May 8

May 8 Thee Process is thee Product (A Linear View of Making Music)

Any artist should stay challenged as long as possible.
-- Genesis P-Orridge

I love thinking about the relationship between a work and it's artist. Even the most staunch materialist must admit there is a mountain of time and effort that you put into a work before you release it into the wild wide world. It's that ineffable relationship between the work and the artist is fascinating. That sense when everything falls away and the work is moving through you. There's something special about that.

When I first started makin-a-the-techno (get offa my lawn you whippersnappers) the idea that a single person could conceive, write, produce, record, mix, master and distribute a song was insane. First you had to get yourself an instrument, then a band, then play a few shows at a friends party, then maybe a bar or two, then if you got really lucky you'd get noticed. Then there would be a legion of people to produce, record the song in a multi million dollar studio. Nevermind the logistical nightmare that is distribution. Now it is not only conceivable, but practically standard for a single person to do all of these things on a single computer. The desktop publishing punk rock revolution has come.

Each step on that ladder of conception, writing, producing, recording, mixing, mastering and distributing a song is a skill, process and art. Some I have mastered, others I really suck at.

A Linear View of the Process

Starting

There are a million reasons not to start working on a song. All of them suck and are dumb. Starting a work is hard to do sometimes, and you'll get a lot of internal resistance. Sometimes, I know I "should" sit down and get started on a work, but for whatever reason I just don't have the motivation to get started. If I'm on my A game, then there are a few things I have found to help kickstart inspiration. First off, there is the concept of "mise en place". Is my desk clean? How about Ableton? There are always plugins to manage, presets to build, synths to play around with, and default sets to update. How organized are my samples? Once I get started on that inspiration will start to take hold.

Alternatively, maybe if I'm not feeling the groove, it's time to practice some deep listening on tracks I admire. Or maybe read or watch some tutorials. The trick is to circle around the edges of making music and before long the muse will strike like a snake.

Just push through that. Soon you'll wonder why you were such a prince(ss) about it in the first place.

Conception

The easiest way to have ideas is to have ideas and respect ‘em once you got ‘em. You're just going to have ideas, it's part of being human. The trick is getting more and better ideas. Submerge yourself in the media you love and let it inspire you and teach you. Also, expose yourself to new media and new mediums. Try going in blind. The less you know about it, the more likely you are to experience it for the first time and get the most out of it. Walk in knowing that you could hate it, but that hatred teaches you more about what you live for.

Write your ideas down. Ideas are fleeting, and unless you have a truly great system for remembering them they will go away. I keep a list of all my song ideas in a notebook. This really helps when I want to make music but I am not feeling inspired on what I want to make. I can just scan through my list and see which one tickles my fancy and away I go. If your brain knows that you're going to respect the ideas it generates by never losing them, it'll generate more of them

Writing

In my process there are 3 main steps to writing a song: getting the initial idea together, building a variation, and writing a beginning and end. If I'm writing a song about an existing character, that character already has had a song written about them (probably by the venerable Mr. Fox). I like to plumb the existing work to see if I can pull out a leitmotif of sorts. After that it's all up to the muse.

The end result of getting the initial idea together is a loop that I am able to turn parts on and off and have something that, if you squint real hard, sounds like a song. After that it is easy to build out variations like a breakdown and a recapitulation. I find that they just flow out from me and require no coaxing. It's writing the beginning and end I find a little more challenging. On the one hand you can just add (or remove) parts and... walla![sic] You have a beginning and an end. You also have one that is boring as shit. Lately I've been playing around with focusing on what I'm trying to say with the song and consider how that could apply to the beginning or end. A perfect example is the end of "Crash: Terezi" which has a booming crash finale, a literal space ship crashing.

Producing

A music producer is the one who sits above the whole show and is able to see the forest for the trees. They make the whole thing sound cohesive, coax out the best performance from the musicians, get all right personnel working on the record at the right time... is this role even relevant as a solo creator? You're the one performing, engineering, and recording, why even bring it up? Emotional distance from the song is why. A producer has the ability to step back and maintain distance from the work and guide it. Being able to step back and be reflective about the work is an important skill. It's an entirely different way of approaching the work. The way I do it is to make a number of rough recordings and listen to them during my commute. I'll take notes about what I can improve on and how. Does this provide the same distance as a dyed in the wool producer? Of course not. But it only costs a train ticket, not hundreds of dollars an hour.

Recording

I love to play live techno, so once I have all my loops and patterns assembled, I go through the process of putting those loops together in real time and working the knobs on my analog synths. This is a full performance that I record. Generally I do anywhere between 2 and 20 recordings until it's about right. It has been a number of years since I have done a performance in front of people, so my skills are atrophying. I hope someday to change that.

