[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[sc-dev] graphics synth

On Wednesday, June 5, 2002, at 04:32 PM, ccos wrote:

it would be great if the synth engine itself could serve up graphics as well as sound.

as mentioned, a separate graphics synth program running parallel to the audio synth server is probably
the wisest.

i have daydreams of using the sc language to do sound and visuals which are intimately connected, not just in terms of managing and manipulation of events with the language, but also doing visual dsp using the servers

when wanting to take an audio signal and directly transform into/with visual synth, i suppose it might be good to have a unified audio/video server. that might be worthy of a separate version.

on the other hand inter-application communication isn't that hard at all. and you have better stability when pushing
the limits (which you are going to do).

in most cases, the audio server can be made to osc message to the visual server. think about how much info iTunes takes from the audio: its not much, and could easily be sent over osc in real time. even down to australia.

i suggest exploring osc messaging until it breaks under the strain. on the same machine, its going to be pretty fast. if you want to fft something and animate every signal, that might be too much.

just to conjure up an example of what i'm thinking of,
wouldn't it be fantastic to be able to do things like algorithmically generating 2/3D shapes with pattern classes, and have the positions of all vertices be controlled by LFOs, whilst having the same LFOs
used simultaneously as control sources for sound synthesis?

hey and what about smells ?

why am i mentioning this in this thread? well, i originally thought that with just a few basic graphics rendering primitives people could just build a gui system if they wanted to, but would there be a way for the user to interact with the graphics and more importantly, for the server to pass that info to the interpreter? no idea, maybe someone could enlighten me. in any case, i always thought that if sc should do graphics it should do them the same way it does sound. we should have a ugen count and cpu meter indicating what is going on with visual synthesis. if this means that graphics can't be used as gui material, so be it, after all we aren't building guis out of sound either:->

also keep in mind that interfaces can get quite complex on the sclang side, including 3d, openGL (strictly for the interfaces, not trying
to do vid-synthesis).

JIT i guess does support building interfaces of large graphic complexity. i was talking to randy from cycling74 about his JIT based interfaces.

JIT is quite nice ! and can be used with scserver, but on different machines (its not osx).

any thoughts?