certain amount of headroom between when the red peak light is activated and when the actual audio is clipped. Watch the meters but also use your ears to experiment with recording levels and find what works well for your system. In general, if a signal peaks very occasionally that's usually fine. However, if the signal peaks often then decrease the input level.
Sample Rate and Bit Depth
Now that you've had some experience recording digitally into Pro Tools, let's talk about two primary determinants in the accuracy of a digital recording: sample rate and bit depth. These two parameters provide a grid onto which the analog audio signal is plotted and determine how well the digital recording represents the original sound.
Here we see an analog waveform and a digital representation of it. Faster sampling rates and larger bit depths provide more accurate digital representations of analog sounds.
In discussing sample rate, consider photography as an example. If you set up a camera to take a picture of the sky once every hour, you could follow weather patterns in a very rough way. With this low "sample rate" you would probably miss many significant events. However, if you took pictures every second you would see much more detail (like the moment rain began to fall or the sun broke through the clouds).
Digital recording is like taking pictures of music at a speed determined by the sample rate. If the sample rate in your session is 44.1kHz, Pro Tools takes 44,100 pictures of your input audio every second. Each picture captures the amplitude (level) of the audio signal at that moment.
Each sample is digitally "mapped" to an exact digital value and converted into binary digits (or bits). The number of bits in the system is referred to as bit depth. The higher the bit depth, the more accurate the digital representation of the analog sound.
The precision of the amplitude value depends on the bit depth - small bit depths yield less precise representations of the audio signal. Back to our photography example: in color photography, a photo taken with a bit depth of 4 bits (24) would only allow 16 different colors (24 = 2 x 2 x 2 x 2 = 16). If the color of a photographed object is not precisely 16 colors allowed, the closest color would be assigned to it. Obviously, 16 colors can't possibly describe all of the shades and hues found in our colorful world. The same logic applies when describing the myriad nuances of sound with an audio signal - usually, the more bits the better. That's why depths of 16 bits and 24 bits are offered in Pro Tools: 16-bit resolution offers 65,536 (or 216) levels of audio amplitude and 24-bit resolution offers 16,777,216 (or 224). With bit depths this high, it's like having thousands or millions of colors to choose from instead of only 16.
For each new session you open in Pro Tools you're asked to choose between 16-bit or 24-bit recording, as in figure 3.2. Be aware that you can't mix bit depths and sampling rates within sessions; you'll need to convert any audio being imported if the file was recorded at a different bit depth or sampling rate than the session.
Beware: Recording at 24 bit resolution requires 50% more disk space than 16 bit recording. Using a sample rate of 48kHz or higher also uses more disk space