Professional Analog Mastering.
Professional Analog Mastering.
226K+ subscribers
Stereo expanders are often used to widen a track or bus, but this method sounds much more natural and retains a driving center image.
In the past, I’d use a M/S compressor and compress the mid-image separate from the side. When done subtly, it has a nice dynamic stereo expansion effect.
That said, I recommend trying the opposite instead.
Using a M/S compressor, de-link the mid and side channels, allowing for independent processing.
Then, on the side image, lower the threshold, likely a fair amount, until you achieve about 1-2dB of attenuation on the mix bus, or more as needed on an instrument bus.
This should narrow or center the stereo image. However, if you use makeup gain on the compressed side image, you’ll achieve a dense and impressive side image.
It’s an aspect of a mix that rarely gets compressed, but the sound is both unique and often what’s missing from a mix.
In short, by compressing and then amplifying the side image, we reduce its dynamic range and amplify all of the quiet parts of the signal - resulting in a good amount of detail brought to the forefront.
Let’s listen to the effect used on a drum bus in the context of a mix, and then listen to it used subtly on a full mix bus.
Watch the video to learn more >
After experimenting with different saturators and super complex techniques for about a decade, I’ve found this simple method. It works almost every time since its effect is so musical.
So, let’s say I have a drum bus, bass bus, guitar bus, vocal bus, and so on.
On each I’ll insert a frequency-specific saturator - I usually use Saturn 2 for this.
For each bus, I’ll find the range occupied by the fundamental or fundamentals of the notes performed.
With that range isolated, I’ll saturate moderately and use different saturation models for instruments in a similar range.
For example, if I use this warm tube setting on the bass’s fundamental notes, I’ll achieve even and odd harmonics that will be in-key with the track due to being direct overtones of the fundamental.
I could do the same for the drums and isolate the kick and snare fundamentals, but use a tape or transformer setting to ensure the harmonics aren’t overlapping too heavily with those generated from the bass. With this method, the saturated elements fit together, with the harmonics generated from each bus occupying slightly different spaces.
The same could be done for guitar and vocals - I’ll isolate the ranges and saturate while using different algorithms to impart varying harmonic formations.
The overlap of harmonics is reduced, meanwhile, the musical aspects are emphasized by making it much more likely that the generated overtones come from in-key elements.
If you’re using a different saturator and want to understand how it forms harmonics, run a sine wave through it and monitor the output with an eq or analyzer.
Then, try to memorize which algorithm generates which harmonic formation. Of course, use your ears to confirm what sounds the best as you experiment with various options.
Let’s take a listen to this method being introduced, and notice how full the mix becomes, without becoming unbalanced.
Watch the video to learn more >
When EQ matching first came out, I was excited to match the frequency response of my mix to one of my favorite mixes, but I noticed it rarely if ever, works well - here’s why.
Let’s assume we’ve matched the amplitude of our mix with our reference mix.
First off, we’d need to ensure that our mix is in the same key as the reference mix.
If not, EQ matching will likely attenuate a lot of in-key frequencies while boosting out of key frequencies.
Although an analyzer might make it look like frequencies are ubiquitous, or as if they’re evenly distributed across the spectrum, they absolutely aren’t.
Each performed note has a fundamental and overtones that relate to the fundamental. Even a complex arrangement adheres to its key - meaning it occupies very specific frequencies, especially in the lows to mids.
In this instance, frequency matching will absolutely mess up your mix.
But let’s say they’re in the same key - still, they won’t have the same notes played the same number of times.
As the analyzer measures the amplitude of the overall spectrum during the matching process, it will measure the amplitude of the notes played most often to be higher in amplitude.
So, for example, if your song contains notes equally, meaning each note in the key is played an equal number of times, but the song you’re matching to plays the note A 4 times more often than the others, then you have a problem.
One of your notes will now have a higher amplitude than the others and for no good reason. It won’t improve your mix, just emphasize a particular note.
There are dozens of other specific examples I could give, but you get the idea.
It’s not like you’re balancing your mix by EQ matching to another balanced mix. Instead, you’re reshaping your mix to one that’s been balanced for its specific contents - which, unless you’ve completely copied this other song, won’t be the same contents of your song.
To show this, I’ll find a song that’s as close as possible to the mix I’m working on.
After matching, notice how nothing is improved - instead, seemingly random aspects are accentuated and attenuated.
Watch the video to learn more >
Most of these are well known, but in case it’s useful, avoid:
Aggressive compression or limiting on the mix bus. Save this type of processing for mastering.
Avoid any equalization that causes more than 2dB of change on the mix bus. If a range is lacking or too aggressive, it’s best to address it on a track-by-track basis.
Avoid HP filters greater than 12dB/octave on an instrument bus or the mix bus. The phase rotation around the filter will emphasize frequencies above the band as well as interfere with transients.
And of course, ensure that no peaks exceed 0dB by the time the signal hits the stereo output. It’s alright to go over on individual tracks or instrument buses since DAWs use 64-bit processing, but by the output, it needs to be below 0dBTP.