||What's your workflow?
| br>Workflow is one of the most common terms used in these posts, and I am curious-- How do you use your instruments? (I'm talking about workflow as organization of process, not where you mount your modules.)
As an example, my workflow is most often as follows:
1) Build, buy or borrow a collection of devices that will be used in the composition. I started out in the era when DIY was the only practical path to electronic music, and I view the construction of contraptions as part of the compositional process. This part of the process can take days (if I'm writing code) to months for building elaborate systems. (My most ambitious building project was wiring a gamelan orchestra to capture MIDI data.)
2) Learn what the system can do. Contraptions never do exactly what I expected, sometimes they need a bit of tweaking, sometimes they give me happy surprises.
3) Define and practice a composition. They generally have a heavy algorithmic and improvisational content, so I have to learn how to work with whatever the system gives me.
4) Perform it when I get a chance. I'm not so much into recording.
So what's your workflow? Are you a new gadget = new piece composer like me, do you jam with a fixed set of instruments, do you score everything ahead of time? I know there are lots of ways to wiggle, so tell us yours. br> br>
| br>Interesting question - the kind that makes you question what you do and why you do it, which perhaps leads to ideas about doing things differently.
I don't differentiate between research and composition (and one often leads to the other), and I still have so much to learn about all this stuff that I can see that approach keeping me going for a while yet. I'll usually start with a loose idea of "exploring this feature of this module", or "what happens if I use X with Y", or try to approximate a complete patch I've seen someone else do. The more things I learn through this process the more patching nous I have, and when it does start to become music I have more tricks up my sleeve to make it interesting.
More recently I've made attempts at designing my own systems of generation from scratch - with mixed results. I'll get there, but in the meantime learning and having fun is enough motivation. br> br>
| br>Well I'm still rather a novice at all of this, and the modular is my only instrument, so often "research and composition" are happening at the same time. I'll often start with an idea in mind, but sometimes while just noodling about I run across a sound that gives me an idea and I extrapolate a composition from that point.
Either way I have a Zoom H4n at the ready, and when I have something i like I record it for say 6 - 10 minutes. After that I make adjustments, maybe change a voice, and record it again in a different manner. Sometimes I'll do this enough times that I'm recording something entirely different from what I started out doing in the first place, but all the while I'm tweaking and learning. This has provided me with a sample library over the years that I can use together to multitrack into larger compositions, which I have done a couple of times.
If I have a composition that I really like then the real work flow begins; I let it run it's course through whatever engine I've built it with while I clean house, do the dishes, and other such things as I listen to the subtle changes being brought about by what I've put together as I try to understand the magic happening before me that it may be reproduced by me in the future.
I don't know that I would have nearly as much fun were the process to become a job and have to be disciplined as such... I like just putzing around at my own pace. br> br>
| br>Fog Door
| br>Wow guys, my flow is much more humble - I've got a notebook full of pretentious track titles, I pick one that I'm feeling the "vibe" from, then imagine what it could sound like. Then I start running through a few different set ups of different gear, because I don't just use modular. Then I record a whole load of one take stuff to the DAW, keep it all, no matter how perfectly awful it sounds. Once I am satisfied that enough layers of mud have been flung against the wall for at least some to stick, I try to "find the track in the mix". Rinse/repeat br> br>
| br>1) I turn on all my equipment and let it warm up.
2) While waiting for (1) I surf the internet or turn on a game.
3) 1 hour later I say to myself, "Who the fuck are you kidding? You have no talent and lack even the smallest amount of motivation to write any music. Just give up."
4) I turn off everything and continue with whatever I was doing in step (2). br> br>
|Are you a new gadget = new piece composer like me, do you jam with a fixed set of instruments, do you score everything ahead of time? |
I don't score anything note by note. The collections of modules and the patch are a partial score I suppose. I may start with a particular 'gadget' and see where that leads for a while. Start somewhere, listen to what you have, let that suggest the next step. Be like a sculptor examining different aspects of your rock from different points of view. Maybe focus on timing or pitches or timbre or scale... always listening, reevaluating, changing your point of view.
Mostly my work flow starts by thinking in terms of instruments. What does it take to come up with something you can sit down and play? Rack up some modules and try it. You probably want more than one voice. How many? Struggle with optimizing it to not be too big, hard to use, or too narrow in scope. Keep going over it and think of metaphors - what does it remind you of? Listen, tweak, build, rinse and repeat.
Next comes the compose/improvise divide. I like to improvise but that doesn't mean you can't have intentions in mind when you start.
