POV: caught junkerqueen and zarya in the gym shower :o
LINK
Credits
Zarya
BordeauxBlackVA
Junker Queen
BordeauxBlackVA
Audio Engineer
LewdSoundV2
Models
doublenyl
-----------------------
This animation went smoother than most of my previous projects. I didn’t run into any major hiccups aside from a few rendering mishaps—which is pretty normal for me. If everything had gone perfectly, the whole project would’ve taken about 14 hours to render. With the extra re-rendering for volumetrics and foreground props, it ended up being closer to 15 hours, give or take.
Using render layers really saved me here. Without them, I would’ve had to re-render the entire project from scratch. I also experimented with a new trick I’m still testing: rendering characters and key elements at a high sample count, while cutting samples for foreground and background elements that will be heavily blurred by depth of field. In some cases, I can cut those samples in half—or more—without losing quality. The amount I can reduce depends directly on how much the layer is affected by DOF: the stronger the blur, the fewer samples I need.
Animating this project wasn’t too bad overall. The hardest part was nailing the base motion for the characters. I wanted it to feel like Zarya was actually being held by Junker Queen, not just moving around freely on her own. That meant doing multiple passes, scrapping what didn’t work, and combining the parts I liked into the final motion. In the end, I’m happy with how it turned out. It’s definitely something I’ll carry forward into future animations.
I also learned a lot about physics in this project, especially when it comes to jiggle effects. In the past, I mostly relied on tools like wiggle bones and Simplicage, but this time I used cage physics almost exclusively. The only exception was the characters’ hair, since bone-based jiggling works perfectly fine for that.
Not much to say here—just that all my Houdini practice is finally starting to pay off.
For this project, I tried a new voice workflow that turned out to be extremely useful. I’ve always wondered how I could animate characters’ dialogue without having to wait on voice actors. The solution came with an addon called Push To Talk. It lets me record my own voice directly inside Blender.
Here’s how it works: I animate to my own temporary voiceover, send that track to a VA, they record over it, and then I just swap my voice out for theirs. It gives me complete control while animating, keeps things moving forward, and makes for a much smoother workflow overall.
That’s about it! It was definitely an eventful week, but overall I’m really happy with how this project turned out.
Skye
2025-08-31 06:36:43 +0000 UTCItsKawaiiNoon
2025-08-28 14:59:47 +0000 UTC