For this video, the tips will be in no particular order - we’ll cover various concepts and explain when a tip would work well, but be sure to use your best judgment when introducing some of these ideas to your mastering chain.
Also, I’ll use this FF-Pro Q 3 for a good amount of the video, but a great free alternative is MEqualizer by Melda Audio.
This first idea is simple - in short, we don’t want to alter the frequency response of any one filter by more than 1dB. The reason is, we don’t want to create noticeable changes - most listeners hear an EQ change that’s greater than 3dB, but still, 1dB of change is more than enough to make most needed changes.
Another idea is that we don’t want to deviate too greatly from the mix engineer’s intentions - and if you were the mix engineer, it’s best to create a mix that more closely fits the frequency response you want, and not to rely on the mastering stage to make these changes.
Let’s listen to 2 EQ settings, one with small changes used to balance the spectrum, and one with changes greater than 1dB, and notice how the second one makes the track sound unbalanced and like it’s been equalized.
An emphasis technique means we use one type of processing to drive a group of frequencies into another processor - in turn making that second processor have a greater effect on those frequencies.
For example, I could use an EQ to drive the vocal’s clarity range, usually 2.5-5kHz into a saturator, causing additional saturation to occur on that range.
Then to balance out the spectrum, we could follow our second processor, again, in this case, the saturator, with a subtractive EQ that dips that same range.
This is a great option if you aren’t using a frequency-specific saturator, or the processor you want to affect a particular range more doesn’t offer frequency-specific processing.
Let’s take a listen to this example, and notice how the saturator is working harder on the affected frequency range.
Some engineers use linear phase EQ when mastering - so I want to briefly discuss some of the pros and cons. The main thing that linear phase EQ addresses are phase changes caused by minimum phase equalizers.
For example, if I want to use a high-pass filter on my lows, with an aggressive slope, I’d likely affect the phase around the cutoff - this often results in a frequency bump, that becomes more aggressive the higher the slope value.
But if I was to use a linear phase EQ, there would be little to no changes in the phase around the cutoff, and subsequently, no frequency bump or boost.
That said, this small boost to the lows may not sound bad - so we shouldn’t consider it something that’s always a problem.
Additionally, there are some cons to linear phase settings - since the filter type works by delaying the signal by a set amount, our DAWs need to compensate for this delay. When it does this, the original signal and the delayed one don’t align exactly, causing very mild destructive interference.
This effect is the most noticeable on the lows and quick transients.
All of this to say, you’ll always have to use your ears when making these decisions -may be the linear phase pre-ringing suits the track, or maybe the phase changes caused by a minimum phase filter suits the track.
Let’s listen to a track equalized with minimum phase EQ, and one with linear phase EQ, and let me know if you can hear the subtle differences.
When building a mastering chain, I almost always use EQ as the first insert - it’s a great tool for removing frequencies that I think will cause issues later on.
For example, say I find a small group of frequencies in the low mids that are making the track muddy - if I was to start with saturation, the harmonics I introduce would likely amplify this range, in turn exacerbating the problem they cause.
If I use subtractive EQ first, I can find the issue first and attenuate it, instead of having to fight it later.
I find that attenuating the lows from the side image up to about 80Hz, dipping a little of 250Hz, anything that feels disharmonious in the high mids, and then some of the sibilance is a good starting point, and from there I can adjust as needed.
Let’s listen to this particular EQ curve, and notice how it makes the track a little more balanced, and ready for some additive processing.
When we get to the end of the chain, a Pultec emulation works really well for shaping the sound - most engineers think of these not as EQs by as shapers, since their filters are gradual and affect wide frequency ranges.
If we look at the curves a Pultec creates by using this EQ curve analyzer, we’ll notice how large of a bandwidth the filters have.
If we put it at the end of the chain, before the limiting stage, we get the chance to do final tone shaping - for example, I could boost the lows for more kick and bass - a little around 3.5kHz for more vocal, and then some of the highs to add a little clarity.
I’ll use this Softtube emulation, but a great free alternative is RareSE by Analog Obsession.
Let’s take a listen to some important ranges being boosted, and notice how it augments the track.
For this next tip, I’m going to use this specialty plugin called SplitEQ which separates the signal into transients and tone - so the transient side would cover the Attack and Decay, while the tone is more of the Sustain and Release.
Unfortunately, I don’t know of a free alternative to this EQ, so if you do, please let me and others know about it in the comments.
A combination of a transient band filter over the kick’s fundamental, mixed with panning the band toward the mid-image creates a huge and driving 808 or kick.
