For this master, and for rock tracks in general, I want to get it to a loud final sound, without over-emphasizing the distortion in the track.
To start, I’ll use a Mid-Side EQ to first attenuate low frequencies from the side image, causing the mid image’s kick and bass to stay focused. Then I’ll dip a little around 200Hz on the mid image to reduce masking to the vocal.
On the side image, I’ll boost this same range to increase the warmth of the guitars and instrumentation**, before dipping roughly 2kHz on the side to carve out some room for the mid-image vocal.**
Since this effect will be subtle, let’s listen to the full before and after - I’ll use peak normalization to make their loudness more comparable, but keep in mind I couldn’t match their LUFS without causing the original mix to clip.
Although this mix sounds great, there are some aggressive resonances on the side image due to the heavily distorted guitars - I want to remove some of these so I’ll use this Soothe 2 plugin, and set the processing to the side image, before isolating it to the right range.
Notice that I set the quality to higher settings to reduce phase cancellation and other unwanted artifacts.
Let’s take a listen and notice how the sides sound a little more balanced.
Next, I wanted to make the kick stick out and have more power - to do this I used this Weiss compressor and isolated the compression to the mid-image and the lowest frequency range. With a low ratio and carefully set threshold, I only needed to attenuate by about .5dB.
Then I used the auto-make-up gain setting to amplify what was being attenuated. Let’s take a listen and notice that although it’s subtle, the kick has more of an impact.
Up next, I’m going to use this SplitEQ by Eventide and affect my tone and transients separately - for the transients, I amplified the lows and highs, as well as subtly amplified the full transient band at the output. I dipped a little at 250Hz. and boosted the vocal range with the tone.
As a result, the lows and highs are punchier, while the vocal comes through better.
Let’s take a listen.
After the SplitEQ, I’ll introduce this Omnipressor by Eventide to cause both compression and expansion - I started with a mastering preset and adjusted as needed. I set the amplification and attenuation ranges to 1.5dB. and set a slightly quicker release time of 80ms to work quickly but avoid distortion.
By compressing with the signal is over the threshold and expanding when it’s under, we control dynamics both from the peaks down and the noise floor up.
Let’s listen, and notice how the 2 processing types control dynamics but also create a more powerful-sounding mix.
With very subtle saturation combined with frequency-specific mid-side panning and input/output panning, we can create a lot of depth in an otherwise flat mix. The process to create this is complex, so I start with the 3D preset before adjusting to make the effect more suitable for mastering.
Notice that the drive amounts, mid-side panning, input and output levels, and panning are all triggered by envelope followers being triggered by the transients. Let’s listen and notice how flat the mix sounds when the processor is disabled.
With a bus, I’m going to create a parallel auxiliary track, and first use the plugin MSED to mute the mid image, in turn isolating the side image. Then, I’ll insert this transient shaper to expand my side image’s transients, resulting in a more detailed and dynamic stereo image.
With the track’s channel fader, I can blend in how much of this effect I want, depending on how wide I want the master, and how punchy I want the side image. Let’s take a listen to it.
Moving on to the master output, I like to use this Gullfoss EQ to subtly amplify and attenuate incoming frequencies for the sake of reducing masking and making the master louder and clearer as a result. I isolate the processing to the low-mids to highs and brighten slightly.
Let’s listen, and notice that although the changes are small, this processor has a big impact on the clarity of the mix.
When limiting, I like to use 2 separate processors in series to keep one from working too hard. First I used this Oxford limiter and used the enhance function set to about 30% to bring up quieter details - notice I only have less than 1dB of attenuation on the pre-process side.
Then I used this L2 limiter set to its dynamic algorithm to increase transient impact, but again, I needed very little attenuation from the peaks down.
Let’s listen to how these 2 limiters in series can make the mix a lot louder, without the need for aggressive peak reduction.
Last up, I wanted to add a small amount of loudness but more importantly, hard clip the transients - this will add a small amount of white noise whenever the transients cross the threshold. These added high frequencies will actually make the transients stick out, making it sound slightly brighter and more dynamic.
The final master was -8 LUFS, but feel free to master quieter if you want something more dynamic.
Let’s take one final listen to the full before and after.