Digital Audio II - Bit Depth and Sample Rate

Introduction

In our first post, we saw that digital audio is a "photograph" of analog sound. Now, let's analyze the two elements that define the quality and fidelity of this photograph: Sample Rate and Bit Depth.

Understanding these two concepts is, perhaps, the most important step to comprehending sound in the digital domain. They are the fundamental settings you will configure in your DAW (Digital Audio Workstation) even before hitting the record button.

Sample Rate: The "Photos" per Second

The Sample Rate determines how many times per second the Analog-to-Digital Converter (ADC) "photographs" or measures the amplitude of the sound wave. It is measured in Hertz (Hz) or kilohertz (kHz).

• Visual Analogy: Think of Sample Rate as the "frames per second" (FPS) of a video. A video with more frames per second captures motion more smoothly and in greater detail. Similarly, a higher Sample Rate captures the sound wave with more points of information, resulting in a more accurate representation of frequencies, especially the higher ones.

  
  

How does this affect the sound?

The main function of the Sample Rate is to determine the maximum frequency range that can be recorded. According to the Nyquist Theorem, the sampling rate must be at least twice the highest frequency one wishes to capture. After all, we need at least 2 points to determine a complete cycle for each of the frequencies.

Since the human ear hears, on average, up to 20,000 Hz (or 20 kHz), the sampling rate would need to be at least 40,000 Hz (or 40 kHz).

Common Sample Rate Values:

• 44.1 kHz and its multiples: The standard for audio CDs. It was established because it is slightly above double the range of human hearing (20 kHz x 2 = 40 kHz), ensuring the capture of the entire audible spectrum.

• 48 kHz and its multiples: The most common standard for audio in video, films, and broadcasting. It offers a small extra margin of safety compared to 44.1 kHz. It is the most popular choice for most music productions today, mainly for its ability to sync with video.

• 96 kHz (or higher): Used in very high-fidelity productions. Although the benefit in terms of audible frequencies is debated, many engineers argue that higher rates improve the response of filters and processing within the DAW, resulting in fewer digital artifacts. Note that the way we evaluate sample rate changes whether we are talking about recording versus sound reproduction.

  
  

[Image link: https://cdn.shopify.com/s/files/1/0970/0050/files/Sample_Rate.png?1870870584085320152]

On the left, we have a lower Sample Rate, and on the right, a higher one. The higher the sample rate, the greater the definition in the recreation of the sound wave in the digital domain.

Bit Depth: The Richness of Detail

If the Sample Rate defines how many photos we take of the sound, Bit Depth defines how detailed each of those photos is. It determines the number of "steps" or possible values to register the amplitude (the volume) of each sample.

• Visual Analogy: Think of Bit Depth as the color palette or pixel resolution of an image. An image with a 2-bit depth would only have 4 colors to form itself, resulting in a "pixelated" and poor visual. A 24-bit image has millions of colors, creating smooth gradients and rich details. In sound, a higher Bit Depth offers more "steps" to measure amplitude, resulting in greater precision.

  
  

How does this affect the sound?

Bit Depth is directly related to the Dynamic Range of the audio. The Dynamic Range is the difference between the quietest sound and the loudest sound we can record without distortion or having the background noise mask the signal.

Each additional bit in the depth adds approximately 6 dB of dynamic range.

Common Bit Depth Values:

• 16-bit: Offers 65,536 amplitude values and a dynamic range of 96 dB. It is the CD standard. It's functional, but in a recording environment, the "noise floor" is higher, requiring more care with recording levels.

• 24-bit: The industry standard for music recording and production. It offers over 16.7 million values and a dynamic range of 144 dB. This vast range allows for recording with much more "headroom," without worrying that the quietest sounds will be lost in the noise or that the peaks will distort.

• 32-bit float: Used internally by most DAWs. The great advantage is that it has an almost infinite dynamic range, making it impossible to "clip" or distort the audio within the digital environment. This is why you can lower the fader of a channel that is "in the red" and the sound returns to normal without distortion.

  
  

Summary: Sample Rate vs. Bit Depth

To simplify it once and for all:

• Sample Rate (horizontal axis): Precision in time and frequency. Related to the smoothness of motion.

• Bit Depth (vertical axis): Precision in volume and dynamic range. Related to the details and depth of the image.

  
  

For a high-quality recording, both are crucial and work together to create a faithful digital representation of the original performance.

Conclusion: Which Settings to Use?

For the vast majority of music productions, recording at 24-bit and 48 kHz is the perfect combination of high quality, processing efficiency, and compatibility.

In our next post, we will explore how AD/DA converters work in practice and what "dither" is.

Have any questions?

Leave a comment!