Compression?
Hi Winslow,
I was wondering if you had an opinion on whether using compression (Dynamic Range Compression) is a good idea when mixing, and is it used often?
After all, we are taught from the beginning the importance of varying dynamics.
I'm asking because I recently re-recorded an early study song as a recording/mixing/editing practice, and after much tinkering I couldn't set the harmonica gain at a level that I was happy with consistantly throughout the song. Either the soft parts were too soft, or the louder parts were slightly too loud to the point that it didn't feel like it fit into the mix any more.
Is it likely to be a playing problem (playing soft part too soft/ loud part too loud), or a recording problem (too close/too distant), or a tech problem (poor mic, software)?
In the videos, David frequently tells us he only uses EQ, delay and sometimes reverb. But surely that refers to amplified play. I don't even have an amplifier. I feel like playing amplified would naturally compress your output somewhat, since a mic can only output up to a certain amount, and then anything louder just adds distortion, but not volume. So maybe compression isn't needed when playing amplified, but it might be appropriate in small amounts (low ratio) for acoustic playing?
For this recording, what I ended up liking the most was applying EQ (roll off the extreme low frequencies, slightly up in the mid), compression (ratio 2:1), and no delay or reverb.
Cheers. - Jeremy.
Hi Jeremy! Let's not forget that mixing is not a static process. You rarely set up everything at the onset and let it go for the whole piece. Volume of independent instruments is a variable that will vary a lot within a song; modern softwares even have an automation function to handle that and to offset the fact that the board operator only has two hands. Back in the analog days, mixes of classics such as Pink Floyd's songs could require up to five people at the board and tedious repetitions before committing the mix to tape.
If whole sections are too loud or soft, then I'd say that's a matter of playing dynamics, distance from the mic, etc.
Where I find compression helps is with the difference between loud and soft within a phrase. I remember playing acoustically in a group of mostly fiddlers, maybe 15 people, and the leader remarking that he could hear my attacks but not my sustains (I think he was surprised that he could distinguish a harmonica at all in that sonic environment). And when I record myself and examine the results, I note a similar phenomenon, though I've made improvements since that observation, made maybe 15 years ago.
Compression is widely used in recording, both for individual instruments and for overall mixes and sub-mixes. Look at the waveform for any popular recording, and you won't see a lot of peaks and valleys. The signal is cranked to really stand out and dominate the aural space of the listener.
Individual artists vary in their use of signal processing tools.
David may not favor using compression, and I suspect that if you queried artists who use the traditional tools of amplified playing, you might not find it being used as such. That said, bullet mics tend to to do their own EQ and compression internally by how they process sound.
Less traditional artists may have a whole effects chain that includes compression; Jason Ricci comes to mind.
Yes, compression can reduce or even obliterate dynamic shading. Like any tool, it can be over-used. Learning to use it well is something that takes patience and experimentation.