FabFilter Pro-L online help

Table of contents

Oversampling

The limiting algorithm often needs to make very quick changes to the audio in order to remove peaks while preserving transparency and apparent volume. These sudden changes can introduce aliasing, which causes distortion and generally reduces the quality of the audio signal. Oversampling is a way to reduce that aliasing by running the internal limiting process at a sample rate that is two or four times higher than the host's sample rate.

"When do I need to turn on oversampling?"

You need it more when the limiting process operates faster (using short lookahead times), and when limiting more heavily, both leading to a higher level of aliasing. The aliasing will cause inter-sample peaks and these can cause distortion later on, for example during D/A conversion or conversion to mp3. There are only two small drawbacks to oversampling: it increases CPU usage, and it can introduce a very slight pre-ring due to the phase-linear filtering that is needed. Generally this effect is so small that it's inaudible, but it's good to be aware of this and not blindly assume that oversampling is always better.

"Why can my output level exceed the specified Output Level setting when oversampling is enabled?"

When using oversampling, limiting is applied to the upsampled audio (two or four times the normal sample rate), ensuring that no sample value in the upsampled result will exceed the specified Output Level. However, even though most aliasing is filtered out during the final downsampling stage, still some inter-sample peaks may exist. Because of these peaks, the downsampling process which reconstructs the audio in the original sample rate, can generate waveforms with a slightly higher level than the specified Output Level. The amount of that overshoot highly depends on the speed and amount of limiting. In most cases, using a minimum lookahead time of 0.1 ms keeps the overshoot within the range of 0.1 dB.

Using ISP metering and oversampling

As we just explained, oversampling might already reveal the presence of inter-sample peaks in the resulting audio. But to fully expose them, you should use ISP metering. This also clearly visualizes the benificial effects of oversampling. Using 4x oversampling will dramatically reduce the inter-sample peaks, which in turn allows you to increase the desired output level without inter-sample clipping. You can also clearly see that if you use a slightly longer look-ahead time, there will be less inter-sample peaks, both with and without oversampling.

Notes

  • When you use very short look-ahead times of less than 0.1 ms (clipping), it may become difficult to keep inter-sample peaks within a workable range. Again, in most cases, choosing 4x oversampling in combination with a minimum lookahead time of 0.1 ms, keeps inter-sample peaks within a small range of 0.1 dB.


Next: Dithering and noise shaping

See Also
Overview
Recommended workflow

Table of contents