Samplerates: the higher the better, right?

Summary

The video script delves into the intricacies of samplerates and oversampling in audio production, highlighting their impact on sound quality and the importance of understanding the Nyquist limit. It explains how higher samplerates don't necessarily improve audible sound quality but can help avoid aliasing when using non-linear effects. The discussion includes practical demonstrations using various audio editing tools and explores the benefits of oversampling within plugins versus increasing the project's overall samplerate. Ultimately, the script emphasizes that while technical aspects are important, the ultimate goal is achieving a pleasing sound, regardless of the specific settings used.

Takeaways

• 🎥 Higher sample rates do not improve audible content quality below 20kHz but can reduce quality due to increased intermodulation.
• 🔊 The benefits of higher sample rates are only seen when using non-linear processes that add extra harmonics above the Nyquist limit.
• 🌐 Oversampling individual non-linear stages is usually more effective than just increasing the project sample rate.
• 🚫 Aliasing is often inaudible in most cases, even without oversampling, and does not necessarily render a plugin useless.
• 💡 If a processor struggles with the demands of oversampling every plugin, it may be necessary to scale back the production ambition.
• 🎶 Ultimately, if the audio sounds good, the technicalities of sample rates and oversampling become secondary.
• 📺 Standard definition TV had a resolution of 720x576, which was acceptable in the past but is clearly inferior to modern HD video with 1920x1080 pixels.
• 📀 CDs use a sample rate of 44.1kHz and store each sample with 16 bits of resolution, while modern high-resolution audio uses 96kHz and 24 bits.
• 🔉 A digital system can represent all frequencies perfectly as long as they're less than half the sample rate, allowing CD quality audio to recreate a 20kHz sine wave.
• 🌀 When non-linear plugins like saturators or compressors are used, oversampling can prevent aliasing and improve the quality of the distortion effect.
• 🛠️ Using oversampling with non-linear plugins can reduce unwanted intermodulation below 20kHz and result in a cleaner, more analogue-like distortion.
• 🔄 Running multiple non-linear stages in series benefits from band-limiting the signal between each stage to prevent aliasing and intermodulation.

Q & A

What is the main topic of the transcript?

The main topic of the transcript is the discussion of samplerates and oversampling in audio production, particularly in relation to dynamics plugins like compressors and limiters, and saturators like Saturn.

Why might higher samplerates not always be better in audio production?

Higher samplerates do not improve the quality of audible content below 20KHz and can result in lower quality due to increased intermodulation. They are beneficial only when using non-linear processes that add extra harmonics above the Nyquist limit, but in such cases, it's usually better to oversample each non-linear stage individually rather than just increase the project samplerate.

What is the golden rule in digital audio systems?

The golden rule in digital audio systems is that a digital system can represent all frequencies perfectly as long as they're less than half the samplerate.

How does oversampling work in the context of audio plugins?

Oversampling in audio plugins involves upsampling the input signal to a higher rate before processing, which allows for more accurate processing of high-frequency content and reduces the risk of aliasing. After processing, the signal is then downsampled back to the original rate, with the unwanted high-frequency harmonics filtered out.

What is aliasing in the context of audio and what causes it?

Aliasing in audio is a form of distortion that occurs when a signal containing high-frequency content is not sampled at a high enough rate to accurately represent the original signal. This results in the high-frequency content being incorrectly represented as lower-frequency content, leading to a harsh and unmusical sound.

Why might a higher project samplerate not be the best solution for dealing with aliasing?

A higher project samplerate can help avoid aliasing by providing more headroom for high-frequency content, but it also fills up that extra headroom with inaudible content and can lead to increased intermodulation. Additionally, it requires more processing power and larger file sizes, which may not be practical or necessary for the desired outcome.

What is the benefit of using oversampling in audio plugins like Saturn and Pro-C2?

Oversampling in audio plugins like Saturn and Pro-C2 allows for more accurate and higher-quality processing of audio signals, particularly when dealing with non-linear effects like saturation and compression. It provides extra headroom for the creation of harmonics without audible aliasing, leading to a more natural and musical sound.

How does the use of oversampling filters in Pro-C2 affect the audio and the CPU usage?

Pro-C2 uses linear phase oversampling filters, which add extra latency and an extra CPU hit. This comes on top of the cost of running the processing at a higher rate, but it helps to avoid aliasing and maintain audio quality.

What is the significance of the 48KHz samplerate in comparison to 44.1KHz?

The 48KHz samplerate allows for less severe anti-aliasing filters compared to 44.1KHz because it has to remove content above 24KHz instead of 22.05KHz. This gives the filters more room to work in, potentially leading to better audio quality and allowing plugins that use internal oversampling to relax their filters slightly.

What is the conclusion drawn from the discussion about samplerates and oversampling?

The conclusion is that simply increasing the samplerate is not always the best solution for improving audio quality. It's more effective to use oversampling for each non-linear stage individually. However, in most cases, aliasing is subtle and difficult to hear, so the focus should be on achieving a good sound rather than strictly adhering to higher samplerates or oversampling.

How does the use of multiple non-linear stages in a signal chain affect the audio quality?

Using multiple non-linear stages in a signal chain can lead to increased aliasing and intermodulation, especially if the signal is not properly band-limited after each stage. This can result in a loss of clarity and the introduction of unwanted artifacts in the audio.

What is the role of brick wall low pass filters in a multi-stage non-linear signal chain?

In a multi-stage non-linear signal chain, brick wall low pass filters are used after each non-linear stage to remove extra high-frequency content. This ensures that the next non-linear stage is processing a band-limited signal, avoiding aliasing and intermodulation, and making the overall setup more efficient and cleaner.