February 4, 2014

It's Process Day!!!

by Chris Randall

My attempts at creative endeavors over the weekend were utterly and completely foiled by hardware (not software!) problems. My Maschine Studio got crashy all the sudden, I had to whittle a new tape loop for the echo, the old Doepfer modules/Stackables problem reared its head, and basically everything was conspiring to keep me from making music.

However, creativity struck last night and I was able to pull things together on a track I've been making on and off for a couple weeks now. Just for fun, I let the GoPro Hero 3+ run while I was trying to come up with a part on the DK Synergy for this track. So the video above is actually a snapshot of my writing process, not a finished and arranged song (or real-time improv, like most of my videos, although it does smell like that.) When I'm doing a track with full production that isn't real-time, I like to separate the parts out in the Clip view in Live, blow up the UI so I can run it from the touchscreen, and dick around with different arrangement ideas while I'm trolling for sounds, and that's essentially what I'm doing in the above video.

Sidebar: the DK Synergy is a strange and wonderful beast, and I dearly love owning and playing it, but Jesus fuck the fan in that thing is loud. Something needs to be done about it.

I'm intrigued to learn about your writing process. Since electronic music is almost more about sound design than songwriting, do you play parts first, then do sound programming like me? Or do you come up with cool sounds, then figure out how to use them? Or some other method entirely?


Page 3 of 6

Feb.06.2014 @ 5:41 PM
I program my synths now and then and hopefully end up with a few usable presets. I often get some ideas during the programming that I try to memorize - I rarely record anything at this stage. Otherwise I just sit down listening through those presets, adjusting them, adding effects and turning knobs. If something comes up I start recording clips in Live. It can be anything - a melody, bass riff or a pad. I never start with drums though. When I feel I have enough it's time for sketching in the session view. Many projects ends here with just piles of eight bar clips. I always think they might be useful in the future but to be honest I never go back. I even still have eight bar clips on cassettes that I never use. If the session begins to look like a song I record it into the arrangement view and continue my work there. If I'm lucky I have a finished song.

Feb.07.2014 @ 7:23 AM
beauty pill
It usually starts with a melody in my head or just a feeling. It is usually totally inconvenient and I end up having to clandestinely sing into my phone at a restaurant or something like that.

- c

Feb.07.2014 @ 7:34 AM
Jason Duerr
Writing practices: The best songs i write are instant an have almost no effort. Lots of time in polishing, of course, but the main themes and ideas are really fast. 15 mins to and hour at most.

If I'm not in that hook, then i revert to practice - shaping a single idea or method.

Feb.07.2014 @ 3:26 PM
In Black
I usually have an idea for a bass line in my head, punch in the midi data to Live, then feed it to whatever synth I feel is most capable of generating what's in my head. Often designing the sound leads to an entirely different bass line, which I used to fight, but recently figured it best to let ideas evolve. Record and repeat for rest of song.

Feb.08.2014 @ 2:28 AM
Yeah, I do the sing/whistle into my phone a lot. The Ableton audio to midi feature is pretty awesome in that context.

Then again the process of translating what you hear in your head into a DAW, is often full of happy accidents. Just gotta follow them and there are often ideas there that are better than the whistled stuff in my head.

Feb.09.2014 @ 12:08 PM
mike kiraly
I'm not sure that I'm the only one who does this, but I'm the only person I know or I have heard of who writes in this particular fashion.

I have a Live template which I've refined over the years that is ready to receive some basic input from me before the process kicks off. Primarily, I need to define a scale and input it into a few MIDI or Max4Live devices peppered throughout the various tracks. I have to select a few drum sounds to serve as a foundation (a kick, snares, hats, etc). I also need to decide on some sound sources, i.e. which synth and starting sound or preset. On each track there are anywhere between 1 and 20 modulating devices which will create organized chaos. Mostly it's subtle randomness, but no matter what, everything is changing constantly.

Here's an example of how it might work on a single track. Let's say I need a bassline. I'll have an instrument track loaded with a plug-in, perhaps Trillian. In front of that I will have some Max4Live MIDI note generating plug ins that will generate notes based on a defined scale and a defined probability of certain notes and timing divisions. I might also have step sequencers which are modulated by other sources, perhaps an LFO continually altering the step length. Also on the track will be other modulation sources set to alter various key parameters, usually in a subtle fashion but drastic changes can come into play as well. Common selections are things like cutoff, decay, release, FX wet/dry, send levels, tuning, transpose octaves, add chords, etc. I'll also have MIDI plug ins that will create very, very subtle variations to velocity and timing

So when I hit play, I have an instrument track "writing" a bassline on it's own, but in a controlled fashion. And the beauty is that it's constantly evolving. But that's not even the fun part. There will also be anywhere between 20 and 50 clips on the track (in Session View) which are loaded with Follow Action instructions, all set to play "Any" after the currently active clip is finished playing. Each of these Follow Actions are set to occur after varying time lengths - the first clip might have an action after 1 beat, 2 beats, a 16th note, 1 bar, whatever. And they're set to legato mode, so the clips won't retrigger when launched. Each of the Follow Action clips have different clip modulation envelopes inside of them which will do all sorts of different things - turn devices on and off, change presets, mute the channel, or anything else that I can think of. So hitting play will create a chain reaction of events that I have guided but cannot predict.

