Underdubbing

When we are mixing audio there is a temptation to always reach for EQ, Compression, or something even more fanciful. This is because we are in the mixing mindset, this is how you mix right. Its not that these tools don't have a place in the mix, they really do. Its just that sometimes we need to remember that taking things back to basics can be helpful, and so we instead reach for volume and panning controls. We try to balance what we have in front of us on the screen or coming through the console and apply the tools we are used to. The problem is that often, as with so much in audio, the cause of the issue is actually in the arrangement. There is simply too much going on. Trying to fit too many parts together at once can lead to a lack of focus and can render any attempt to massage the parts together as pointless.

Reasons

Accepting that the arrangement needs changing or altering in some way can prove one of the absolute hardest things to take on board for an artist. Their hand crafted, painstakingly sculpted work of art isn't perfect ? How dare you even suggest that ! (Cue Mozart in Amadeus proclaiming there are just the right amount of notes) Sadly it is often the case that a cluttered, unfocused arrangement can tie the hands of a mixer to a massive degree and some careful pruning would actually yield an astonishingly improved result.

So why does this occur? How does this parts explosion get started?

1.) Emotional attachment to parts

Artists are humans and as such they have emotions and feelings. Creative people are often very in touch with these emotions and use them to fuel their creative habit. The problem is that they can get very attached to parts throughout the writing process, even to the detriment of the track as a whole. They vividly remember the 6 hours of pain and misery trying to record that guitar solo, even though its going on at the same time as the lead vocal and blocking out the singer's tasteful vibrato.

2.) Lack of understanding of the function of parts

Often song-writers like the sound of a part but do not really understand the function it would play in a full arrangement. Each part should add something to the arrangement without stepping on other part's toes but also crucially adding to the full during of the arrangement. A part might fit beautifully at a certain point but could be to the detriment of the track when considered chronologically. Consider a crowded verse part when the chorus is crowded as well. The chorus will not soar, because of the verse part's fullness. It will sound weak in comparison.

3.) Good sounds in isolation

Sounds, when isolated, can often sound very good. The problem can be that they don't sound good in the context of the rest of the track. One particularly bad culprit for this is synthesiser patches. The point of many patches is not mix well but to advertise the sonic capabilities of the synth.

4.) Saturated ears

Often listening to the same parts of a song over and over, while writing, can lead to recognising the song in that form. A sort of over-saturation occurs because the brain can recall and latch onto that pattern within the song. This saturation leads to an inability to see the song as anything other than exactly what it is currently. The tendency of the songwriter is to then try to mix exactly that, and not consider other arrangement choices, which might yield better results.

Personal Reasons

I have personally come across this issue, within my own musical work, largely due to the my particular musical background. I am originally a guitar player, who picked up all the instruments I now play later. I started, fairly early on, recording on a computer and playing along with a metronome. I played in bands for a while but this became less and less a part of my musical life compared to solo work. It was only much later in my musical life that I had the confidence to start recording my own vocals.

As a guitar player my tendency is always to create music on guitar and then try to shoehorn the guitar parts into every bit of the track. Electric guitars in particular can suffer from a lack of dynamics if played from the top of the track to the bottom. They also eat up the frequency range like nobody's business. As such they are a fast track to creating mixes with no space to play with for other parts.

On top of this because I create the music, usually, before the vocals I almost always don't leave space for them in the mix. I create the track as if it was finished, including parts that are functioning as lead parts. Then when I do record them they sound thin and horrible because they are being crowded out. Everything sounds muddy and the mix never seems to sit in a way that blends.

Solution

The solution to this problem is something mixerman, in his book "Zen and the art of mixing" coined "Underdubbing". Essentially this means dropping parts when you have the full mix available to you. Its a simple process of editing yourself, judiciously. If the part doesn't add something, remove it. If the part detracts from another part, pick one or the other. Whenever I finish writing a piece of music and recording the vocals, I go back through it and remove anything that doesn't work, using the above two rules. I find that almost always the track becomes much much easier to mix and opens itself out. Reverbs have more space, delays and effects can be heard and often small details, that were being drowned out, come to the surface.