What's the current best practice workflow for rendering out Normal/Depth/AO passes in COPs 2.0 for quick and dirty compositing work? Is it possible to merge these passes into a .exr file? Unless I'm missing something I'm not really seeing any straightforward tutorials dedicated to this yet.

Normal and Depth yes, AO not that I know of unless you count rolling your own with a wrangle and intersect(). The ROP output node can write multiple channels to the same exr file.

jsmack
Normal and Depth yes, AO not that I know of unless you count rolling your own with a wrangle and intersect(). The ROP output node can write multiple channels to the same exr file.

Thanks! You wouldn't happen to have an image of example network, would you?

Im still a little confused about this, wondering if anyone can help. So if we want a normal pass, do we need to do a render with Karma? Or can we just set up a camera and use the ROP Image node in a SOPs context?

All the tutorials on COPs 2.0 demo a UV based workflow, we wouldn't need to do this for the normal pass though right?

I'm not sure what you're after, but here's a simple demo showing render pass generation using COPs.

There's currently a bug that the ROP tags the image planes as Linear Rec709, causing the color to be interpreted incorrectly when viewing in MPlay, however, the color data in the file is correct.