Quote:
Originally Posted by dnf777
I see the emoji, so I'm not taking your comment entirely literally, but I do find it interesting. I once asked here if much of the less dynamic acoustic folk-type music really needed compression as much as most other styles of music with bigger peaks and troughs. One reply in particular really caught my attention; that "when used properly, compression should be barely noticeable". Well heck...that could save lots of money and time! But I get it. I think.
I go back and forth on whether I think I understand compression. For now, I'm going to take some tracks and mixes and really play with them and see what I learn. I made a lot of progress on Reaper this weekend, so its now a little bit easier to play around with fx and hear real time results.
|
I'm gonna go down this road although I ought not to.
Find 1 minute of an audio track (preferably an acoustic) you've recorded. Call up your volume automation in Reaper. Without too much micro-managing draw an upside-down volume dip on all the peaks across that 1-minute piece of audio. Don't sweat the details, you may or may not pick all the peaks that sonically need to be addressed but that's fine for this demo. For the purpose here in, I'd bring each peak down -3dB. Bear in mind this is an experiment not a finished audio solution.
At this point, each peak should now be aligned (volume-wise) more consistently with the other, smaller peaks that inevitably occur. Ostensibly you've brought higher peaks in uniform with the lower peaks. Two things now happen. Since you've lowered the larger peaks down -3db you're free to raise the volume of the overall track since you've now gained 3db of headroom. Secondarily (or maybe primarily) the phrasing of the 1-minute acoustic audio is now a consistent, even track. Sort of like a great player with a great right-hand attack.
It's important to note that this technique is not always called for especially when the intent is a broad dynamic acoustic guitar track. That is to say you "want' to hear the high and low of a track. say for instance a quiet passage that evolves into a heavy strumming. In that case, leveling peaks would destroy the intent.
A compressor works similarly although the above example is singular in its intent, that intent is of course leveling a track's output. A compressor can take that methodology several steps further. Common compressor parameters include the ability to address the Attack. The attack is how quickly the compressor itself essentially turns on. A very slow attack, on say a snare drum, would allow for the wack of the initial snare hit but then slowly kick in post wack. There is also a
Threshold parameter on compressors. Threshold is fairly simple. You simply set a dB level in which you wish the compressor to start..compressing. If for instance, I set a very low threshold, say -2db EVERYTHING audio-wise below -2db remains uncompressed (more or less, depending on the type and brand of the compressor). The compressor simply leaves that audio untouched. No compression exists until the audio exceeds -2db. Another control parameter of compression is the
Release. Again a very simple concept. It merely tells the compressor how long it should attempt to hold on to a note, or even a passage before it stops processing. That brings us to the
Ratio parameter which seems to be the most misunderstood. The Ratio specifies the amount of compression applied to the signal. This setting is expressed in decibels. For example, a ratio of 2:1 indicates that a signal exceeding the threshold by 2 dB will be attenuated down by 1 dB. A signal exceeding the threshold by 8 dB will be attenuated down by 4 dB, etc. Finally, there is often a "make-up gain" control that does just that. Makes up for gain. If one compresses a signal to the point it's lowering the volume this gain control can be handy.
Tryin' to summarize here: There is no doubt that a good compressor has more tools in the toolbox than good old-fashioned volume automation BUT, and it's a big but (at least for me) there's a sonic price to pay in deferring to compression. Going through a compressor, no matter how lightly used, creates an audio footprint. Some are very strong and I tend to recoil, like fingers on a chalkboard, others subtly but still mostly obvious, unless of course, one gets into some of the great hardware boxes out there, and honestly at that point the game changes for the better concerning compression, but at a substantial financial investment.
In the end, I can use volume (or clip gain) automation to do most, if not all the tasks a compressor provides and in many, if not most, cases a much better job, and sonically (at least for me) substantially better audio quality.