A convolution reverb reproduces the sound of a real space or a piece of gear using audio data measured from it. You can call up the character of halls, studios, plates, and springs. Unlike reverbs that build the tail with an algorithm, the character of the IR you load is the foundation of the sound.
Recreating a space with an IR
Convolution is the process of combining the input sound with an IR (impulse response) by calculation. Adding the reflection and decay characteristics captured in the IR to the input creates the sense of having passed through that place or that device. Its strength is how well it reflects the character of the recorded space.
Which IR to choose
A short room IR gives a close sense of space, while a long hall IR creates width and a long tail. Plate and spring IRs give you the color of the actual devices. Do not decide by name alone; compare them against the tempo of the song, the length of the source, and the sense of distance from the other instruments.
Natural spaces and sound design
IRs are not limited to real spaces. Using a short percussive hit, noise, or a metallic sound, you can create textures that do not exist in reality. When you are adding a natural space, keep the amount low enough that the position and the clarity of the dry sound are not lost. For sound design, deliberately choosing an unnatural IR is also an option.
Points to watch when adjusting
If the IR itself is long, the sound gets crowded even without much reverb. Use pre-delay, decay, EQ, and Dry/Wet to separate the roles of the dry sound and the tail. When low-frequency tails pile up, high-passing the reverb signal is effective.
How it differs from algorithmic reverb
An algorithmic reverb builds the tail from parameters, so it suits sound design where you change the time and the density a lot. A convolution reverb starts from the character of an IR. Neither is better; choose according to whether you want to reproduce a sound or reshape it freely. For the basics, see the article on reverb.
