Arrangement: The Secret Weapon of a Professional Mix (Part 3)

Fix Undefined Mixes with Better Arrangement & Reference Tracks

You can do some pretty amazing things with a song at the mix stage BUT honestly, a great arrangement makes or breaks a song and a mix.

I’ve written a number of times about this before so I won’t rehash those points. What I will say is that getting a great arrangement means serving the song. If you’re trying to squeeze that guitar/synth/vocal part you really love into the mix but it makes the mix fall apart and starts to take away from the song, get rid of it or something else. 

Subtractive Mixing (Cutting Out The Fluff)

Being brutal with parts is essential for mixing. Why?

Auditory Masking: The Battle of Frequencies

It’s always interesting when I’ve mixed songs for a client and they’ve told me that they think I’ve added something to their mix. What’s happened is that their own rough mix had a bunch of instruments and voices that were covering up each other. I’ve pruned out the overlapping frequencies with EQ as well as balanced each part, suddenly it sounds like there are new sounds in the mix.

What’s happened is I’ve balanced the mix out and removed auditory masking using EQ. The auditory masking is really noticeable when you simply dip -1.5 dB around 250Hz on a mix. For most bedroom mixes, it immediately ads clarity in the midrange because you’ve reduced masking by partials (upper harmonics of a sound) in this frequency range by lowering the fundamental at 250Hz slightly.

Let’s look a little closer at auditory masking. There are three types of masking:

1. Amplitude (volume)
2. Temporal (timing)
3. Directional (panning)

Right now, we’re just going the focus on amplitude masking.

Amplitude Masking: is where a loud sound masks a soft sound especially when they occupy the same frequencies. This is where using EQ and balancing your mix can help reduce this type of masking BUT the bottom line is, if you can remove the part altogether … yay! More frequency bandwidth for crucial sounds that make the mix sound amazing.

The Tiny Music and the Playlist Test

On a practical level you can determine the necessity of a part by turning the volume right down on your mix so you can just here distinct parts. Now if you suspect you could lose something, try muting and seeing if you notice it’s missing. If you don’t it might you might be able to remove it completely.

Repeat this process until you remove all the unneeded parts. Then try bouncing a mix with those parts and without. Put these mixes into a playlist of 2 – 3 other mixes and hit shuffle. While you’re doing something else outside the studio and not critically listening to the mix, put the playlist on and take a listen to the playlist and wait for both tracks to play. At the end of second time listening to the track, go and see which version you’re listening too and work out which mix sounded best.

Obviously, you don’t need to go to these lengths if you can confidently say – yes, this part can go.

Don’t Add, Layer To Fix

Once you’ve brutally removed any fluff from your mix if any instrument/vocal is feeling like it’s not quite right – don’t add anything. If the sound is just weak or the performance is lacking and needs to be recorded again, do that instead.

If the sound, part and performance is solid but it’s still not strong enough you might need to layer the part by doubling it. What this does is reinforce the part by adding more harmonic information (not harmonies) to the part thereby thickening up the sound.

Important Note: Don’t just copy and paste the same recorded section in your DAW – this isn’t going to add harmonic information, it’s just going to double the volume. You want to micro inconsistencies of a brand new recording or sound. You can get away with this when laying synths BUT if you make some micro adjustments to the new synth sound in timing (and sometimes pitch) you’ll get some added texture. Experiment with this.

Go High, Go Low

Sometimes you need to double but think horizontally. If a part doesn’t have clarity, try doubling up the octave. If the part feels thin, go low and double it down an octave. What you’re doing is adding harmonics to the top an bottom of the fundamental sound that no amount of EQ or saturation will achieve.

Here are some layering techniques I have used to help improve the arrangement of a track to improve the mix:

Drums: Add samples to help firm up poorly recorded or weak sounding drums. Kick and snare are the usual culprits. Blend judiciously and pay attention to how the sample sits with the room mics and overhead mics. You might need to adjust the pitch up or down and/or phase of the sample to help it blend.

Melodies: Double the part either in the same range or an octave above for below. For example, if the chorus of a song is in the upper range of a singer, record 2 tight doubles of the part and then (if possible) have them double the part an octave down. You pan the doubles left and right and blend them in as much as you want and then tuck the octave double in underneath. It just needs to feel fuller and not actually be heard. This is why getting super tight doubles is important.

Harmony/Chords: This can be approached similar to melodies although you need to be careful not to create a wall of frequency. For example the “high strung” guitar technique which originated in Nashville. Or layering a synth part with a brighter sound an octave higher. Sometimes you might need to remove some notes (especially 3rds) if it gets too thick.

Bass On Tiny Speakers: The reason bass gets lost on tiny speakers is that the fundamental is usually well below the lowest frequency a small speaker can produce. To be able to hear bass on smaller speakers you need to enhance harmonic information above the fundamental. If you’re recording synth bass it can help to double the bass part up and octave, sometimes with a pointier sound and then carefully blend that in.

The Mix That Sounds Great Unmixed

The best test of an arrangement is the mix balance test described in the first post of this series. If you can push up the faders of the tracks of your mix raw, balance them and get a great sounding mix, you’re 75% of the way there.

The decisions you make in the arrangement process are crucial so it’s worth your time getting the arrangement right before trying to mix. Use the mix balance process as your guide as you go. As you add new elements, pull all the faders down and build that balance mix again, it should only take 5 – 10 seconds per track which means the process should take no longer than 2 – 4 minutes.

Work intuitively and quickly. Use the 3dB increment method to help make adjustments once the faders are up. If 3dB is too much, split the difference.

After practicing this quick mix balance approach you’ll learn to love how it helps cleanse your mix palate and stops you getting into the weeds and feeling lost about how to fix the mix.

In the next post we’ll talk about plugin presets and why they are generally useless if you’re learning mixing.