03) Simple Tips for Mastering
I am a minimalist at heart. I want processes that are simple, direct, and powerful. As engineers, we tend overcomplicate things by inventing novel solutions to problems our ears may have exaggerated.
We try to create scientific solutions to the problems we hear, but sometimes those problems are not issues at all. Or our solutions create unintended consequences we might not notice in the excitement of our solution. We have to remember we are creating art, we are not doing scientific experiments. It is not our job to polish every problem away. That’s why I have come up with the ethos of “Don’t Do Too Much”. Because a lot of times doing too much is what makes music sound overcooked.
First Step - Start Clean
My first step in mastering is to push the record up to a loudness that is appropriate for the genre with Gold Clip and my favorite final limiter. Some people are anti-loudness, but I am not. I believe some genres benefit from loud masters. They create a sense of cohesion and density that is sometimes necessary for the production.
I'll start with the default Gold Clip preset and adjust the input level so peaks go to 0.0dBFS on the input meter (essentially no clipping), and then add 1dB to 2dB of Modern Gold. I then insert my favorite final limiter after Gold Clip and get the rest of my loudness. 2dB of Modern Gold means the final limiter will do 2dB of less gain reduction to achieve the loudness you desire.
When I want my master extra clean, I may turn the clipper off (command clicking the clipper) and only use the Modern Gold processing. However, turning the clipper off comes with some risks. Read more about those risks here.
It's rare, but for some songs, Gold Clip and a final limiter may be enough. However, this simple and loud version of the mix gives me the sound that I call "digital urgency."
Digital Urgency
Digital urgency is the pure, in-your-face version of the song. It is also most likely how the producer & engineer heard the song when they said, "This record is done. I usually get my master to the exact same loudness that I hear in their reference mix. By the way, I always ask for the pre-master and the reference mix so I can hear what was approved by the team behind the song."
Now that I have that loud, clean, and true-to-the-artist's-intention version of the song, the limiter may start to tell me about what is wrong with the mix. Most specifically, if the low end is pushed too hard, I'll hear the limiter's gain reduction bite down on the bass and kick hits and modulate the mid-range in a weird and ugly way.
If the mix is bitter or harsh sounding, that will come forward too. If the mix sounds boxy, then that boxy tone will be pulled forward even more. I then have to come up with solutions to fix those problems.
I tend to fix these problems with the cleanest solutions first and then move to using color tools. If the low end is buckling down into the limiter, it may need some low-frequency shelving or filtering with a clean EQ. I then ask myself, did that adjustment make the low-end interact with the limiter in a more natural way? Did I go too far and lose the power of the mix?
When To Compress
If things feel too loose or flappy in the mid-range, I may need some compression to give the song cohesion or bounce.
For mastering compression, I tend to filter out anything below 150Hz into the key input of the compressor. That way, the compressor is responding to and creating clarity in the mid-range and not bouncing around with the kick and bass.
Many modern mixes are so hyper compressed that compression is not always needed in mastering. Be judicious and consider how compressed the mix might already be. Also, listen to how the compressor interacts with the limiter. I often use a compressor to increase the attack of the kick as it hits the final limiter. Be intentful with the compressor. If you do not have an intended dynamic outcome in mind then most likely compression is not needed at all. If I do use compression, I always filter out the lows and tend to use slower attack times of 30-50 ms. That lets some transients through and increases punch.
Finding Balance
Clarity and power in the mid-range is the key to perceived loudness on streaming services. Sustained resonances in the mid-range will raise your average LUFS and reduce your perceived volume on streaming. Finding balance in this range is the where I spend a lot of time when mastering.
Does the mix feel tight, balanced and powerful in the midrange?
Does the mix now feel tight and balanced in the midrange?
If the mix is too brittle, pointy or harsh in the tops, maybe high-frequency shelving and some soft-knee de-essing will lessen that high-frequency fatigue.
Sometimes records have what I call "instrument-frequency discontinuity". What is that? It's when one instrument is particularly resonant in a frequency range and is poking out harshly in a mix. Maybe a tambourine is pointy around 12 kHz and mixed too loud. The easiest solution is to ask for a new mix, but sometimes that is not possible. In those situations, a tight EQ reduction to that frequency range is my first move. If the discontinuity is in the mid or high frequencies, I'll use a linear phase EQ. If it is below 2 kHz, I'll most likely use a natural phase EQ. If that EQ move is not enough to fix it, then I might move to a dynamic notch filter with Weiss de-esser or Kirchoff EQ.
Is the top end bright, but gentle enough to not fatigue the ear now?
These are all clean solutions to getting a balanced “true to mix” version of the record. I usually start with these or similar concepts first, and then move to more colorful approaches.
Adding Color (or not)
Once the loudness, balance, and dynamics are in place, I then start to think about how far I can push things, what I can experiment with, and what is appropriate for the genre.
We already made the decision about the loudness for the genre, but we have to ask ourselves what best serves the song.
Is this song best served by digital urgency and clarity, meaning no tubes, tape, or transformers, just that clean in-your-face sound?
Believe it or not, some productions sound much better that way, but others may need some extra flavor.
Some records may need some saturation, clipping, tape emulation, transformers etc. All of these devices start to add color, but move the record away from that digital urgency I mentioned earlier.
Know Your Role
Your job as an engineer is to decide what's appropriate. Is this going to be clean and pure like a Doja Cat or James Blake record, or is it going to be soft, rounded, and sauced up with color like a Tame Impala or Alabama Shakes record? Or maybe it lives somewhere in the middle?
Those decisions are yours, but I generally tend to start clean and make it sound as whole as possible, then add color as I see fit for the genre and the issues that the production presents.
In mastering, I always tell myself "don't do too much." In mixing, I am more often smashing things around more aggressively.
However, I usually solve problems with clean processors first and then move toward the colorful and crunchy processors after that. But keep in mind these approaches are interactive, and I often find myself circling back to my clean processors as the colorful ones make their mark.
Give it a try and let me know how it works for you. We all have our own solutions to the sonic issues we hear in our music, but these are the approaches that worked best for me and I hope they are meaningful for you.
Be well,
Ryan Schwabe
Grammy-nominated and multi-platinum mixing & mastering engineer
Founder of Schwabe Digital