Mixing

After I finish with the recording comes mixing. This is where all the subtle magic happens that make a track go from sounding okay, to sounding good. I used to think of the mix as just the shine and sparkle and would rush through it. A real eye opener was hearing that a good mix can take just as long to craft as the song itself.

The listening environment will have an effect on the mix one does, and mine is shit right now. It's important for me to try the mix out on different systems and see how well it translates. As an aside, due to the nature of my room a lot of bass frequencies cancel out right at the listening position, so during a mix I need to stand up once in awhile so I can better understand what is happening in the bass range. I call it prairie-dog mixing.

Given that the mix should take as long as the creation of a song, here is the process inside of the process.

Setup and Gruntwork

I start by making sure all my tracks are properly sorted, labeled and coloured. My scheme is:

Snare Drum - Orange
Other Drums - Browns
Kick Drum - Red
Breakbeat - Purple
Bass - Indigo
Synths - Blue
FX and One Shots - Green
Vocals - Yellow
External Analog Synths - Black and white

Ableton live has decent track colouring, so if I have 2 bass's on a particular track, say a sub bass and a higher register synth bass, I might colour the sub a deep indigo, but the synth bass a lighter shade of dark blue. The same with synths, more percussive, weirder synths might get coloured more towards the green end of the spectrum.

I also route each instrument type to it's own sub-mix channel. This makes it easier to manage the side-chaining. My kick side-chains almost everything, my snare sidechains the kick, my vox sidechains the synths and FX... it's like an uncomfortable analogy involving clandestine BDSM.

Rough Mix

The first step is to do a rough mix. I like to start with my most important instrument (the kick drum, it's techno!) and set it's volume, then in order of importance I set everything else to the kick drum, starting with the bass and ... whoops, now I'll need to fart around with the sidechain! The goal here is to turn down the bass when the kick hits, but then turn it up again. Getting the timing and settings on this right takes a bit of time. Then back to setting the volumes of all the other instruments and probably play around a little with my send and return effects like reverbs and delays. Just enough to get a sense of how I want them to be.

Next up is EQ bracketing. A high hat has a surprising amount of low end and high end that just doesn't matter. On it's own it sounds impressive, but in context of every other instrument playing, it just gets lost. All that noise there just eats into your headroom and makes it harder to get every instrument in focus for your mix as well as give you the ability to be louder. I use the same trick with setting levels here as well. I focus on the central instrument and work my way down the hierarchy. With a kick drum I'll be very gentle with my bracketing, maybe cut at ~20-30Hz for the lows. By the time I am getting to the last few instruments, I may be cutting deep into 600Hz territory.

After that I'll start working on things I know will need a lot of effort: vocals, analog synths, drumloops, busy drum patterns. All of these need very dialed in compression to sound good.

The last step is all the other side chains. Everything is a slave to the kick drum. Synths get ducked, fx get ducked, everything: get ducked. This gives room for the kick and really locks in the tempo of the track with the kick. It should be subtle, but there. Very definitely felt more than heard. You have to feel a good ducking.

That's it! That's the first draft of the mix! About 20% there now. Then it's just the process of bringing out the character of every instrument in the mix and getting its sense of place figured out. If you leave everything panned center, you're going to have a very boring and hard to mix track. This is unfortunately a bad habit of mine. The trick is balance, and how to find that balance. If there is a high hat and a shaker both sounds are going to take the same general spectrum, so you can pan one hard left and the other hard right and be done with it. Autopanning helps a lot too. It adds interest and variety while getting the chosen instrument out of the way. Again I keep the most important instruments panned to the center, and move everything else to the sides as much as I can. One should really pan first then EQ bracket, but I'm a slow learner when it comes to panning.

On the subject of autopan, automate all the things. Seriously. The effect is subtle, but it does so much to add interest to the song. As with most things mixing, you are making a lot of small changes (or automations) that holistically gestalt into something awesome. Automate your FX sends. Automate your reverb time. Automate your MAMA!

Smoothing Out the Rough Mix

By this time my volumes are all messed up, and I'll need to re-balance a few instruments. Or lots if I'm not lucky. Righto. Draft 2 is done. Print it. Time to step back and listen to it on a couple of different systems. At this point the steps get less and less generic and more and more focused on the individual needs of the song. Here is a list of some of the changes I made during the mixing process of Orange Hat:

The Clap sound needed repitching to tune it to the song better.
The snare build needed some distortion so the snares popped out of the mix better.
Carve out ~800Hz of the kick drum to give it more space, and boost its top end to give it lots of air and emphasize the hit.
SO MUCH WORK DONE TO THE BREAKBEAT:
- Run it though Ableton's Drum Bus to trash it and compress it hard.
- EQ it hard to bring out the sound (a boost at 200Hz, shelf at 2kHz, cut everything else, low pass at 120Hz).
- Run it through another Drum Bus to trash it again and compress it harder and differenter.
- EQ it hard again to bring out the sounds I want. This is a series of gentler boosts and cuts.
- Finish running it through a glue compressor with very gentle compression.
Add saturation to make the bass poke out of the mix more.
All the synths get EQ cuts in the midrange to make room for other synths. This is really just checking the frequency range where the synth hits the loudest, and cutting that from other synths.
The Analog synths have so much distortion. How many layers of distortion are they on Bro? So many distortion layers Bro. So many.

