Reverb - Chamber / Concert Hall / Cathedral

Getting Started
1

Hi,

I need to create three options so the user can simply select one of those options without having any knowledge of what low-level settings need to be selected to create the desired reverb.

Two out of three options should represent two different extremes e.g. Chamber and Concert Hall and one very different from the other two (e.g. Cathedral).

Looking at parameters in Reverb class JUCE: Reverb::Parameters Struct Reference
What values should I set for each of the following options to give maximum effect?

Chamber

roomSize = ? (probably 0)
damping=?
wetlevel=?
drylevel=?
width=?
freezemode=?

Concert Hall

roomSize = ? (probably 1)
damping=?
wetlevel=?
drylevel=?
width=?
freezemode=?

Cathedral

roomSize = ? (probably 1 but what will make Cathedral sound different from Concert Hall - not sure)
damping=?
wetlevel=?
drylevel=?
width=?
freezemode=?

Please suggest.

2

I am not an expert on reverb algorithms, but the difference from a cathedral and a concert hall is the material. The old stone cathedrals have no means of limiting the reverb, so you will notice especially high frequencies not decaying as they would in a more wooden room or even with drapes or other acoustic treatment.

Also cathedrals usually have quite simple shapes, so the echo directions will not be very diffuse but very direct, while a good concert hall aims for a very diffuse reverberation.

I wouldn’t expect too much detail from the juce reverb.

3

I agree but ultimately I think they want to know what parameters best fit the reverb they’re trying to go for, rather than anything highly accurate (which is a whole field of study in of itself)

Ultimately those parameters would probably be best determined via trial and error, but the JUCE reverberator is relatively simple (and the documentation for the reverb seems like it’s catered moreso to audio developers then standard musicians). I’ll further elaborate on what the parameters are actually doing:

  • roomSize adjusts the gain level of the internal feedback comb filters, resulting in a longer tail.
  • damping does a little filtering operation with the comb filters to attenuate higher frequencies, resulting in a faster decay of high-frequency sound in the wet signal.
  • wetLevel is just the gain value for the wet signal.
  • dryLevel is the gain value for the dry signal.
  • width two signals are being processed in parallel in the reverb if it’s running in stereo mode, and width combines these into a mono signal at 0, or keeps them seperate at 1.
  • freeze creates an endless reverb tail by feeding the output back into the input without gain attenuation (and no longer takes input) and probably isn’t anything related to what you’re going for.
4

Thanks @daniel and @momentumm.