...watching input/output meters is sufficient if adjusting a sinusoidal tone, but a crude insufficient method of adjusting parameters (like attack time, depth of threshold). The best compression limiting is that which is "not heard", which is often far before it is seen just in an output meter.
I do it by ear anyway, and I need to squash it pretty good. I manage a combination local and remote meeting, and I can't mic the local room very well. And some of the remote people are hard of hearing. So I need it to be exactly full-scale regardless of how far each person is from the mic. It's taken a while to get an acoustic solution that is also acceptable to the other local people - had a collection of wireless cardioids on stands for a while, distributed around a large circle, and mixed with VoiceMeeter before going to OBS (still dry except for a "dumb mix" and a "telephone EQ" on the mix master) - but I think we're finally settled on a small circle around a single beamforming mic with automatic beam direction, still running through VM and the 'phone EQ. Still varying distance and varying volumes, so I still need the brickwall with lots of makeup gain.
For my own personal stuff, yeah, I'll lighten up, but according to my ears at least, I think I've done a pretty good job with the brickwall in the meeting. It *is* pretty much full-scale for everybody, and it doesn't sound half bad either, considering how aggressive I am with it. That's even with OBS's built-in compressor, which I don't really like so much either.
Better yet, is there another option for choosing a VST-2 (or VST3) Compressor/Limiter plugin of our choice in the filter adjustments? If so, how do I do that? (I would prefer to use professional WAVE plugins, etc.)
I think Google would have a better idea than I would, of how to get your preferred processing in VST format.
That said though, I'm in the process of building a new rig while still running the old one, and for the new one I'm using Ardour on Ubuntu Studio Linux to do all of the audio stuff:
ubuntustudio.org
Ardour is a quite capable DAW that comes preinstalled with Ubuntu Studio, which is all free, and also comes with a TON of plugins preinstalled, including the ones that became my favorite even before I starting using that platform:
lsp-plug.in
The finished audio then goes to OBS as an unchanged passthrough.
OBS v27 is also preinstalled, but you can update it to v29 by installing the direct PPA (see one of the sticky threads on the Linux support forum here), and then doing a normal system update.
