Video editing/processing in supercollider

Hello all,

I want to do some video editing to accompany some of the music I am creating in SuperCollider. For the video editing I intend to do some of the following:

  • Have multiple layers of video playing different section lengths of the video
  • Have the videos change their levels of transparency dynamically
  • Have video layers be created and ended in real time
  • Calculate multiple video editing lengths algorithmically

Now to the best of my understanding this is not possible with SC, which is a shame, but I fully understand why SC has kept its focus on music.

So the next thing that sprung to mind was Jitter in Max for Live. I opened it, played around a while, and though it’s an elegant and slick piece of software I realised at this point how much more I prefer the way I work with SC now; I am beginning to think with the greater flexibility that I sense I have with SC, rather than thinking about patch cords, bangs, and often repetitious copy and paste editing – as much as I have a huge respect for that way of working, and the excellent software that it is.

So my question is in two parts: is there any way to editing video in SC (I guess not)? And if not, is there an equivalent video editing software like SC where communication between the two programs isn’t too tricky to cultivate? Or put another way, what do people do here when they want to add video to their projects?

Thank you in advance for any responses to this question :slight_smile:

I don’t think the program you use will matter too much… But to use OSC to communicate between the two is pretty easy. I actually use unity for all my visuals - processing might suit your needs better.

The important problem with OSC though is that it encourages a distributed state. For example, you might end up with separate timelines running in different applications, this is incredibly difficult to manage at any kind of scale. Instead, make all the OSC controls for the sound and the video be unified in the same program so that you can easily see what happens together. Supercollider is pretty good for this! As an example you might have a Task that sets off patterns and sends OSC. Or you could do sound and video both inside of patterns. Or, use buses for each osc message, continuously sample them and write all of the pieces logic on the server (my preference).
The aim is to write code that makes it easy to see what events happen together, and to completely describe your piece that way. This means the audio and visuals should do absolutely nothing without the OSC sequencer running.

I don’t know if Scintillator is abandonned or on hold ?

1 Like

Scintillator is great - the author does post here @lnihlen

It’s on the back burner right now while I focus on Hadron. I’ve been meaning to come back to it. If someone had specific feature requests that might be interesting, please file feature requests/bugs on the github (or better yet, PRs!)

1 Like

Thank you to for all the helpful responses to this thread.

Jordan, that’s a very helpful overview on how to achieve a good working practice with OSC. I will definitely keep that in mind, and think about coordinating the messages.

Scintillator (great name) sounds fascinating. I will take a look into it. It would be great if everything could be achieved in SC.

Thanks everyone for getting me off to a good start with this!