I like FogDoors notebook idea. br> br>
| br>Basically what Fog Door said.
1. Notepad in phone full of song names
2. Hit record on my DAW and jam on the modular
3. "Oh, that sounds pretty good" -> put marker in DAW
4. Render selected region to .wav
5. Pick most relevant track name from list and save
This is how all the pros do it, right? br> br>
|Fog Door wrote: |
|Wow guys, my flow is much more humble - I've got a notebook full of pretentious track titles, I pick one that I'm feeling the "vibe" from, then imagine what it could sound like. |
IIRC Ray Bradbury used a similar method for composing short stories, damned if I can find the corroborating quotation though. I think he may have explained it an a foreword somewhere; or maybe I’m just misremembering.
I think with instrumental music particularly, a good title can go a long way towards framing the piece and giving it enough of a whiff of meaning that an image can begin to form in the listener’s mind. “East of the River Nile” is a great example.
My own workflow, the last time I was actually being productive and finishing projects (four years ago now ) revolved almost entirely around listening habits, which worked because the only four pieces of gear I had to use were an MPC, a korg EA-1, a bottom shelf Yamaha keyboard, and a pc of 2004ish vintage with an equally vintage DAW (Cubase 3).
I kept a running Ur-playlist on Spotify of music that consistently grabbed my fancy, mostly 60’s psych and pop, but anything was game. I would listen to whatever felt right on this playlist with the electribe in the signal path going into my PC and sort of “play” the track - the EA-1 can do some gnarly filtering and delay stuff if you know where the sweet spots are.
I recorded hours and hours of this, and then turned around and used those recordings as raw materials, sampling out of the PC to the MPC via-SPDIF so as to avoid hitting the 2kxl’s converters (which have a certain sound that does not lend itself well to “dense” material).
Then I cut about ten thousand loops from those. Figured out a concept I wanted to pursue (ended up being a riff on Dostoevsky’s Demons, which I actually can’t even vaguely get into because of this forum’s policy on political content) and then just started smashing samples together, all while daily growing the Ur-playlist and looking for musical, thematic, and lyrical synchronicities between basically anything and everything.
I worked on this thing for about three years, even finalized nine-tenths of it, before I realized 1) that I have no interest in the music business whatsoever, and that if I wasn’t going to commercialize this project there was basically no reason to lose my marbles trying to finish it, and 2)that some important decisions I had made early on in production re: mixing the low frequency information meant I was never going to be able to press it to vinyl (something I’ve wanted to do for a couple decades now, put out an actual record) without radically reworking it. Basically I just said the hell with it.
In the intervening years I started having some disposable income, started building modules, and instead of recording stuff I spend my free time troubleshooting problem builds and reading this forum br> br>
| br>I usually start at some point with two clocks and a or module and then got lost. br> br>
| br>Noodle Twister
| br>I think the idea of thinking up names for tracks first is a good one.
Once you have a few creative seeds (names) you can think about what to include what to leave out, try out some ideas and get started.
I still feel like I need to build some more modules first but I sometimes think,
just record some stuff with what you have. Great things can come from simple ideas.. br> br>
| br>I just purchased one of these:
I had an idea driving home yesterday to make a few lists of instruments/patches/effects/functions/etc. and let the dice roll choose what I combine for a performance/recording. I have a great variety of equipment and use it all, but some way more than others. This composition by chance will force me to use combinations of things that I may not have considered before and will help me be more familiar with devices I may have used as pigeon holed functions.
So roll a 7 on list 3 and you get ARP Trumpet. Roll a 6 on list 5 and get Korg Kaoss Pad Program 21. Etc. Down to which beaters to use on which percussion or which VCF to use on which looper...
Should be fun making the lists to start. After that... we'll see what happens. br> br>
| br>Really interesting thread.
I do quite a lot of jam style, real time track building.
1- patch a few voices. Maybe Vco- vcf - vca with envs, maybe vco - folder - vca, maybe self contained voice - nothing. Maybe something more esoteric like sample> rings, or clouds ( not both) Rarely, more experimental stuff such as looping envelope as oscillator etc etc.
2- plug these into a mixer, along with some drums ( pico/ bitbox/ bia) and samples ( movie dialogue, chanting, ranty evangelical or political speeches etc)
3 - choose a few notes in a sequencer, and mult them everywhere
4- run a metric shit ton of clocks, S/H, Euclidean sequencers and linear BSP sequence tracks, relying heavily on quantisers to keep it roughly sane ( Sinfonion FTW)... I’m up to about 17 triggers usually.
