[ardour-users] Newbie help with dynamics in Ardour...

[Stephan Neuhaus]

> So I'm trying to figure out a good way to produce a final "mix"
> that eliminates the pops, brings the general volume up, but 
> still preserve some of the dynamic range.

As one correspondent has already said, you could use Jamin for 
mastering.  Jamin includes compressors for various frequency bands, so 
you can compress the dynamic range dependent on the frequency.

For a quick solution, though, I'd try this:

1. Put a compressor plugin on the track pre-fader
2. Set attack and release times to your liking (for speech, I'd choose 
something like 100 ms attack and 400 ms release, but your mileage may vary)
3. Set the threshold really low, like -20dB or so.  Your problem is that 
you need to boost those low parts
4. Set the ratio really high, like 5 to 7 or so.
5. Adjust the compensation gain so that you end up with -6dBFS on the 
track, *not counting the popping sounds* (you can see the maximum level 
next to the fader and you can reset the indicator by clicking on it)
6. Use gain automation to eliminate the popping sounds.  Having -6dBFS 
on the rest of the track will give you some headroom so that you can 
afford to be a bit sloppy in the gain automation.
7. (Optional) Boost the master volume so that you end up with a peak 
level of -2dBFS or so.  (If you have anything *above* 0dBFS, you *must* 
do something about it, because it will distort terribly once you export 
it to WAV.)

If you want to use EQ, use it before the compressor.  (Funny things can 
happen when you heavily compress voice tracks, since many microphones 
have frequency responses that vary with the distance from the 
microphone.  So you may really want to add some EQ and remove some bass.)

Speakers on the radio usually have a very low dynamic range; their 
voices have been heavily compressed (I'd estimate ratios of 10 or more), 
but since their mic positions are known and controlled, they don't need 
extreme settings for the threshold.

Fun,

Stephan