What is your sound design process?

[Immerse]

Hey guys, lately ive been struggling to get consistent results when synthesizing. Mostly due to the fact that i dont have a routine to fall back on and improve upon.
So i made this thread in hopes of picking up some techniques you fellow heads use to get your ideas out.

Topics:

• What parameter do you find yourself tweaking first?:

  • mainly just go about shaping the original osc waves first ofc, then filtering, etc. specifically, when i rewire reason to logic and use malstrom, i prefer to find a waveform that i enjoy, and then i move the motion knob way down and then slide the index slider around to find individual grains that i like. its crazy really how many possibilites there are with this. once i find a waveform and an area of the grains and a rate of motion i like, i move to the second osc to shape the sounds to what im looking for.

    i normally try different combinations of oscs until i find the kinda style im going for then route the filters (in massive) going through each other, so osc-filter-filter-instert-fx-master, if im not making a wobble theres still usually a small amount of movement from lfos or envelopes to keep the sound interesting.

    Rough osc/sample sounds, amp envelope, filter envelope, tweak all above, effects. That's about mine.

    Sound -> Filter -> Envelopes for filter -> tweak sound

  Do you design around a loop, or in the context of your song?:

  • I'll usually just start with a saw wave, I'll carve out a bassline that I like, and THEN from there, I'll start to design the sound I want. I find it easier that way because it's much simpler for me to design the sound around the melody rather than the melody around the sound. It gives me an Idea of what type of sound I'm looking for as opposed to just fumbling in the dark for a random sound that I may like. BUT, then on other days (pure sound design days), I go into whatever synth I'm using with a goal, Either I want to make a pad, a stab, a lead, or a bassline. And then I just go to work.

    create 'em all on the spot... but they are already there to begin with. i've made so many damn sounds, i know what i like to hear... but creating them fresh each time gives them chance to evolve and take different directions. i very much like to use classic simple sounds in new ways, i get a lot of mileage out of saw and detuned saw, lots of lowpass filters (especially with downward envelope) lots of gritty claps and hats, and deep kicks.

  According to ill.gates, it is best to keep left and right brain process separate, such as arrangement and sound design. If you design in the context of your song how do you get around this, or do you just disregard this advice?:

  • I rarely sound design while writing a song. I have sound design sessions separate from my writing sessions (because spending too much time tweaking a knob can kill the inspiration) that generate me a lot of patches or ableton instrument racks to use later.

  How long do you find yourself doing sound design sessions?:

  • It varies. Sometimes I wind up with a sound I like in 5 minutes, sometimes I spend hours tweaking something only to find the next day that it sounds god awful. This is part of the reason why I bounce out religiously and save new versions of every project so often. Maybe I made a sound that is decent, but then just spent the last hour ruining it with effects that I thought sounded great at the time. If I didn't resample, I'd pretty much be fucked and stuck with a shit sound, but since I did, now I have the older version or versions of that sound to go back to, and the methodology is saved in the effects chain if I ever want to go back to the original patch.

  How long do you tend to spend on one patch?:

  • Sometimes it takes me days, sometimes it takes me minutes; it all depends on the type of the sound I'm trying to achieve.

  Do you bounce your patches to audio for organization? If so, how do overcome obnoxious things such as making the audio file endless, or preventing the sound from changing speed from note to note when using a sampler?:

  • I am also really well organized (had to learn this the hard way when I had separate projects with basses and leads named the same thing all in the same folder-> needless to say I scrapped those songs). If you look into the "my beats" folder each song will be organized in its own project folder. Furthermore, I separate all the elements in the song. I'll have a folder for "drums" and "lead 1" and "lead 2" and "bass 1" and bass 2, etc and "samples". And I also use the first few letters of the track name before every sound so that I don't have 30 projects with files named "snare 1".
    In regards to the second part of the question, I know there are plenty of samples that can change pitch without changing the speed, but I'm not usually interested in that. I don't typically take on filter movements until AFTER it goes into a sampler, so it doesn't matter if it gets sped up or slowed down a bit. I usually just export 8 bars of one note (usually C, but sometimes I like the movement better on other notes), and that way I know if the speed changes it won't matter, because I still have 8 bars to work with.

