BUTTERFLY IN THE DARKNESS

My workflow bottleneck? Endless re-renders due to minor tweaks (backgrounds, focus, lighting). Compositing solved this by letting me edit render passes (depth - for focus tweaking, lighting) in post (using software like Nuke or Fusion) — zero re-renders. This project stress-tests that efficiency. Final result:

Before diving into the main part, a small side note: the character model was based on my friend. I made a scan of him and used it as the foundation for creating the final, animation-ready head in ZBrush:

Now, moving on to Nuke. Final comp script overview:

I started by Blender's shader rebuilding, killing the red light, then graded the base image. Next: dust for atmosphere, focus tweaks via depth pass, and stylistic touches — chromatic aberration, motion blur, handheld camera jitter, and a subtle vignette to frame it all

Chromatic aberration was added by splitting RGB channel into Red, Blue and Green channels, moving them a bit and then merging them back together. The effect is subtle but essential

The main disadvantage of the workflow is the size of files containing all the information needed (exr files). Typicaly you render out animation from, lets say, Blender as a jpeg/png image sequence and the exr variant of the same render will be ~10 times heavier (like in my case). Storage ↔ Time. The eternal trade-off.