2008-06-04 1 Comment
What if a NLE/sequencer would use tracks only in the sense of having rows to place stuff on, tying all routing and parameters to objects in the timeline?
You would have a lot of flexibility. It would become easy to apply effects to specific regions. You could just import sessions and chain them in a master session. But if being able to set everything on a per object basis becomes having to do so, things become cumbersome.
There could be containers for wrapping objects. Objects could inherit settings from their containers.
For an NLE, audio steams, video streams, automation data, processing nodes (plugins) and connections could all be shown as objects on the timeline. Connections as separate objects means they wouldn’t be tied to the lifespan of source or destination.
A very basic example of how this could look like:
The first container defines a time reference, position and transport state (paused or playing, relative speed). One case where you might want to use 2 transport scopes would be video within video, being able to change the speed and playing direction of the embedded video (here I assume each transport scope is fully automatable).
Since the horizontal axis is taken for time, nesting must be expressed vertically. Think of Lisp-like parenthesis 😉
The slider after Group is just to show that controls could be placed on container headers.
A concept for showing detailed info and options for all objects that cross a location (could be the playhead or a marker):
The basic idea is actually years old, but I had to think of this again because of the Lumiera project.