In the first part of this article we talked about fast hand syndrome and how proper preparation can help to slow ourselves down, be more proactive in our artistry, and less reactive during the mix.
I want to continue the conversation by discussing a topic that is more technical in nature. I proposed last time that in my opinion one of the largest contributors to fast hand syndrome was a lack of control as it relates to audio dynamics. I know that this was something that I really had to work through in order to get more relaxed while mixing audio.
The term ‘dynamic range' refers to the difference between the loudest part of a signal and the quietest part of a signal. If you were to have an announcer on a microphone and the signal meter for that microphone reads -20 dBu when the announcer is speaking softly and goes up to -5 dBu when the announcer is speaking loudly, that would represent 15 dBu of dynamic range. That is, at the announcer's loudest point in time he is 15 dBu louder than when that announcer is speaking at his softest.
In the mixing world, some audio sources have a small dynamic range while others have a very large dynamic range. Interestingly enough, the human voice has a tremendous dynamic range. A practical example of this would be the pastor who speaks at a relatively normal volume but then gets excited and yells something out causing the congregation to jump in shock because of the sudden change in volume. Another opposite example is the worship leader who is a very strong singer but has a very quiet voice when speaking or praying (typically right in the middle of the song where it causes you to struggle to make them heard.)
As you can imagine the dynamic range of any two inputs (sources) are rarely the same. This can make mixing particularly interesting as most live content (especially music) can be quite dynamic as a whole. For instance, many of today's modern worship songs build in energy quite a bit between their start and their end. To help understand the concept, consider this contrived scenario; you have a worship team comprised of exactly two inputs. The first is a flute, and the second is a vocalist. Together they are going to perform a song. Generally speaking, a vocalist will have a much larger dynamic range than a flute will. At the beginning of the song you have them mixed such that the flute is a little bit softer than the vocalist, providing a beautiful accompaniment yet allowing the vocalist to sit forward in the mix. As the song progresses, however, it builds in volume. Each input, the flute and the vocalist, continually get louder until they are at the upper end of their dynamic range, meaning that they have gotten as loud as they can, respectively. The result, however, is that instead of the flute being a little bit softer than the vocalist, as they were when they started, the flute is now much softer than the vocalist and is getting lost in the mix. This is because the flute is not capable of growing in loudness as much as the vocalist is. Its dynamic range is smaller.
What is the sound tech's first reaction? My guess is that most of us would reach over to the faders and either pull the vocal down or push the flute up. If you were to add more instruments to this scenario, such as a bass, an acoustic guitar, a keyboard, and multiple singers, etc., you would find yourself mixing reactively to all of the changes in level throughout the course of the song. And with that, as they say, we're off to the races. To put it in terms of our previous conversation, you would be suffering from fast hand syndrome as you try to continuously match the dynamics of each individual input.
There are multiple solutions to this problem, though the most widely accepted solution is the use of compression. After all, asking a very talented singer to pipe down doesn't generally end up very well nor is it respectful. Trust me on that one, really.
The compressor is a device (currently being built directly into quite a few consoles) with the sole purpose if limiting the dynamic range of a signal. This is done by limiting the rate of growth that a signal is allowed to have once it has grown above some pre-set level (called a threshold).
The simplest form of compression is to set a threshold at which point the signal is no longer allowed to grow at all, regardless of what the input does. This form of compression is called limiting and it effectively sets the rate of signal growth to zero once it meets the threshold value. For example, if the loudest part of the vocalist's performance provides an input level of -5 dBu, we could set the limiter to a threshold of -15 dBu and the result would be that the signal coming out of the compressor never grows past -15 dBu, even though the vocalist is outputting more than that. We would have effectively taken the loudest 10 dBu of the vocalist's capability and thrown it away.
While limiting is a simple concept to understand, it can make a poor choice for managing dynamic range because it doesn't really sound natural. In our contrived example, if we were to limit the vocalist as described, we would get into situations where the flute is moving up and down the upper part of its dynamic range while the vocalist is being restricted by the limiter. This would result in the flutes volume fluctuating appropriately with the piece that they are playing, but with the vocalist having no change in volume at all. That just wouldn't sound natural.
A good solution to this problem is to modify our limiting example just a bit to allow the vocalist to have changes in level while still limiting the maximum signal. We can do this by bringing down the level at which the compressor starts reducing the signal (i.e. threshold) and having it do so at a rate (called ‘ratio') that is proportional to the growth of the input signal. So for instance, instead of limiting as we did above, we set our compressor so that it allows the output signal to grow at 1/2 the rate that the input signal grows. Now, let's reset the threshold such that this behavior starts when the input signal gets to -25 dBu. With this configuration the compressor will only allow its output to grow 1 dBu after the input signal has grown 2 dBu and it will start that behavior once the input signal reaches -25 dBu. In this scenario, the end result is identical to that of the limiting example. In both cases the output signal will not go over -15 dBu. However, the big difference with this method is that the output signal of the compressor is allowed to move up and down the upper part of its (now smaller) dynamic range reflecting what the vocalist is really doing, rather than throwing that movement away as the limiting example did. This is a much more natural sounding method of limiting dynamic range.
An astute observer will note that by compressing the signal this way, we have effectively reduced the maximum loudness that a signal could achieve and thus we are affecting how much signal is available to use for mixing. This would be a correct observation. This is why compressors generally have a gain knob on the output side of the unit that allows you to make up the gain that was lost by limiting the signal. Using this setting effectively shifts the entire (and now smaller) dynamic range up in level.
The three main settings that we have discussed here (threshold, ratio, and gain) are the three main controls that you find on most any compressor on a modern console. You may find other controls such as knee, attack, etc., that allow you to get pretty nuanced, but for someone just stepping into the foray of managing dynamics, these can be left alone until after you are comfortable with the main controls.
Making use of compression on your individual inputs allow you to effectively normalize their associated dynamic ranges to better match one another. Doing so will minimize your need to chase faders as signal levels ebb and flow and will allow you to spend more time making small adjustments to the mix to accentuate the details of what is going on. When you start getting really good at this you will find that you are making mix changes during songs that are less frequent and smaller in size.
The idea here is to get to a point where we are using the mix to draw people's attention to specific parts in the sonic presentation that you want them to experience. It really provides us with a more precise ability to paint an audio-based picture. This is akin to our friends in the lighting world using key lighting to focus an audience's attention.
While the use of compression may seem like some sort of complex magic, it is quite reasonable for any tech to learn and to make regular use of it.
As we mentioned earlier, practice, passion, and discipline are the bedrock for improving our technical capability. If you don't have much experience with compression, I highly recommend that you give it a try. Alternatively, if you have been doing this for a while maybe it's time to have some fun and play around with different settings or try a new application. Remember, God invested his time to give us our gifts and He loves to see us invest our time in making use of them and improving them.
