The Mage’s Workbench – Episode 1: Dynamic Mixing

- Posted January 21st, 2020 by

Welcome to The Mage’s Workbench, a mini-series which tackles various audio production topics spanning from trackers to DAWs (and everything in between). While the material may not always apply directly to you, the goal of each episode is to present production techniques in a broad, universally applicable fashion. In short, there should be something here for nearly everyone that works with audio. For the inaugural installation, we’re going to take a look at how you can use automation to craft a polished mix that evolves throughout the course of your song.

Episode 1: Dynamic Mixing

Not every mix will leverage automation to the same extent, but it’s safe to say that automation has the capacity to elevate your song to new heights. Some like to work with automation earlier on in the mixing stage and others will opt to do this well after the essential components have been locked in. Bearing in mind that neither approach is wrong, we’re going to focus on the latter school of thought. A helpful place for us to start is to define the difference between a dynamic mix and a static mix:

As its name likely implies, a static mix is the non-kinetic iteration of your song’s mix. It is a representation of how you intend each voice to interact with one another in a global sense. For example, if your lead voice is the primary focus of your song (which would make sense), your static mix will likely prioritize that voice throughout the majority of your song. Similarly, your song’s supporting voices will maintain a predominantly auxiliary relationship with the lead voice. In a static mix, these relationships do not vary significantly beyond the dynamics inherent to the source audio. In the case of a dynamic mix, however, these relationships are not nearly as rigid.

“… bring all channel faders to unity gain (often represented as 0 dB)”

For our purposes, a dynamic mix relies on having an existing static mix. Generally speaking, you should already know where your guitar is going to sit. You should know how distant the drum kit needs to sound. You should know how present the vocals need to feel. A dynamic mix acknowledges all of these decisions but respectfully asks: ‘Okay, but what about this verse? What about this phrase? How about this riff over here?’ While static mixes do a great job of establishing the fundamentals of your mix, they can’t always provide a universal answer to these questions. Using automation, however, you can craft a more responsive sound that is tailored specifically to the needs of your song.

With functional definitions in place for both static and dynamic mixes, let’s tackle the former. If you’re relatively happy with your current mix but it’s relying on channel faders to determine output levels, solo each and take note of their average level. One-by-one, bring all channel faders to unity gain (often represented as 0 dB) and then use a gain trim plugin to set each channel to the appropriate level. This might seem like extra work, but hang tight. Your toils will not be in vain.

Finalized static mix with minimal processing and no automation

Next, we need a reverb send. Typically, this will manifest as one reverb plugin in a single channel bus that you pipe other channels through in varying amounts. How you configure your reverb settings will largely depend on your song’s needs, but it’s generally not a bad idea to do the following:

Remove dry signalPrevent stacking with source
Return to group bus or masterMaintain control of routing
Roll off channel’s low frequenciesKeep your low-end free of ‘mud’

Now route each track that needs reverb to the reverb bus and adjust the send amount to taste. Some channels may need as little as 5% whereas others may be able to get away with something like 75%. Since your reverb lives happily in its own isolated channel, you can also adjust its overall volume as you see fit. Just be careful not to overdo it—reverb has a way of commandeering a mix if left to its own devices!

It’s time to take stock of all the active voices in your mix. Create a list that places all your channels into two distinct categories: the elements that will remain constant and the ones you anticipate changing. For everything in the latter bucket, assign the following insert FX (in this order):

• Equalizer*
• Gain trim

* There may be an additional gain trim before this insert FX from when you configure the static mix
† It’s possible that your equalizer also has a gain trim knob. If this is the case, you can consolidate these two FX into a single plugin.

“How you configure your reverb settings will largely depend on your song’s needs”

“… route each track that needs reverb to the reverb bus and adjust the send amount to taste”

Automation clips controlling volume, high frequency content, and reverb content

The goal of this signal chain is to give us some means of controlling both the volume and the high frequency content. As you have likely concluded, we’re going to create automation clips that control both of these parameters. In addition, we’re going to attach an automation clip to the reverb send. If you’re working with a stereo signal, create one more automation clip tied to stereo separation. This parameter may be baked into your software’s mixer or it may require an additional plugin.

The question to ask now is: ‘what takes priority?’ Each section should have a primary voice and—to a lesser extent—a series of supporting voices. To illustrate, let’s use a vocal-heavy song as an example. In this type of song, vocals are often prominently featured in the mix. However, it’s perfectly conceivable that there will be moments without vocals. So what happens when the vocal track drops out? Do we just leave a gaping, vocal-shaped hole in our mix? No way! We want to fill it with the next most important voice. Depending on the song, this could mean bringing up the guitar or introducing a new synth. Suffice to say, it’s situations like this where establishing an order of operations really comes in handy.

So what does all this add up to? The answer is the ‘secret sauce’ of a dynamic mix: the illusion of distance. Let’s say we want to move a voice further back in the mix. We don’t just want to make it quieter. We want it to sound as though it is further away. Conversely, if the sound source needs to move up in the mix, it needs to feel like it’s closer to us. All of these components play a role in our perception of space and distance. And as it turns out, we can make use of those newly-created automation clips to directly influence our listener’s perception.

In terms of how to manipulate your automation clips, the ‘formula’ (which is also listed below) is fairly simple: if the source needs to move forward, the volume, high frequency content, and stereo separation (if applicable) increase while the reverb content decreases. If the source needs to move backward, the volume, high frequency content, and stereo separation (if applicable) decrease while the reverb content increases. The degree to which you make any of these adjustments will vary from section to section (and from song to song), but we can use a value of 50% to represent the base level for each parameter in your static mix. In other words, if we leave all your automation clips at a level of 50%, the mix will sound identical to your static mix. The one exception to this will be reverb automation level since it’s not likely that all channels will be sending 50% of their signal through the reverb bus.

High Frequency ContentIncreaseDecrease
Reverb ContentDecreaseIncrease
Stereo SeparationIncreaseDecrease

Comparison of isolated pulse wave without automation (first 16 bars) and with automation (last 16 bars)

Comparison of pulse wave in global mix without automation (first 16 bars) and with automation (last 16 bars)

After spending time dialing in your dynamic mix, you may find a few imbalances that surface in the process. Pulling a certain channel forward in various sections may serve their self-contained purposes, but it’s possible that channel may now feel a little disjointed with the rest of the mix. Since the initial static mix was configured in a way that left all channel faders at 0 dB, we now have the freedom to yank those knobs and faders around and make some big (or little) changes to the overall mix. Having this ‘second static mix’ allows you to make proportionate changes to your dynamic mix. Now any time you bring a channel fader up or down, the automation will scale in tandem. This last stage of mixing is where you can really dial in all your FX and finalize all the fine details. You may need to tighten up the automation as you go, but you should now be sitting on a fine-tuned mix that grows and adapts while maintaining consistency throughout.

Comparison of unprocessed static mix and processed dynamic mix, alternating every 16 bars

Thanks for reading this episode of The Mage’s Workbench! Got your own process or workflow that you’re excited to share? Have a question about the material in this article? Drop a comment below or chime in on Discord! Don’t stop sharing and don’t stop creating!

Note: traducción al Español por Pixel Guy encontrado aquí.

Dig this article? Then consider supporting us on Patreon!