For this video, we’ll cover 9 back-to-back steps for mastering your track, and then at the end cover a very unique and more advanced way to master, so stick around for that if you’re curious.
Before we start adding processing, let's make sure we have enough headroom - if the track is less than 3dB from unity, it’s best to reduce the overall gain of the track with clip gain. Similarly, if the track is too dynamic for its genre, it may need to be remixed.
Otherwise heavier compression or limiting will be needed to bring the level up to the desired amount. Let’s take a listen to the mix without any processing.
Next, let’s start our processing on the channel strip, not the master output for reasons that’ll become clear later on.
We’ll want to attenuate aspects of the signal that we don’t want to amplify and do so in a precise manner. The Kirchoff EQ is a great option for this that offers an extended demo mode - Big thanks to our viewer Chris for bringing this plugin to my attention.
With it, I’ll listen carefully to the mix and subtly attenuate any aspects that I don’t enjoy.
We can also open the piano roll, and see which notes would be out of key with the track and subtly attenuate those.
Let’s take a listen and keep in mind that the effect may be subtle.
Saturation causes soft-knee compression at higher levels, but more importantly introduces harmonics to fill the spectrum, and even emphasizes or tames transients. For example, Tube saturation will amplify transients and create a warm sounding second-order harmonic - while tape tames transients, and creates a 3rd order harmonic.
What you choose is up to you, but I typically emphasize my transients with tube saturation. Saturn 2 lets you use envelope followers to make the distortion in time with the transient, so I’ll include this as well to the highs to further emphasize transients.
Something I’ve really grown to enjoy is the sound of sending my signal to a bus, then isolating the mids with a linear phase EQ - after which I introduce an upward compressor. This means that I can bring the details of the mids forward significantly, then blend in the effect.
The EQ needs to be linear phase to avoid weird artifacts between the 2 signals. Alternatively, you could simply put the upward compressor on the main track, but for reasons I’ll explain next chapter, this setup gives us a lot of control.
Let’s take a listen.
Since we have our mids isolated with their detail brought forward, we can further process these in any way we see fit. For example, I could use a dynamic EQ to expand certain mid frequencies, or use parallel downward compression to make the signal even denser.
For the time being, I’ll do a little of both to really shape these parallel mids, and give myself a lot of control over both the master’s timbre and dynamics.
Let’s take a listen and notice how this addition processing augments the mids.
For the remaining processing let’s move to the stereo output - this way all of our processing is being added in order, starting with the channel strip’s EQ, through the parallel bus, then to whatever we add to our output.
If you’re making dance music or anything in which loudness is expected, a clipper will definitely help bring your close to your desired LUFS - that said, streaming services will still normalize the volume so keep this in mind. When I use a clipper, I’ll put it before the limiter.
This way added harmonics from the processing won’t cause clipping in my actual DAW, which I don’t have the ability to control the shape of.
Let’s listen to clipping be added, but I may take this off after this demonstration.
I’m going to use the Kirchoff EQ again and try some of its useful analog emulations to create some gentle curves. Using the Left Ride fader, I can determine to what extent this band affects the left or right channel, in turn giving me control over my stereo image.
Around 3 to 5kHz I’ll create 2 of these bands at slightly different frequencies and lean one more toward the left and one more toward the right channel.
I’m then going to create a high shelf filter and choose the mid-side placement, before leaning the filter more toward the side image to cause high-frequency stereo expansion.
Additionally, I’ll cut the lows out from the side image using a high pass - this way lows remain more mono and focused. Lastly, I’ll introduce low latency linear phase to avoid phase issues in my low frequencies, and enable 2x oversampling to avoid aliasing from my air filter.
Let’s listen to what this EQ adds to the signal.
Next, I’m going to add limiting to my signal, but avoid true peak detection since it’ll negatively impact the transients. The limiter you choose is up to you of course, by the Sonnox Oxford limiter lets me increase low-level detail, in turn reducing the need for affecting my peaks.
As a result, I get a loud and full sound without the need to completely destroy or alter the peaks.
Let’s take a listen.
Last up in this chain, I’m going to go back to my earlier processors and consider where certain functions can be automated for a unique effect. For example, I could increase the level of my parallel mid signal during an important part to make it sound fuller.
Or I could reduce it for a softer passage of the song. What you alter will completely depend on what you’re trying to accomplish and the song itself.
Let’s listen to some of the automated changes made.
DDMF’s Metaplugin is a truly interesting one that lets you create complex routing - so I’m going to use it for exactly that and create a somewhat orthodox way to master this track. First I’ll use 2 instances MSED by Voxengo to solo the Mid and Side respectively.
Now I can process the Mid and Side channels individually, and recombine them when needed.
Next, I’ll create a crossover and route my Mid image to it - this breaks the signal into lows, low mids, high mids, and highs, which I can then route separately.
I’ll route my lows to the oxford inflator, low mids and high mids to Saturn 2, and my highs to the satin tape plugin - giving me a lot of control over how the signal is saturated.
I’ll route the Side image to the Oxford inflator and avoid breaking up the signal into distinct frequencies as I did with the mid image.
Then I’ll equalize my mid and side images separately, and limit them separately. What’s cool about this is that whenever the mid image is attenuated by the limiter, the sides will be louder in comparison, causing program-dependent stereo expansion.
Lastly, I’ll route them to the output. Be careful when using this plugin since there are definitely some quirks and phase cancellation can definitely be an issue, especially when introducing oversampling for some reason, so I’d recommend leaving this off.
Also, you can even automate these parameters in your DAW, as we did in the last chapter, but I’ll leave this be for the time being.
Let’s take a listen and consider how you could use this plugin for a unique mastering session.