Well, I usually do a basic setup on my modelling app and once done I export 2 main obj files and some extra ones depending on need. The first two would be the main mesh model, and the alpha model mesh. The alpha model mesh is the same than the main, except I change the material assigment to all black diffuse for all opaque materials, black diffuse with .5 opacity for glass materials, and black diffuse with opacity maps for clip textures as leaves and such.
With those two I open octane and then import the main mesh, I do a basic light and camera setup and use a low pixel resolution to work around and start to tune up the materials, here when the scene is pretty complex and too hard to move around (or harder than what octane has us used to) I do an extra obj export for certain objects which require more fine tuning on the materials. I import it in the same scene this extra obj mesh node, and link the material node to just this object, the scene mood (camera, light, etc) is already done, so I can more easily navigate and work on it with less polys laying around.
I repeat the above steps as needed, and when everything is to my liking I add another mesh node, the alpha one, sometimes it needs some tuning, but usually it ready to go. Then I start saving a octane scene for each camera I want, in really really heavy scenes I use a "dummy" mesh node just to speed up this process. Then I close Octane and create a command batch file to render to the desire max samples and res, one for each node and camera, the "real" one and the alpha one, then I go to sleep and wake up to find all the neat images and a fresh coffee, ain't life beautiful?
Like I said, pretty easy workflow, but if you have any improvements or just plain different ones, I'll love to hear it, and hopefully we can all learn a little bit more.
Cheers
Kubo
PS: I know this would be neater with graphs and pictures, but I'm on a tight schedule lately so I would apreciate if someone has the time to add those up.