5- hit record, sometimes stereo, sometimes 6-8 tracks in a DAW.
6- run all and tweak/ mute / modulate - often clocked mods.
7- delete or save result as appropriate
8- edit and basic mixing
10- remember effective patches/ sequences / sounds for reproducing at live shows
11 - play a couple of live shows, according to above method.
12- review, research, add modules, subtract modules, consider a larger case, and the validity of my life choices.
13- go to 1
Other than that cptnals technique speaks to me.... a combined exploration/ research/ practice/ happy accident gathering exercise that sometimes ends in a recording, and sometimes not.
Apart from when patching and practicing for a show, which can be 48 hrs- a week or so, I pull every cable at the end of each session, with the exception of clock distribution and a handful of BSP>drum trigger cables. br> br>
| br>I have a slightly different definition of workflow; to me, the definition in the OP is more 'what is my process' than 'what mechanisms do I have in place to achieve my goal'.
The last thing I do is turn on my synths when I am working on something new. I rarely resort to open exploration or hunt-and-peck to find ideas. For me, it starts with an overall concept: This is going to be a piece using X form of approximately Y length; it will have a tempo of Z, I will explore time signature and tempo changes of AA and BB. Oftentimes in concert with that, I will pull out manuscript paper and start working on motives, themes, and ostinatos -- still keeping this purely as a thought-exercise, working out the structure and rhythm in my head.
Once I have a composition outlined and know what I am going to do, only then do I fire up the synths and start figuring out, timbre-wise, what voices will go with what melodies. Many times I will sketch out motives using my DAW just to make sure my voice leading is sound.
Once I have a few bars complete, I'll play back what I have, make any needed changes or improvements, and then keep going until the composition is done.
When the composition is complete, I'll record it as-done and then listen to it over a period of days/weeks/months, rewrite it as needed, and then decide if I'm going to commit to it or put it aside for further work later down the road.
Sometimes I will take a completed idea, and just pick elements from it to complete another piece which I think matches it better.
That is my process; my workflow is designed so that when I get to that last stage, I can get all of the sounds I need without having to rewire anything. This allows me to record live in toto even though I am often a solo performer and there are multiple instruments happening at the same time. br> br>
| br>I also have a list of song names waiting to be put to use. Once and a while when I'm experimenting I'll run across a sound and think hey... that sounds like...(A, B, or C) and off I go! br> br>
|Noodle Twister wrote: |
|I think the idea of thinking up names for tracks first is a good one. |
I've been keeping a small notebook with me for the past 35 or so years, and whenever I come across a phrase that I think would make a good song title I write it down. Even though my first notebook was stolen 20 years ago, my replacement notebook currently contains far more titles than I will ever be able to use before I die. br> br>
| br>Thanks for your replies-- it's fascinating to see how many ways there are to make music.
It never occurred to me to base a track on a name. I usually stick to descriptive titles like Twirl for a performance driven by gyroscopes in a Nerf ball or Push for one involving a broom. I have lots of notebooks with designs and patches, but no titles. br> br>
|Pelsea wrote: |
|Thanks for your replies-- it's fascinating to see how many ways there are to make music.
It never occurred to me to base a track on a name. I usually stick to descriptive titles like Twirl for a performance driven by gyroscopes in a Nerf ball or Push for one involving a broom. I have lots of notebooks with designs and patches, but no titles.
Now I feel basic, so many of my song titles are simple things like a time or place or object the music reminds me of, or often just random words/phrases that happen to sound good to me at the time. br> br>
| br>I may have described my working process somewhere recently, but I don't remember where. It's a little unusual, as (apparently) are the outcomes.
1. I have no ideas in advance about how it will sound. I begin to imagine complex control structures in my mind. Complex feedback loops involving lots of LFOs and sample & holds, usually. Sometimes with gate modifications and logic.
2. When it becomes clear enough, then I draw out an abstract patch diagram of the control structure (i.e. not naming specific modules).
3. With power off, I patch up the control structure. Then power it and watch the blinking lights.
4. Next step is to come up with an audio patch that includes VCOs, envelope generators and VCAs, maybe filters, and some CV-able FX. It usually involves some kind of cross-modulation between VCOs, such as dynamic depth linear FM.
5. Then comes the patching of the controls over to the audio. This is always exploratory. I have a bunch of LFOs, gates, and S&H outputs available to patch to control the audio.
6. Once I have a suitable patch, then comes the twiddling and wiggling. And some re-patching. This takes the longest, because of sensitive dependence on initial condition. Just a small tweak to an initial rate or CV depth can make a lot of difference. I fiddle, lie down and listen for a while, make some short recordings for reference. This can take several days, part time. Eventually I decide it's ready to present to an audience.
