Skip to Content

Sample Rate and Bit Depth: Explained Simply

It’s likely that you’ve come across the terms “sample rate” and “bit depth” in your DAW and have wondered what they mean and how they affect your sound. I put the below guide together to help you understand the basics! 

The “sample rate” is the measurements of audio wave taken per second and is measured in Hertz (Hz) and Kilohertz (KHz). “Bit depth” is the range of amplitude information that can be measured in one sample and is measured in Bits. Computers use this information to create a digital signal from an analogue source.

When they’re set correctly, your recording will have a clean and clear sound with plenty of headroom for mixing and mastering.

Many audio engineers have different opinions regarding which sample rate and bit depth should be used for which purposes, which can get confusing at times. I’ll explain which sample rate and bit depth you should use and why, below!

What is sample rate and bit depth

Every time you record audio using a computer, you’re recording the sound waves made by your voice or instrument. These sound waves are “analogue” and need to be interpreted by your computer in order to record and reproduce them.

To do this, your computer takes quick snapshots, or “samples“, to measure each wave at different points in time to create a digital signal representing the original waves.

The sample rate is essentially how often your computer takes these snapshot measurements. A higher sample rate means more measurements. It’s measured in Hertz (Hz), which is the measurement of one wave cycle per second. Look at the below images for an example.

An image demonstrating a sample rate. The wave is measured 8 times.
An image demonstrating a higher sample rate than the above image. The wave is measured 16 times.

In audio production, we most often measure in Kilohertz (KHz) which is 1,000 measurements of a wave cycle per second. A common sample rate for recording audio is 44.1KHz, meaning that the wave cycle is measured 441,000 times per second!

When taking these measurements, your computer needs to record how loud the sound wave was at that point in time, which is called the “amplitude”. The bit depth determines how many amplitude values can be captured within in each sample. The higher the bit depth, the more accurate the amplitude value is, thus the closer the digital signal will be to the original sound wave. Look at the below images for an example.

An image demonstrating bit depth. There are 8 possible values for the wave to be measured at.
An image demonstrating a higher bit depth than the above image. There are now 16 values for the wave to be measured at.

An increased bit depth increases the signal-to-noise ratio (SNR), also called the dynamic range. This is the difference between the desired signal and the noise floor. The noise floor is most often made up of the electrical “hum” generated by your equipment.

By having more amplitude values to choose from, your computer can better differentiate between the desired signal and the noise floor. This will make the noise floor imperceptible in most cases, unless you have very noisy electronics.

Top tip: The lower you can keep the noise floor, the more headroom you will have when it comes to mastering. Make sure your equipment is set up correctly so you’re not getting any loud humming.

You can see why we need both sample rate and bit depth: they’re what your computer uses to record audio!

What sample rate and bit depth should I use?

This is a heavily debated topic in the audio engineering community, and there are some merits to both sides. I recommend that you record at 48KHz at 24-Bits.

  • It’s the standard requirement for music that accompanies video
  • You will be less likely to have audio errors in your sound than a higher sample rate
  • It’s is accepted by most online streaming platforms

We’ll look into the specifics as to why 48KHz at 24-Bits is best, below.

Why I recommend a sample rate of 48KHz

There’s a mathematical theory called the “Nyquist–Shannon sampling theorem” that states in order to reproduce a certain frequency, the sample rate must be at least double that frequency.

In context of an audio recording, a sample rate of 48KHz (48,000 samples per second) can produce a maximum frequency (the “Nyquist” frequency) of 24KHz which is the sample rate divided by 2. This is over the frequency of human hearing (generally between 20Hz to 20KHz).

On paper, it would make sense to record at the highest sample rate possible, but this can negatively impact your sound due to “aliasing”.

Simply put, aliasing occurs at when your computer tries to identify a frequency above the Nyquist frequency, which it can’t do. Check out this video to see a 20 second demonstration.

So why not record at 44.1KHz as this would record a maximum frequency of 22.05Khz, which still covers the human hearing range? The answer is that the anti-aliasing filters may roll off the frequencies around 20KHz, which may have a negative impact on the high-end of your recording. Some people argue that there is no difference with modern anti-aliasing filters, in which case 44.1KHz is fine.

48KHz definitely provides enough headroom for your anti-aliasing filter to work without impacting your hearable frequencies.

Most recordings will be “down-sampled” when uploading to a service like Spotify or iTunes. For this reason, I recommend avoiding any sample rate above 48Khz, as it may make your recording sound worse. This is because a higher sample rate will be recording more supersonic frequencies, which become audible when down sampled.

Why I recommend a bit depth of 24-Bit

As discussed, bit depth is the number of different values that can be recorded in one sample. The higher the bit depth, the more accurately the digital signal will track the analogue waves amplitude. This directly influences the signal-to-noise ratio (SNR) and dynamic range of the recording.

For context, a bit depth of 16-Bits gives your computer 65,536 unique amplitude values to choose from, which is a maximum dynamic range of 96.33 decibels (dB). 24-Bits provides 16,777,216 different values, which means a maximum dynamic range of 144.49dB.

Realistically, 16-Bits will give you perfectly usable results, but 24-Bit will give you that extra headroom to get more from your recording when it comes to mastering.

Summary

Sample rate and bit depth can be a complex subject to understand, and there’s definitely more I could go into, but this guide has given you the basics. To summarise:

  • The “sample rate” is the measurements of audio wave taken per second and is measured in Hertz (Hz) and Kilohertz (KHz).
  • “Bit depth” is the range of amplitude information that can be measured in one sample and is measured in Bits.
  • Computers use this information to create a digital signal from an analogue source.
  • I recommend recording and exporting at 48KHz and 24-Bit depth. This is also the maximum you should use.
  • The lowest sample rate and bit depth you should use is 48KHz and 16-Bits. Anything lower than this, and you will start to be impacted more by the noise floor.

Related Questions

Should I record using a high sample rate and bit depth?

As discussed, recording at a higher sample rate won’t give you any clearer sound, and will likely negatively impact your recording. There are instances when you might want to use a higher sample rate, though, such as recording for archiving. Some audio engineers prefer to have as much frequency information available as possible for archiving purposes.