This first tip is a little counter-intuitive - by definition clipping will attenuate transients; however, clipping adds high frequencies, somewhat similar to white noise. Since this noise is added whenever a transient hits a clipper’s ceiling, it’s almost like high frequencies, or white noise is added to each transient.
Granted, if you smash the signal into a clipper this won’t work too well, but, if it’s done carefully and isolated to transients you can actually improve the listener’s perception of transients.
Let’s take a listen.
Ideally, when mastering you’ll have access to a 96kHz. sampling rate file, but if you don’t it’s not the end of the world. Oversampling is typically better at avoiding aliasing distortion since it increases the signal’s sampling rate and introduces low-pass filters above 20kHz.
This means many of the harmonics generated from aliasing will often be filtered, or at least have their amplitude reduced significantly.
An audio example of this is a little difficult to demonstrate, but let’s heavily distort our mix’s high frequencies and AB with and without oversampling. This mix has a sampling rate of 44.1kHz, but you’ll still notice a significant reduction in aliasing distortion.
In my opinion, modulation is going to be a big part of audio processing in the future - FabFilter popularized linking modulation to plugin parameters, and UVI has expanded on that in their Shade EQ. One thing in particular that interests me a lot is the idea of randomization.
That said, I’ve linked randomization filters to 4 EQ bands to very subtly alter the frequency response of the overall mix.
With the filters I was also able to vary the left and right channels for each stereo band, meaning the stereo image modulates subtly as well.
Let me know in the comments if I’m imagining this, but to me, this slight variation gives the mix a lot of life and this somewhat unexplainable dynamic quality.
Like I said in the last chapter, Saturn 2 lets us modulate saturation - with an envelope follower that measures the transient I dynamically distorted each frequency range to varying degrees, as well as dynamically panned that distortion and the overall level to the mid or side image.
Furthermore, I modulated the dynamics of each band, so that each transient caused the range to be more dynamic before compression occurs on the back half of the waveform.
I could even increase the amount of the overall effect by linking the follower to the mix slider.
Let’s listen to how this saturation fills the sound as you’d expect, but includes a lot of complexity like dynamic stereo imaging, and more.
I like this effect on drums usually, but it works just as well on a full mix - in short, I’ll duplicate the main track and use MSED to solo the side image. Then I’ll use a transient expander to amplify the side image’s transients, which adds a lot of detail.
Lastly, I’ll blend the processed side image in with the original using the channel fader. Although most DAWs will compensate for latency, be sure to listen for any phase cancellation that might occur.
We often get in the habit of using one limiter for all of our projects, but it helps to line up a few different ones with similar amounts of attenuation to get a good idea of what’s needed. I typically just put these on my master output and cycle through.
When doing this, you might find that a limiter you don’t use too often or usually don’t like the sound of actually works best for the mix. Let’s listen to 3 different limiters and let me know which one sounded best to you.
Our stereo output can be separated into left and right channels, on which processing respective to that channel can be added. Although I usually like to process my stereo output collectively, I’ve found that using separate limiters on the left and right reduces unwanted artifacts.
The reason is, that each limiter won’t have to work as hard, letting you achieve a louder sound without aggressive attenuation. Let’s take a listen.
There are a few great upward processors out there, but I’ve found that each one works a little differently. That said, if you combine multiple types, like the Oxford inflator with the Weiss maximizer, you affect different aspects of the quieter parts of your signal and in varying ways.
Although I don’t think this works for every master, combining upward processors often achieves a complex and desirable full sound, and reduces the need for aggressive limiting later on.
Let’s take a listen.
De-essing is probably the least appreciated form of processing - but its role in mixing and mastering is crucial to creating a balanced sound. That said, definitely don’t avoid de-essing during mastering, especially if amplifying the signal brings up unpleasant sibilance that could easily be attenuated.
If using a de-esser during mastering seems strange, then use a multi-band compressor to attenuate sibilance, or a dynamic EQ - but definitely don’t make the mistake of trying to change your chain around sibilance, when a subtle de-esser would work well.
Let’s introduce de-essing to this track to illustrate that it still has a place in a mastering chain.
This last tip is a little weird, I know, but it’s helped me avoid mistakes in the past so I wanted to share it. One thing I always do is remove any processing that isn’t playing an active role in the chain - for example, metering plugins like a LUFS meter.
And I won’t just bypass these I’ll actually completely remove them - because although it probably isn’t affecting the sound, why not be 100% sure?
Or maybe if I’m doing a revision to a track I previously mastered, I won’t simply make the change and bounce it out, I’ll listen to the full track to ensure nothing inadvertently got changed.
Maybe this seems like overkill, but, true story I know one engineer that didn’t do this on a revision, and someone had accidentally sent him the instrumental instead of the full revised mix. He sent back the mastered instrumental, no one checked and that’s what got distributed.
So just always double-check and take any precaution you can when mastering to ensure that the final product is as good as it can possibly be.
There are 8 unique value-points inside the Sage Audio membership, so let’s go over each one so you can decide if it’s the right fit for you.
1. 50 Free Mastered Songs Per Year
2. 1-on-1 Mentorship
3. Unlimited Mixing Feedback
4. Sage Audio University™ (SAU) — (Mixing and Mastering Education Platform)
5. Mixing and Mastering Competitions
6. Community
7. 70+ GB of Curated Sage Audio Downloadable Content, Including:
8. 700+ Video Catalog
