Almost all reverb plugins (as well as hardware digital reverb units) use one of these two digital processing methods.

Algorithmic reverbs use calculations based on hypothetical rooms and other spaces to generate their reverb sounds. Generally this gives a sharper, more artificial sound, typified by most hardware digital reverbs of the last 30 years. This is not necessarily a bad thing though – as mentioned above, musically we’re not always after the most ‘natural’ sound, but the one that has the right ‘character’ for the track. Algorithmic reverbs are also far lighter on the computer’s CPU than…

Convolution reverbs use pre-recorded samples of real rooms and spaces to build Impulse Response (IR) files of those spaces. The impulse response is then ‘convolved’ with the incoming audio signal you want to process, hence the name.

Convolution reverbs then, are generally far better at simulating real spaces than algorithmic reverbs – the only major downside is that they also require significantly more CPU processing power to work, so you are more limited in terms of the number of instances of the plugin you can run simultaneously.