Is there a rule of thumb regarding audio dymanic range compression?

riredale wrote on 3/16/2004, 11:48 AM

I've noticed for years that my miniDV camcorders do some dynamic range compression on the audio coming in when in "auto" mode--they have to, since otherwise the sound would either be clipped or too faint much of the time.

In a studio setting, though, how do the pruducers of music CDs determine how much compression to use? Is there a general consensus that says something like, "average volume should be about 10db below peaks?" When I hear an orchestra on a CD, is there no compression at all? What about popular music? Vocals?

Finally, if I wanted to make audio recorded via Minidisc (no compression) sound like audio recorded via camcorder, what compression settings would I use?

Back to post

Comments

TorS wrote on 3/19/2004, 12:42 AM

Marc
The compression presets you find in Vegas (and Sound Forge) all have volume compensation set as default. Which means that they first compress the signal and then normalize it. The settings are adjustable, of course. You do not normalize prior to applying compression. In fact - once you get into the finer details of compression you do not need to normalize ever again.
Tor

StormMarc wrote on 3/19/2004, 1:44 AM

Thanks for the answer Tor,

I guess where i'm confused though is that if you take two instances of the same file make one peak at -10db and one at -1 and they use the same compression on both of them the results will be differant. So this tells me that all files should have some uniform starting point right?

Thanks,

Marc

RichMacDonald wrote on 3/19/2004, 7:51 AM

>Regarding compression, in my explorations I had tried applying 6:1 to the vocal and it made it too "strident", but changing this down to 2:1, thanks to this thread, worked wonders - it now adds a bit of strength without over doing things.

One thing worth trying is to dup the vocal track. On one track set the compression to something high like 5:1, cut the low and high frequencies, use no other fx and pan it center. Drop the level a bit. On the other track, set the compression to something like 1.5:1, boost the lows and highs to compensate for what you did on the first track, add some reverb, and perhaps some tiny delay between the left and right channels. Then blend the two tracks to taste with the idea that the 2nd track is "primary" and the 1st track is used for "strength". You want the 1st track to be felt, not heard. Note that the reverb and delay on the 2nd track are optional.

I use this approach primarily to draw out the vocals of live music recorded with a stereo on-board camera mic, but it can also work with a single mono track recording as well, afaik.