With the breakbeat I had a goal in mind. I love the sound of The Amen Break, and I wanted to recreate that sound with this other break. That's important to mixing. Have a goal.

For the analog synths the ones I use are two to-the-circuit recreations of the Roland TB-303. They beg to be distorted, and there are so many good ways to distort these synths. Do you want something that screams and squeals? Something that growls and barks? This step takes so long because distortion by nature adds so much mud into the mix that needs to be EQed out. This is actually common in mixing, you'll add an effect and then you find you need to add another one to fix the problems that the first one created.

I also need to spend time on the space of the mix. Reverbs and delays. This is not really my area of expertise. The only advice I can offer here is have a vision of how you want your track to sound in space, and work towards that. One of the things that I learned recently was to do heavy EQIng in the mids/sides. You can EQ out a lot of Mud in the side and give yourself a lot more space.

This is a trick I learned from a master class by Rinzen on his "'Torus' Track Breakdown". I cannot overstate enough how important it is to learn from track teardowns like this. I have learned so much from these kind of in-depth breakdowns in process. Hopefully you will too!

Just as a side note, I found myself re-evaluating some mix decisions as I compiled this list. I don't think that's smart. You need to commit to a sound. Some people go so far as to render their tracks down to the audio only and start there. This isn't my workflow! I prefer flexibility, but this requires discipline to say "no, I'm not going to revisit this synth sound." Until the song really needs it.

Mastering

Mastering is an oft-misunderstood aspect of music. While it's true that if you were to take off the mastering plugins from my song it won't sound as good, all the work to make my song sound good is in the mix, not the master. The master only adds sparkle and shine...wait a sec, that sounds familiar... So, what kind of sparkle and shine? Well there are 4 primary plugins that I use, a sound-gooder-izer, a Tape Saturator, an EQ and a Limiter.

First up, at the advice of a Mixing and Mastering teacher I grabbed the "Oxford Inflator" plugin on sale. It really is a sound-gooderizer. Just do a proper gain stage (you are gain staging right?) and then turn up the knob until it sounds great! Actually with the Inflator I found that you should read the manual and tweak the settings a little to get it perfect. Just read all the manuals for all your musical software. And take notes in your notebook. It will make you a better producer.

Next is the EQ. This is a character EQ, so it's less about surgically cutting or boosting specific frequencies but instead about "warmth" and "character". Right now I am using the Luftikus plugin which you can get for the low low price of 0 dollars. The idea is not only to EQ the sound but to impart some nonlinear analog distortion into it. Generally I boost the bass and the top end, and cut the mids.

After that comes the tape saturator. It literally emulates a reel to reel tape machine both sonically and visually. This is essentially another sound-gooderizer.

The final step in the process is Isotope Ozone. This is a mastering plugin suite with it's own EQ, Limiters, etc. This is a great one to work with because there is literally a "make my music sound good" button that will generate a custom preset based on your music, which gives you a good starting point. It's pretty expensive, however you can usually find this guy for sale. Black Friday usually. As with most plugins, learning to use this effectively means reading the manual and learning how this thing works, and this thing is a beast.

The Process is not Linear

Before I even wrote Orange Hat I had a template set that sets up a bunch of shortcuts for me. I'm almost always going to have a kick drum track, it's always going to be colored red, it's always going to be assigned to a submix, and it will almost always be the source of a sidechain. Having a template like this makes the creation and mixing process so much easier. In a lot of ways I started this song way before I even conceived of it just by having sane defaults.

By the time I have a basic arrangement in place, I will already have my outputs set, my tracks coloured, a rough mix in place and even a crappy rough master ready. As I keep working up and up the chain, everything gets more and more focused. Once I am done recording my mix will be in better shape. By the time I hit my final mix, my mastering chain should just need a few more tweaks.

During the mixing phase, I'll frequently think of flourishes and additions that would make the track stand out, so in that case I'll add them as needed.