  Misc. Tips and Tricks to better synthesis workflow:

These are just a few to start, and ones that ive been curious of. They arent amazingly thought provoking but hopefully it will spur some conversation. If you try to keep your tips in Q & A form, ill try to keep the topics updated with responses. But feel free to post in any format, as all contribution to conversation is appreciated, lets see what you got fellahs

[Eat Bass]

good thread idea. im wondering the same myself.

lately i've been trying to get all the sounds im going to use in a song before i start recording/arranging. tonight i spent 2 hours making bass patches for a song. i made about 3 nice bass patches, but i only want to use one for this song. im having a tough time designing a bass patch that fits my theme idea for the song. so sometimes ill spend days in the sound design process for a song.

as far as what i go to tweak first. mainly just go about shaping the original osc waves first ofc, then filtering, etc. specifically, when i rewire reason to logic and use malstrom, i prefer to find a waveform that i enjoy, and then i move the motion knob way down and then slide the index slider around to find individual grains that i like. its crazy really how many possibilites there are with this. once i find a waveform and an area of the grains and a rate of motion i like, i move to the second osc to shape the sounds to what im looking for. malstrom is definitely my favorite synth.

im also pretty clueless about the when to resample and when not too. personally i like to have the original synth so i can tweak and automate.

[Immerse]

usually, i tend to do adsr, then osc 1, then osc 2, filtering is usually dead last. dont follow this it really hasnt been working for me hah. tbh most of the time i just mess randomly with knobs and sink tons of time into a patch that turns out sounding either a. like shit or b. good but doesnt fit the tune.

[bigdaveo11]

also extremely interested in this thread, hope it takes off. Anyway still a n00b so can't contribute much here except that i have been learning my chosen synth inside and out before moving on. I have been practicing frequency splitting but still can't seem to "understand" when I should go about splitting up the sound or layering up the sound. I am improving on analyzing my favorite tracks and attempting to replicate them as this way has helped me learn quite a lot, would be interested to hear people's thoughts/process in terms of attempting to replicate other's sounds.

[JTMMusicuk]

i normally try different combinations of oscs until i find the kinda style im going for then route the filters (in massive) going through each other, so osc-filter-filter-instert-fx-master, if im not making a wobble theres still usually a small amount of movement from lfos or envelopes to keep the sound interesting.
After watching ill gates workshop i have been trying to sepearate my right brain tasks from left brain tasks which means only doing technical things like sound design in periods where i dont have enough time to complete the structure of a track or think of new ideas to put into it, it really does work, you end up having that 'creative flow' for alot longer than usual, as soon as you try and switch between left and right brain then you get stuck and end up listening to your half finished song 100 times without doing anything at all.

[Aufnahmewindwuschel]

envelopes are the shit
add them for example to a filter and vice versa to the pitch or resonance of the filter and you can shuffle your bassline however you want without lfo trouble or resampling and cutting adding a distortion to a specific kind of range can lead to intervalls between the basssound padsounds while rising or lowering the filter its like breathing before the bass hits or breathing after the bass hit

[therapist]

Rough osc/sample sounds, amp envelope, filter envelope, tweak all above, effects. That's about mine.

[e-motion]

I rarely sound design while writing a song. I have sound design sessions separate from my writing sessions (because spending too much time tweaking a knob can kill the inspiration) that generate me a lot of patches or ableton instrument racks to use later.

Then, during writing sessions, I just load the patches/racks, tweak them and bang!

I may sometimes make a specific new patch that I think is what the song needs, but most of the times I load a patch I made before.

[Immerse]

the few times i have had pre-created patches/samples i really loved the result. i seem too make a song exponentially faster, and with better more flowing result. if i design mid song, it tends to sound forced until i spend mucho time tidying every little bit up

[Samuel_L_Damnson]

Im gonna be honest, when I make sounds I never seem to do the same thing twice. It seems like I just twiddle knobs for a while interpret that as you will.

[JTMMusicuk]

Im gonna be honest, I just twiddle my knob for a while

[Samuel_L_Damnson]

Im gonna be honest, I just twiddle my knob for a while

I try to get my 5 a day if you get me

[Immerse]

and with good results haha, i need to work on my twiddling

[Samuel_L_Damnson]

No homo
Edit: depends which team you play for haha

[didi]

Sound -> Filter -> Envelopes for filter -> tweak sound

[narcissus]

no but really, i just wait until a knob wants to be turned, and then i turn it however much it wants to be turned. sometimes i hold down control when i'm doing this.

this has to be done when the song is being written, or else the sounds won't fit right with each other

[Immerse]

no but really, i just wait until a knob wants to be turned, and then i turn it however much it wants to be turned. sometimes i hold down control when i'm doing this.

this has to be done when the song is being written, or else the sounds won't fit right with each other

as in you create your sounds on the spot? or tweak patches based on the song?