By creating the band, we boost the kick’s level and the level of the transients - by panning it to the mid-image, we make the amplified band more centered.
Like before, I’ll only boost by about 1dB, then select the panning icon, the M/S function, and subtly drag the transient band toward the mid-image.
Let’s listen to it and notice how this type of filter works really well for a rap or EDM track.
As I was saying in chapter 4, some processors will be additive - this is especially true for saturation, in which added harmonics will emphasize particular frequency ranges.
With a multiband saturator, we get a lot of control over how this occurs - if we use a warm tube setting on the lows we often get a second-order harmonic - which will amplify somewhere between 80 to 200Hz, depending on what the fundamental is.
This same concept applies to any other frequency range - for example, if I use transformer saturation on the low mids, I’ll create both even and odd-ordered harmonics that will fill this range, in turn amplifying it.
That said, if I was to saturate one frequency range more than another, or one range and another not at all, it would be similar to boosting the affected frequency range or attenuating the unaffected range.
This effect is even more prevalent when we consider how some saturators introduce behind-the-scenes equalization - if we observe this Saturn 2 plugin, we’ll notice that each type causes a unique EQ curve.
Also, if you want to try this out for yourself, but don’t have Saturn 2, try this free plugin GSat+ which introduces various harmonic formations and distortion algorithms. If you combine it with emphasis EQ, as we discussed in chapter 2, you can greatly control how saturation shapes the frequency response of your master.
Let’s take a listen to Saturn 2 and notice how it both saturates and equalizes the master.
Air equalizers are great for adding brightness to a master - they amplify frequencies above 15kHz to varying degrees. For example, if we observe this Maag Air EQ, we can see that the 40kHz band, affects lower frequencies, resulting in audible changes to the signal.
Some air EQs also introduce harmonics to the high-frequency range, like an exciter - this is true of the Free Fresh Air plugin by Slate Digital. Since they affect highs, they might also amplify sibilance frequencies, which can result in a harsh sound.
For that reason, I’d recommend using a de-esser earlier in your chain if you plan to add some form of an Air EQ - this way you control harsh-sounding frequencies before amplifying the range they occupy. The Weiss De-esser is a transparent option, but T-De-Esser is a good free one if you don’t have one.
Also, if you’re working on Jazz or another genre that has cymbals, but you want it to sound clean, I’d avoid this Fresh Air plugin due to the intermodulation distortion the exciter could cause. Granted, it still may sound good so you’ll have to use your ears.
Let’s take a listen to Fresh Air being used, without and then with a de-esser enabled.
Most songs use 1 key - meaning they have 7 in-key notes or notes that work together, and 5 notes that don’t work together. Each one of these notes occupies specific frequencies which we can amplify or attenuate.
So to keep things simple, let’s say the song is in the key of C Major, we could amplify notes that are in key, which would be all of the whole notes, like A, B, C, and so on.
And then, we could attenuate all of the sharps or flats, like A#, C#, etc.
Although I don’t always use this tip, I have some presets saved to quickly monitor how it sounds with in-key frequencies amplified and out-of-key ones attenuated.
It doesn’t work every time, but when it does it makes the master sound a lot more musical by increasing the ratio of in-key signals to out-of-key signals.
Let’s take a quick listen to it tailored toward the key of our demo, and let me know in the comments if you think it improves the sound, makes it worse, or has no effect.
Any EQ that’s capable of Mid Side and/or Left Right EQ can be used to enhance the stereo image of your master. Again, I’ll use this FabFilter Pro Q 3, but if you’re using the free alternative MEqualizer, you can change the routing in the output section, then right-click any filter you create and change the stereo placement to match what I’m showing here.
When working with mid and side, some helpful filters include cutting the side image lows above the fundamental to center the lows, boosting the low-mid frequencies on the side to create a fuller and deeper sound, and boosting the highs on the side with a shelf to expand the highest frequencies.
If the vocal is buried, I could dip 2.5-5kHz on the side to reduce its masking effect on the vocals. Or if the vocals are sticking out too much I could subtly dip 2.5-5kHz on the mid-image.
If I want the kick to be more driving I could amplify the kicks fundamental on the mid image, somewhat similar to what we did in chapter 6.
If you aren’t a fan of mid-side processing, which some aren’t, you could try amplifying 2 bands, one on the left channel and one on the right channel, then offsetting their frequencies slightly. This will cause stereo expansion.
Let’s listen to a combination of mid, side, and left and right filters, and notice how we can greatly alter the stereo width in a frequency-specific way.