For Orange Hat I tried something different. When I was on the final phases of the mixing I pulled all the volume faders down to 0, and brought the kick drum up just to the point where it's hitting the limiter. Then I did a final re-balance of everything else according to the kick drum. This really enhanced the kick drum on the track and was a real game changer for me. Mixing into a limiter (and the rest of the mastering chain) like that really helped give the track a lot of focus. It was also pretty easy to balance the levels.

Staring to work on the master when you're done the arrangement helped me zero in on exactly the track that I wanted. Just as the mixing phase I reset all the mastering plugins to 0 before I completed the final master. I think that is the trick to working in this way: make a decision early, then when you get to the point where you need to commit to the decision you reset everything and decide again.

Learning from the Masters

One of the most important things about creating music is the ability to listen to music and understand why it makes you feel the way you do. Not only in the chords and melodies used, but the arrangement, sounds, mixing and effects. The technique is called "deep listening", but I'll be damned if I can find a lot about it.

As it was taught to me, you choose a track you love, but haven't over-listened to (good luck right?). You'll want something fresh so that it's not super clouded by your own preconceived notions of that song. You're going to give it 5 solid listens, and for the most part you'll want to close your eyes so you can really focus on the sound. What you're going to get out of this is a lot of subtle techniques of music-making that defy linguistic explanation. Also I find that techniques that I've heard about (or read about) actually gel a lot better when I hear them in practice and in context.

Your first listen is a "Big Picture" listen. What is happening in the song? How do you feel emotionally? Where is your attention being drawn? What is the songs intensity curve and shape? How about the craft of the song? Is it hyper-produced? lofi?

The next listen through is a lot more microscopic. What are the parts? What parts are the primary instruments or the focal points? Where are they panned? How are they placed in the music? List out all the instruments and the relevant details.

The next two listens just focus on the high end an the low end respectively. What's going on there? How are the parts balanced?

the final listen is to listen to the sense of space created by the mix. Are reverb and delay effects being used? This is also a good time to pick out any fills and one shots. How are they used? How do they add to the song?

I'd love to see a language built up around this kind of description of song. The theory of music (thus far) is limited to stuffy Mozart worshippers talking about semi-quavers and B-flats that are the same as A-sharps; and doofy guys in baseball caps on youtube with channels named "Shotgun Bass Killerz" asking "do you even compress bro?".

On Flow and the Ineffable

You likely know this state. When you sit down to work, and after a little while you get up to take a little break when you realize you're shaking because of hunger, it's dark outside, and you're amazed at what time it is. That state is flow, and it is one of the key states to get into in order to complete a work. As a creator if you can understand how to induce the flow state, especially for yourself, you will be more successful at finishing a work, and you'll feel more fulfilled about it as well.

One can cultivate this state. Some of the requirements get pretty existential and philosophical: can I really expect to master a track that is comparable to the mastering houses out there? Am I biting off more than I can chew? Some of them are just hard rules. You can't be in a flow state if you're constantly distracted by everyone around you.

So what are these requirements? First of all, you need to know stuff. In some senses it's really basic, but figuring this out makes the whole process go a lot smoother. You need to know, in no particular order:

Where to go, what is your finished product.
What to do, what are the steps to get to where your going.
How to do it, the process of taking the steps and course-correcting them.
How well is going, are you even doing the right thing?

You also have to believe stuff. You have to believe that the thing is within your skill level to complete. There is a sweet spot of challenge vs. skill. If the thing you're doing seems too easy then it just becomes grunt work. If it looks too hard then you're going to be hesitant at the prospect of bashing your head against a brick wall. The important thing here though is the perception of your skill level, and the challenge of the ask at hand. As long as you believe it is possible and you're exerting the proper effort to get it you can stay in flow. It doesn't matter if you can actually do it, as long as you think you can. This is what I mean by a philosophical question.

But then, what if you believe that you can do it, but it turns out it's beyond your ability? Either all it takes is more time and practice, which is perfect because flow collapses the perception of time; or it is far outside of your ability to complete. In this case you're in for a flow-crash (my word for it). This means you don't have good answers to the 4 main questions (Where, What, How, Howsitogin). Find better answers to those questions.

Finally, the more prone you are to distractions, the less likely you will be able to stay in flow. So turn Discord off. You'll thank me later.

Finishing

Perhaps one of the most important pieces of advice I have to offer is to finish the work. Don't rush through it, but don't leave it hanging. The only way to get better at what you do is to keep doing it. It's better for a potter to throw a thousand mediocre pots then to attempt a single master-work. I have the tendency to get a good idea, and hold it in my pocket until I can make it perfect. This usually results in it not being executed. Perfection is a great ideal to strive for, but a too-tall yardstick to compare against.

It's better to execute poorly on an idea, learn from it, and move forward, than get paralyzed by brain-crack.

Easier said than done.