Multilayer output

Support Compositing and Post Processing
#1

Just a quick question:

Let’s say I have a cube on layer one and a second cube on layer 2.
I set up 2 render layers, each rendering just one layer.

Is there a file format that will save both of these images in one, multilayer file?

I know there’s a “multilayer” option in the format tab, but I think that that is just saving the various channel information as layers, not two seperate image layers. Am I correct about this? I have no app that can correctly view a multilayer exr file. If anyone can point one out for me, that’d be great.

Basically what I want here is one image file that would have two image layers. One layer would have one cube, the second layer would have the other cube.

If Blender does not support this, I can probably work around it by doing multiple render passes and saving the image sequences in seperate folders for each scene, but It would be a LOT easier to just open one image file for each frame and not a dozen or more. I supposed I could write some code that would properly import all of these files as multi-layer frame sequences or something. Again, would be easier to just import a single image sequence.

Any thoughts?

#2

I don’t think you can render two 3D layers as 2 TIFF layers (or whatever format) as a batch, but with nodes you do have some control over the different parts coming out of the scene. You could have 2 renderlayers, each defining the rendering of 1 3D layer, and then outputting that as 2 files (one for each layer). Still, you would then have to combine them in photoshop or GIMP.

#3

That might be a good way for me to at least get all of the rendering done in one pass. I’ll have to make sure it will work with image sequences, but that should work.

I’m actually writing my own tool to do the re-compositing. Goofy as it sounds, I’m actually making a pixel-art sort of animation. Using Blender as a sort of ‘first pass’ for the animation. I plan to clean it up and add more subtle animation in my 2D tool. Things like the facial animation and whatever effects I need to do that don’t look that great or would be overly difficult to do in Blender. I can save a lot of rigging and animating time this way, and end up with better results and have more control over the details and stuff.

That’s a good idea, tho, I’ll try it out. If they get python nodes running I’ll probably be able to make a node that supports file output with multiple image layers, make it export gimp .xcf files or something and then import those in my little tool, or maybe I can use something like Irfanview to quickly batch-process the frame layers into multi-layer files or something…