[cmgoodman1226]

What parameter do you find yourself tweaking first?:
It all depends. If I'm going for a certain sound, I may want to tweak an effect before I start using any oscillators. For an example of my point, One of the things that I use is Thor in reason. It has a really great shaper distortion, especially when set to soft or hard clip. When I'm making a synth using it's wavetable oscillators, I automatically turn the shaper on first because it gives the oscillator a nice edge to start with and really changes the sound. If I did not do this, the sounds I would start with would sound dull and boring to me, so I probably wouldn't use them; but when I turn that on first, it already starts to take the form of something else. But typically I start with the oscillators first. I'll usually just start with a saw wave, I'll carve out a bassline that I like, and THEN from there, I'll start to design the sound I want. I find it easier that way because it's much simpler for me to design the sound around the melody rather than the melody around the sound. It gives me an Idea of what type of sound I'm looking for as opposed to just fumbling in the dark for a random sound that I may like. BUT, then on other days (pure sound design days), I go into whatever synth I'm using with a goal, Either I want to make a pad, a stab, a lead, or a bassline. And then I just go to work.

According to ill.gates, it is best to keep left and right brain process separate, such as arrangement and sound design. If you design in the context of your song how do you get around this, or do you just disregard this advice?:
This is probably true; and it's probably why I'm really good at making loops, but I'm shit at completing songs. But I'm still more comfortable designing sounds around a set melody.

How long do you find yourself doing sound design sessions?:
It varies. Sometimes I wind up with a sound I like in 5 minutes, sometimes I spend hours tweaking something only to find the next day that it sounds god awful. This is part of the reason why I bounce out religiously and save new versions of every project so often. Maybe I made a sound that is decent, but then just spent the last hour ruining it with effects that I thought sounded great at the time. If I didn't resample, I'd pretty much be fucked and stuck with a shit sound, but since I did, now I have the older version or versions of that sound to go back to, and the methodology is saved in the effects chain if I ever want to go back to the original patch.

How long do you tend to spend on one patch?:
Once again, it all depends. Take for instance the second tune in my sig. The reese that plays througout doesn't sound all that complicated by any stretch, but I made it over the course of 2 days. I knew that was the sound I was looking for, but I had to audition a few different synths because I just didn't like the sound I was coming out with. I eventually landed on massive, and after a few hours of processing. So I don't really know that there's necessarily a right or wrong answer for how long it takes. Sometimes it takes me days, sometimes it takes me minutes; it all depends on the type of the sound I'm trying to achieve.

Do you bounce your patches to audio for organization? If so, how do overcome obnoxious things such as making the audio file endless, or preventing the sound from changing speed from note to note when using a sampler?:
Well, to asnwer the first question, I bounce things out A LOT. But I am also really well organized (had to learn this the hard way when I had separate projects with basses and leads named the same thing all in the same folder-> needless to say I scrapped those songs). If you look into the "my beats" folder each song will be organized in its own project folder. Furthermore, I separate all the elements in the song. I'll have a folder for "drums" and "lead 1" and "lead 2" and "bass 1" and bass 2, etc and "samples". And I also use the first few letters of the track name before every sound so that I don't have 30 projects with files named "snare 1".
In regards to the second part of the question, I know there are plenty of samples that can change pitch without changing the speed, but I'm not usually interested in that. I don't typically take on filter movements until AFTER it goes into a sampler, so it doesn't matter if it gets sped up or slowed down a bit. I usually just export 8 bars of one note (usually C, but sometimes I like the movement better on other notes), and that way I know if the speed changes it won't matter, because I still have 8 bars to work with.

Misc. Tips and Tricks to better synthesis workflow:
I would love to say I'm decent at it but honestly ill gates would probably call me a moron for the way I do things. I'm not saying it's the best way, but it's the way that I'm most comfortable with.

[narcissus]

no but really, i just wait until a knob wants to be turned, and then i turn it however much it wants to be turned. sometimes i hold down control when i'm doing this.

this has to be done when the song is being written, or else the sounds won't fit right with each other

as in you create your sounds on the spot? or tweak patches based on the song?

create 'em all on the spot... but they are already there to begin with. i've made so many damn sounds, i know what i like to hear... but creating them fresh each time gives them chance to evolve and take different directions. i very much like to use classic simple sounds in new ways, i get a lot of mileage out of saw and detuned saw, lots of lowpass filters (especially with downward envelope) lots of gritty claps and hats, and deep kicks. lots of my sounds have roots in old skool acid house. i instinctively know how these 'should' sound, which is a combination of how they've traditionally been used and how my mind wants to hear them. this is why knobs "want" to be turned at certain points. and then when the track has all it can take of the 'used' sounds, i sculpt something new to fit the precise combination of elements. for most atmospheric sounds, i bounce, so i can add lots of effects and automation w/ little CPU cost, and reverse, cut up, stretch. i like to have the simple classic sounds interacting with complex new sounds purely from my imagination.

[Immerse]

Ah i see, i will probably find myself beginning to recycle more sounds when i have a bank of sounds i really like
@goodman nice post thanks man