The problem
I needed a simple promo animation of my company logo.
I can’t draw to save my life. I’ve tried, I’ve practiced, and I’m still kind of shit. So when it came time to animate the logo, I had two options: get better at art or make it move in code. Given that I don’t plan on taking up any new hobbies, Manim was the obvious choice.
I also wasn’t interested in image-to-video AI. I wanted something I could control and actually own, and AI video still falls apart the second you need precise motion.
Manim over matter
This was my first time giving Manim a proper go. I had played around with it before but never actually tried to finish something.
First I cleaned up the SVG in Inkscape, simplified it, and made it grayscale.
Then I brought it into the IDE and started animating.
That’s where I got stuck.
The biggest problem was making the outline smooth. First I thought the SVG was bad. Then, when that didn’t fix it, I moved on to debugging the code.
I lost an embarrassing amount of time to that one problem.
It starts working after I throw it away
After hours of banging my head against a brick wall, I reset. I went back to the SVG and rebuilt it from scratch.
That was what I needed. I still do not know what caused the original spikes, but starting over removed them. Hours of debugging vanished because I rebuilt the file in ten minutes. Annoying, but effective.
Now that I had a functioning animation of the logo, it was time to make it look good. The glow and frosted glass effects were the fun part. I wanted the logo to look like it was sitting on a frosted glass card and glowing like a neon sign.
Making the glow effect and frosted card effect was much more intuitive, especially compared to fixing the animation. I knocked out both in under an hour. Getting the glow to interact with the card took a bit, but it sells the effect much better.
The fun part
Once the animation was good enough, I moved to editing and Foley. I know audio well and video barely at all, so I handled the edit first. I tried CapCut because the browser version was free.
This was surprisingly good. CapCut felt natural and not too obtuse or in the way. While I noticed it lacks some tools I would have liked, you can probably get around that with enough skill.
With the video blocked out, it was time for audio, my favorite part. I’ve been into music production since I was young, so I was already comfortable in Ableton Live. For the sound effects, I went for a stylized vintage CRT TV turning on and then off.
After the visual edit in CapCut and the sound work in Ableton, I stitched the final version together.
What I learned
- Manim is powerful but not as straightforward as I expected, and SVGs are annoying.
- The rise of CapCut makes sense after trying it. I honestly think a lot of other software could learn from the simple design language and UX.
- Using Python for motion and timing instead of logic changed how I think about it. It felt less like wiring logic and more like arranging frames.
Finished video
You can watch the version with audio here: Watch the Manim animation on Facebook
See It In Action
Manim is still annoying. It is also the first animation tool I have used where code felt like a useful constraint instead of a workaround.