Thebigcheese
Member
I would like to see a "maximum gain reduction" (sometimes called "range") slider added to the OBS compressor. Alternatively, a dry level slider could serve a similar function. The reason for this is to have better control when using the compressor as a sidechain ducker. As it currently stands, the compressor works okay as a ducker, but since the amount of compression is tied to the incoming signal level, you end up with an undesirable result where the quieter you speak, the less the signal is reduced, but the louder you speak, the more it is reduced. This is basically the opposite of what you'd want as if you are talking loudly, you don't need much reduction. By allowing us to specify a maximum reduction amount, we could crank up the ratio and effectively have the ducker reduce the signal by a flat amount regardless of the sidechain level. The end result is that quieter spoken bits will be heard more clearly.
I'll try to explain better with an example. I have a compressor on my PC audio set to use my mic track as the sidechain input. I've set the threshold to -24 dB (as far as I can tell, the sidechain happens before any FX on the mic track, so it has to be set lower) and the ratio to 7:1. If I am talking quieter, I may just barely be hitting that threshold, maybe it's -23.5 dB, causing little to no gain reduction on the PC audio. By contrast, if I am yelling, I might be hitting -10 dB, causing something like 98 dB of reduction even though I would've been heard just fine without any reduction.
If I was able to set a maximum reduction of, say, -12 dB, and then crank the ratio up as high as it will go, then no matter how much I exceed the threshold, it will always reduce the PC audio by 12 dB. This means that my yelling will reduce it less, which is fine, but my quieter talking will reduce it by more so that I can actually be heard.
A dry level slider can do a similar thing. If I set the dry level (that is, the signal with no compression applied) to -12 dB, then the output signal will always be at least that loud. This isn't perfect as the overall signal level will increase when mixing the dry level in with the wet, but I imagine it's a lot easier to implement. Heck, it could technically be done now by just having the same audio source twice, one with FX and one without. But a maximum reduction slider (in dB) would simplify things and just generally work better.
I'll try to explain better with an example. I have a compressor on my PC audio set to use my mic track as the sidechain input. I've set the threshold to -24 dB (as far as I can tell, the sidechain happens before any FX on the mic track, so it has to be set lower) and the ratio to 7:1. If I am talking quieter, I may just barely be hitting that threshold, maybe it's -23.5 dB, causing little to no gain reduction on the PC audio. By contrast, if I am yelling, I might be hitting -10 dB, causing something like 98 dB of reduction even though I would've been heard just fine without any reduction.
If I was able to set a maximum reduction of, say, -12 dB, and then crank the ratio up as high as it will go, then no matter how much I exceed the threshold, it will always reduce the PC audio by 12 dB. This means that my yelling will reduce it less, which is fine, but my quieter talking will reduce it by more so that I can actually be heard.
A dry level slider can do a similar thing. If I set the dry level (that is, the signal with no compression applied) to -12 dB, then the output signal will always be at least that loud. This isn't perfect as the overall signal level will increase when mixing the dry level in with the wet, but I imagine it's a lot easier to implement. Heck, it could technically be done now by just having the same audio source twice, one with FX and one without. But a maximum reduction slider (in dB) would simplify things and just generally work better.