Good day. In this Pro Recording article, I will compare and contrast an algorithmic and convolution reverb. I will also demonstrate the difference and the important features in both types of reverb. I have recorded a short snare sample that I will demonstrate the different reverbs on. Here is the spectograph results of the snare sample without any reverb on it:
What is Reverb?
Reverb, short for reverberation, is an effect that is comprised of many different delayed signals which creates a background effect that is similar to sound in a room, bouncing around. When you are in a room and make a noise, the sound waves reflect off of various flat surface and this creates an echoey type of noise called reverb. Because all of these reflections happen at different times depending on how long it takes for the sound to reach the surface and back, reverb is basically an incredibly large number of varying echoes. The direct sound hits the listeners ears before the reflections. The difference between the two is referred to as pre-delay. All reverbs will have a dry/wet mix control which will control output signals as a mix between the dry inputted signal and the wet reverberated signal
Digital reverbs can use algorithms to reproduce natural echoes and simulate the effect of reverberation. The definition of an algorithm is a simple step-by-step procedure that is used in computer programming to logically outline the steps the computer will take to solve something.
Algorithmic reverbs use multiple feedback delay circuits which creates a diminishing series of sounds. The early reflections are controlled by variables such as room size, room shape and stereo controls. The reverb time is generally less than 2ms. The computer produces a reverb sound based on the settings on these controls which controls the variables in the algorithm.
The important features are the customization abilities with the room size and shape controls. Another important feature is stereo capability which allows the reverb use up some space in the width of the mix. A popular example of an algorithmic reverb VST plugin is “Ambience” which can be used to create a variety of reverb sounds. Here is a sound sample of a snare with algorithmic reverb on it and the spectograph results:
Convolution reverb is the digital simulation of reverb using a pre-recorded sample (called an impulse response) of the modeled space. The computer generates output based on the convolution of the samples of input audio and the samples of the reverb file.
Convolution is an advanced mathematical equation similar to cross-correlation that produces a third modified version which leaves an overlap that can be utilized to create this reverb effect. These can be real spaces, such as an auditorium or exotic places such as a famous concert hall.
An important feature of convolution reverb is its ability to mimic a variety of different room by importing lossless audio files. A popular convolution VST reverb plugin is Liquid Sonics Reverberate LE which comes with a few samples and also has the ability to import flac, wav and other impulse response samples. I especially like how this plugins shows the visual graph depicting the amplitude of the sound over time. Here is a sound sample of the snare with convolution reverb on it and the spectograph results:
- Convolution reverbs are not designed to be as customizable as algorithmic reverbs.
- Convolution reverbs may be more “creative” than algorithmic reverbs.
- Convolution reverbs are more “real life” than algorithmic reverbs which are strictly computer generated.
- There are many more algorithmic reverbs available.