What inspires instrument/effects coders....?

DSP, Plugin and Host development discussion.
Post Reply New Topic
RELATED
PRODUCTS

Post

Just a quick question to the instrument and effect coders out there - where do you get your ideas, and how do you turn them into practice ?

For instance, if you are coming up with a new effect, do you have an idea of what it is likely to sound like before you write it, based on expert knowledge of synthesis, or do you go, well, I guess a 10 band EQ with built in flanging and arpeggiation might be interesting and then go off and build it. Do people use something like Synthedit to prototype things to see if they like the sounds, and then do it in C++, or do people have their own toolkits of bits and pieces that they glue together ? A lot of the VST's I have seen out there are not really trivial in terms of what the code must be like, so I am curious to find out more about the actual PROCESS of development, rather than the technical details of how it is done (although that is obviously interesting too !).

Hope someone feels like explaining how they go about it. For the record I am still just paddling in the very very shallow end, so I don't have any tales of my own.

Cheers,

Post

It varies, naturally.

For bread-and-butter effects, there's a lot of literature out there about how to implement them, so if you're bothering to create yet another one, it's usually because you want to make some unique variation or integrate it with something else or add some key user interface convenience or something.

Sometimes you're trying to emulate (or one-up) a piece of hardware or model something similar. Throbbing Gristle's "Gristleizer" inspired my "Throbber" effect, but the components I had "laying around" in code allowed me to provide more waveshapes and more filter modes than the original. (I don't prototype in Synthedit because I haven't invested the time to get comfortable in Synthedit. After I finish the Tiny FX suite, I'll be able to do a lot of prototyping in Audiomulch, though.)

User suggestions wind up having an enormous impact on my development. Even when I don't incorporate them as-is, I keep them in mind and let them mingle with other ideas. I spend a lot of time reading about synthesis and effects, reading the same things over and over in fact, trusting that eventually two ideas will collide in my head and yield something new and interesting. The original idea for Meridian was "how few knobs can I get away with," and on that basis it was destined to be a total failure, because everyone was talking about Vanguard and someone said "crapgates", and I knew I had to put crapgates in.

This may sound a little strange, but I don't give all that much thought up front to how things are going to sound. There's no way I can think about the range of sounds a synth with 50 knobs can make. There's too many of them. This is part of the reason that I'm firmly committed to providing randomly generated presets and/or a way to randomize parameters in my synths.
Image
Don't do it my way.

Post

I hear strange sound oddities sometimes (like a radio clipping out, or something like that), and I'll try to make code to create that noise artificially. Other times, I'll work based off of an image of noises that I have in my head, but actualizing this in code is fairly difficult, as is also true with music composition itself. :)

As far as the actual development process goes, I carry a notepad and pen with me wherever I go. I often find myself in very boring situations where I'm unable to code (ie, driving somewhere, being at work, etc.), so I usually develop the algorithms in my head and write them down later for coding. Same goes with debugging; I keep a mental "list" of all the problems I need to fix, and when I'm unable to physically work on them, I'll just think of things that I can do to discover the cause of the problem and fix it.

Post

phutboyslim - I heard the effect on a fatboyslim record and thought - that's probably quite difficult, but if I wrote a plugin......

scuzzphut5/6/6.5-lite - I loved to use midi gates to make rhythmic chops. At the time there were no gates with 808-style step sequencers. Putting a stepsequencer in seemed like a nice way to get rid of the extra chore of routing a midi track to a midi gate.

Post

I make it up as I go along.
Stefan H Singer
Musician, coder and co-founder of We made you look Web agency

Post

sqook wrote:As far as the actual development process goes, I carry a notepad and pen with me wherever I go...
Me too. Sometimes I sketch block diagrams, UI layouts (while playing "what's the minimum number of controls I can use for this?"), waveshaper transfer functions, etc. Or I write down random ideas, e.g. "LFO modulates FM depth?" and do the experiments later.
Same goes with debugging; I keep a mental "list" of all the problems I need to fix...
I have a text file for random notes in every Visual Studio project - I keep problem lists, to-do lists, and user-suggestion lists in it. I tend to misplace my notepads or not always carry one with me - I probably have fifty of them, each 1/4 full, scattered about the house - but only one laptop, which I generally don't misplace, so it's a better place for centralized notes :)

When I have a big task that's blocking me (like, say, implementing cross-plugin modulation in Tiny FX (cough), it's nice to have a list of small things to work on. I can sit down, say "well here, let me finally figure out what's causing this denormal bug and then if I'm not getting into it I'll stop and watch a DVD with the missus, but maybe I'll get momentum going and just spend the afternoon cranking stuff out." (Nice theory anyway...)
Image
Don't do it my way.

Post

can we check some of your products please ?
stefancrs wrote:I make it up as I go along.

Post

topaz wrote:can we check some of your products please ?
stefancrs wrote:I make it up as I go along.
Nothing released really, except what you can find in my sig. Will soon release a synth made in synthmaker, but stuff has been hindering me (both on my and on SM's side), so hold on a wee while longer.
Stefan H Singer
Musician, coder and co-founder of We made you look Web agency

Post

I couldnt find any plugins in your sig.

I cant read sweedish tho :hihi:
stefancrs wrote:Nothing released really, except what you can find in my sig. Will soon release a synth made in synthmaker, but stuff has been hindering me (both on my and on SM's side), so hold on a wee while longer.

Post

cash money,

hoes,

and gold-plated teeth.

those are my inspirations to make it big in the plugin world. :D

(i joke, i joke... i don't know why i subject myself to this miserable lifestyle.)

== chunk

Post

topaz wrote:I couldnt find any plugins in your sig.

I cant read sweedish tho :hihi:
stefancrs wrote:Nothing released really, except what you can find in my sig. Will soon release a synth made in synthmaker, but stuff has been hindering me (both on my and on SM's side), so hold on a wee while longer.
There's no plugins there. But some realtime graphics+synthesis going on, made by me.
Stefan H Singer
Musician, coder and co-founder of We made you look Web agency

Post

Hmm, I try to visualise waveforms in my head, trying to bend, stretch, mix etc...them in interesting ways. Or to generate interesting spectrums.

Post

wolf_of_badenoch wrote: where do you get your ideas
From what I need. I needed an Hammond organ, I did it. I needed a Rhodes, I did it. I needed a Wurlitzer, I did it. Now I need money... ehem... :)
For instance, if you are coming up with a new effect, do you have an idea of what it is likely to sound like before you write it
I just did "conventional" effects and instruments, imitation of existing things, nothing new. So I knew how they would have sounded like.

Post

Thanks all for taking the time to share. While I splash around in the shallows and look up at the lofty heights of what others have done at least I feel that I now have some idea of the path ahead. Happy coding :)

Post Reply

Return to “DSP and Plugin Development”