Then I hit record and walk away for 10 or 15 minutes. Live will do it's thing and when I return I will have an amazing wealth of content that I can now pick and choose from. I'll grab loops, whole sections or specific phrases or transitions.

But since I have set this process up over multiple tracks that are staged for specific kinds of sounds, when I return from my 10-15 minute walk I actually have content for all of the elements I wanted in the track. And playing them all back simultaneously will also unveil interesting arrangement ideas, transitions and combinations.

This probably sounds overly complex, but since I have a template in place, it's only a matter of minutes to set-up and then I'm ready to record.

I know this will invite criticism (hey asshole, you're not writing anything!) and I recognize that this method shifts some emphasis from writing to editing. But if you were to try it you'd realize that it's really not much different than noodling around on a guitar or synth until you hear something you like. My output will always sound like me because I'm the one inputting the instructions and I'm the one who selects from the results. And of course, this doesn't help shorten the mixing process which tends to be the most time consuming part of any song I write.

Last thing: I also use (abuse?) this process to force my brain to avoid typical mixing and arrangement conventions. For example: as I mentioned above, most of the individual tracks are set to send to various FX returns at random times defined by modulating devices. When I hit record, the return tracks are also printing audio. I then take that resulting audio track, chop it into 1 bar clips and place those clips on it's own channel in Session View and repeat the Follow Action process. But I'll also add in numerous empty clips (no audio) and hit record again. Now these FX clips will play back in random order, for random durations with passages of silence inserted in at random too. When I play back this new track, I can come away with arrangement ideas that will defy logic. Perhaps a trailing delay that hits on the 1 from a synth that wasn't playing. Or a reverb wash from a variation of a sound that never made it in to the track. These things add depth and character to the track - and because they originate from elements in the track itself they don't feel like cheap beeps and boops added after the fact. They're cohesive and enhancing, not distracting.

Feb.09.2014 @ 12:30 PM
@mike kiraly, I honestly think that sounds like an interesting way of doing things. I wouldn't approach songwriting that way ordinarily , in a constrained random sense, as I generally philosophically feel "intent" defines art, but I have definitely edited the crap out of random noise to create flowers from chaos.... and this is just one step removed from the old eno axiom "of honor your mistake as your hidden intent...".. You are setting rules, and then defining what it is you consider worthy. Not much different then a lot of needle drop guys, who are definitely putting less thought (and time ((geez, like to see a session of what you got going on)) into the prior conditions.
So no hate here from me!

but I could see it being amazingly functional as a slump buster, and definitely could generate some interesting timbres/counterpuntal melodic lines etc... and as you said, it's not "that" different then noodling... Have any of these results you can post?

Feb.09.2014 @ 1:01 PM
Long-time lurker, first time poster. Loving the conversation about creative process, as I find for me that's the hardest part of making music. That and making the time for it.

My process is probably similar to a lot of you. It usually starts with fiddling around on one device or another. My go-to is the piano since I've been playing for years. But I might also start by designing a pad on the AniMoog or fiddling with knobs and faders on my Minibrute until a get something gritty that inevitably involves a distorted square wave.

I'll eventually stumble across somewhere between 2 and 4 bars of what might be a neat bassline or the beginnings of a melody, and I'll decide to build from that and see where I can take it. Usually that means sequencing it in FL Studio ( I can hear you all judging me, but I tried Ableton and it just wasn't my thing ) and then adding a few drum loops to try things out. Once I'm happy with it, I'll probably layer in a pad or 2, and then start varying the synth patterns and the drum loops until I've got around 4 variations on the theme and about a minute of music. I'll probably toss in some reverbs, delays, and the occasional flanger.

Often I won't get past that point as I won't know where to take the piece.
I have dozens of little 4, 8, and 16 bar loop experiments that sound interesting on their own, but won't necessarily work as part of a longer piece. But I keep them around anyways, and every once in a while I'll find a place or a user for them.

Otherwise, If things seem to be working together well I'll make sure I've got all the loops I need, rip it apart, and start building the elements back up into some sort of song structure.

I find a tend towards certain keys and rhythm patterns. So if anybody has any really good tips on how to introduce a little disruptive noise into the creative comfort zone I'd love to hear them.

Feb.09.2014 @ 1:24 PM
mike kiraly
@ashtarbrian - I'm happy to share some results. The Soundcloud link below is to an EP release that's coming out at the end of the month. The first two tracks are my originals (the others are remixes from other artists).

This track was written and arranged via the described process. The first clip (Original Mix) is how it all gets tied together, but the second clip (Dub Tool) is a better way to hear how it all works. The Dub Tool is essentially a stripped version of the original. There's no kick or bassline so the melodies and sound design content are much easier to focus on and listen to. If you were to break it down, you'd see that there is really only 3 or 4 main elements but that those elements have spawned dozens of different variations.

Oh, and those with good ears will be able to spot heavy usage of Vapor, Eos, & Discord 3 ;)

link [soundcloud.com]

Feb.09.2014 @ 1:31 PM
mike kiraly
I'd also like to add one more thing: this is not an entirely "in-the-box" process for me. I've got 27U x 104HP of modular delightfulness and several outboard synths and drum machines that are also controlled and incorporated into the aforementioned Live template. The modulating devices and Follow Action clips are semi-randomly impacting or triggering sounds, patterns, sequencers, etc etc.

Page 3 of 6



Sorry, commenting is closed for this blog entry.