Node/Sequencer workflow question

Support Compositing and Post Processing
#1

I’m using blender to colour correct / grade and edit a short film. I have about 30 minutes of live action footage that will be cut down to about 9 minutes using the sequencer.

It all needs to be graded, about half with one node setup and half with another node setup.

So I have raw footage to be both edited and coloured, what’s the best workflow for this?

Ideally I’d like to edit the whole raw footage, then route that through composite nodes before rendering the final cut with sound.

Is this possible or do I need to edit>render then node setup>render, then re-render again in the sequencer to add audio?

Here’s what I’m trying to acheive, the pre-rendered cut run through nodes for the final render:
http://img214.imageshack.us/img214/7459/compositeexampmu0.jpg
Any help greatly appreciated.

#2

Well, you can’t use the sequencer output to feed into a node. You could color correct the footage with nodes and use the output to feed into the sequencer, but it seems like that would be pretty slow and cumbersome.
If the built-in color correction in the sequencer isn’t good enough, you will probably have to use some intermediate renders.

#3

render workflow possibilities http://wiki.blender.org/index.php/Manual/Rendering#Render_Workbench_Integration

#4

Thanks for the replies fellas. For now I’ll just render the final cut, then re-render through nodes and then sequencer again. Hopefully in the future the nodes can be run in the same way as a plugin through the sequencer, so each frame only has to be rendered once instead of 2 or three times.

#5

The next big open source blender project is slated to be an experiment in VFX. I would imagine that during the development for that project, the ability to use the 3D renderer, compositor, and sequencer back and forth internally in whatever workflow you like, but for now you’re limited to 3D render -> composite nodes -> sequencer. So you can do it all in one file (although you would have to render 2 separate scenes and make sure the second is offset after the first) but it would kinda be a pain in the butt.

#6

Well, eventualy I think what will happen is that you will be able to put node strips of any size into the VSE, so the final render is everything + node setup.

So instead of your workflow being edit > render > node composite > render > final edit for sound > render, it will simply be vse > render final. It would save a huge amount of time :slight_smile: