A simple approach to mixing a rock song: Part 2

By mic on 6/15/2007

The purpose of this and the next couple of post is to discuss the volume, pan, and equalization of the tracks in a mix. These are related – as you adjust one you may want to consider changing the other. For now though we will discuss them separately.

Each instrument in a song is important. Each instrument carries the song and so during mixing we should ensure that all instruments are audible. It is unfortunate to end up with a song in which the bass is missing, the snare is barely audible, or in which the vocals are so loud and muddy that the rest of the instruments cannot be heard. Many people mix songs with the goal of making instruments work together. Some people (like me) try to make sure that instruments do not hinder each other. Fortunately, it is easy to balance or space out the instruments of a song. To do so, we adjust the volume and the pan of instruments, or equalize them. In other words, we arrange instruments in the dynamic, spatial, or frequency spectrum of the song.


In practice the "right" volume of a track depends on the volume other tracks and may change as you pan, equalize, compress, or add reverb to tracks. Independently of how this all plays out, my plan is always simple: start with the rhythm section of the song and fit in everything else later. Specifically, I work with the kick and snare to properly set up the rhythm of the song. I may have recorded my drums on one, two, or many more tracks (see Our cheap home recording setup – the drums). One way or another, I adjust the volume of the snare so that it becomes annoyingly loud without clipping.

What does this mean? This means that I set my snare as a reference point for my mix and I can now bring in everything else. Most digital equipment will set the maximum allowed output signal level at 0 dBVU and this is exactly where I try to set my snare. That is my reference point. Now, it is unlikely that you will know the exact scale of the rest of your output meter numbers, but let us assume, that whatever is at the bottom, it is the minimum signal which creates any useful noise. When I use CoolEdit Pro this floor is at -80 dBVU. I use CoolEdit often, and so what is below describes my CoolEdit experience, but it is hard for me to imagine that other home recording software differs that much.

I have always said that the snare should make me blink. When I get to this point I may take the snare a dB or so down to the point where it almost makes me blink. If I recorded the bottom and the top of the snare separately I would adjust the volume of the two so that they mix well and, in general, I would work with both tracks together. There is no right choice on how they should sound and that is a choice that you have to make on your own, but I generally end up mixing them at the same output level.

Having set my snare, I would bring in the kick to make it very audible. I actually mute all other tracks to listen to the mix between the snare and the kick. My kick, if well recorded, does not drop by more than -3 dB from the snare. I would then add the bass to get a good rhythm section. Generally, my bass ends up about -10 dB under the snare on the output level meter. And I would add the rest of the drums. My hihats are usually at the level of the bass, so highest and lowest frequencies of the mix carry the same weight. The crash should be as audible as the hihats. Toms should be at least at the level of the kick.

After setting up the basic rhythm section I have no set rules about mixing the rest of the instruments in. Usually the vocals, rhythm guitars, and solos end up at the same level, while clean rhythm guitars are lower, but all that depends greatly on the nature of the instrument and the song. I do, however, have some simple rules about what not to do:

  1. I do not put solo instruments above the mix (solo guitars, vocals). The biggest mistake is to assume that the solo makes the song. When I find a comfortable place for the vocals or the solo guitar I force myself after that to bring those down by a dB or two. There is no reason for the vocals to come up in front of the bass, for example;
  2. The opposite rule to (1) above is: I do not leave hihats and bass under the mix. You do not want to end up with a mid-range song. Hihats and crashes carry the liveliness of a rock song and should be plainly audible. Without the bass the contemporary song would sound hollow;
  3. I do not assume that volume levels have to change during different parts of the song. Just because vocals are coming in does not mean that the rhythm should go down. I do change things sometimes, but not too often; and
  4. I do not sacrifice volume for clipping. My snare is set up at 0 dBVU, which means that it will occasionally clip. If I do not find this annoying I would not change it.

You will sometimes find bands, who are good instrumentalists, and who sound great on an album but not too well at concert. Nowadays, more often than not, you will find bands that are great at concert, but do not sound very exciting on their album, and the reason for that is usually the fact that liveliness was lost when the album was produced. The easiest way to lose the liveliness of a song is to ignore the ends of the frequency spectrum (highs and lows; hihats and bass) and to over-exaggerate mid-range instruments (vocals, overdriven guitars, keyboards). A very important reason to pay attention to that is the increasing amount of digital mp3 distribution. The mp3 compression of your song by design will probably depress your highs and lows and take out some of the excitement of your song.

For the sake of having an example, here are the output meter levels for one of the songs I mixed in CoolEdit Pro. I picked this song, as it is a mellow and more accessible rock song, which includes drums, bass, clean rhythm, overdriven riff, solo guitar, piano, and vocals. I put two values for the piano as I did actually raise its output during the piano solo.

Snare (top and bottom together)0dBVU
Rhythm guitar-17dB
Riff guitar-12dB
Solo guitar-15dB
Piano-13dB, -18dB

This is probably enough for the output level mix of the song. I will discuss pan and equalization in my next posts. Аs you change the pan or EQ of a track you may want to go back and review its volume in the mix. I do revise the volume mix often as I go along.

authors: mic


Copyright 2006 by Kaliopa Publishing, LLC