7. The last step is presenting it to a live audience. I explain that the 'music' is being generated in real time, using no recordings or samples, and that it's all the movement of electricity until it hits the speakers, which transduce it into sound waves. I'll let it run for an arbitrary amount of time.
I sometimes have a specific future date in mind, and a place where I'm supposed to present. The process from step one to six can take anywhere from two days to two weeks. After the presentation, it's over and done with. I save recordings, but the 'work', such as it is, is always a live presentation.
A composer friend of mine asked me, 'Since your pieces have no beginning or end, how do you think of them as beginning or ending, since they do start at some time and end at another time?' Good question. I think of a composition as beginning after step six and ending when the patch is torn down. Sometimes I draw out the patch, but as often as not I don't. It's ephemeral and temporal, existing in the ears of the listeners. Even though the patch and tweak settings exist when the power is off, the cold patch is not the composition. It's really existing only when speakers are on and someone is listening.
EDIT: Copied here:
https://pugix.com/synth/my-work-process/ br> br>
|cptnal wrote: |
|I don't differentiate between research and composition (and one often leads to the other), and I still have so much to learn about all this stuff that I can see that approach keeping me going for a while yet. |
I agree, these things tend to go hand in hand for me too.
I usually find stuff to build a song when I am jamming, exploring a module or practice on the guitar. It is quite often connected to a feeling. So If I am emotional - happy, sad, thoughtful whatever - during for example practice I tend to find something that sounds good.
Then I isolate that initial idea and try so build around it. Absolutely essential at this point is to play it over and over and to write it down and/or record it. Then I play this phrase all the time. It usually ends with a song, or at least the structure for a song.
After that I try to record a bit. This is really fun, because now it becomes a question of composing and arrangeing. The idea usually morphs into something different now. More sections come in, previous stuff is ignored or changed to a different song.
I borrow instruments, call in friends to play stuff, figure out harmonies. It is really fun, but sometimes it makes me anxious because it doesn't sound good and then I feel like a bad musician. I try to ignore those feelings and be humble instead. It is always important to have fun and don't worry that a track take months to finish or that I perhaps have started a recording process but only wanna jam on the modular. It's OK.
After that I print the entire song, listen to it on different speakers, make adjustments and then it is DONE. br> br>
| br>I love to improvise. I played jazz in high school, and performed in a taiko group more recently, and honestly I like improvising a lot more than playing someone else's music, or even writing my own and then performing it.
I also don't want to get into the state I used to get in with 100% ITB production, where I was making lots of little tweaks and adding stuff just to make it different when it wasn't actually getting better.
1. Fire up the modular/DAW/etc. and patch something. Sometimes this starts with something I was curious about, but often it starts with "I'm going to listen to this VCO raw for a while..."
2. Expand on the patch, add modulation and FX and manual controls, just generally mess around.
3. Once I start hearing something I like, or that demands to become a song, I switch mindsets a little. I figure out in a general sense what I want from the song, and I build out the patch with additional voices (if needed) and any sequenced parts, FX chains, control scheme, and I retune things as needed. I jam with it a while longer, discovering what I can do with this patch and maybe practicing particular combinations/phrases that particularly catch my attention. I might make an informal, partial plan for the recording.
4. Record an improvisation. The full stereo mix with the effects baked in, no stems. If I screw up, delete the take and start again.
5. Make minimal patch notes (the flow, not specific settings) and unpatch everything.
6. Edit the stereo file, in Sound Forge Pro. Destructive edits, though I keep a backup of the untouched recording. Head and tail trimming and fades (and extension via delay/reverb/timestretch/copy-and-crossfade), some EQ adjustment or maybe more reverb if I feel it was too dry. Sometimes I make more major sonic changes here -- maybe even re-record the whole thing through my modular, pedals or a dictaphone. And I get levels within a reasonable range with plenty of headroom.
The whole timespan is about 1-6 hours, more typically 3-4.
Other times I choose not to record, and just get lost in a self-playing patch or a drone where I make minimal tweaks for a couple of hours or so. br> br>
| br>Sketch a narrative in the Daw
Fire midi sequences from said sketch
Record an unhealthy amount of takes
Edit and mangle the whole thing
Overdub and redesign some sounds to have a fuller sounding mix
I almost never start wiggling at random, I usually always have some ultimate goal to achieve before I start the machines. br> br>
Powered by phpBB © phpBB Group