This college student 3D-printed a video camera that runs on 35mm film

There’s something special about film, and for a time in my life I was absolutely obsessed with it. But I don’t think I ever had the dedication, determination or drive for film the way college student Joshua Bird did. Because he, well, he went out and 3D printed his own film video camera.

Why? Well, the idea started because he said it was too expensive. According to his blog post which details the entire process of this creative contraption, it would cost about USD250 (~RM1,108) and he’d have to “wait months” for just 10 minutes of footage. Obviously, this was unacceptable so the only workaround was to build his own motion picture film camera that ran on regular 35mm film that you’d use in 35mm film photography cameras.

Immediately he ran into a problem. These 35mm film cannisters typically only store at most 36 shots per roll, which meant that if you converted that into video, you’d get about 3 seconds of footage per roll. Not a terrible problem to have since most people’s attention spans these days are about that long, but that just wasn’t going to be long enough for Joshua.

To solve this, he decided to reduce the image size by a factor of 8, which meant each shot in the 36 shot roll of film would be 8 shots in his film camera. This extended the limit dramatically because now he could record 16 whole seconds of footage with a single roll of film. That’s longer than a standard Instagram story.

Now, I won’t go into the nitty gritty details of how he did it. But the summary was that he 3D-printed the chassis, gears and mechanisms necessary to operate the camera, like the shutter and winding mechanism. Did some magic with a bunch of electrical stuff and coding that I won’t pretend to understand, and boom, he had a “working” camera.

It worked so well, in fact, that he could even control the lens’ aperture and even the optical image stabilisation with his shed-built camera. This is no small feat because the lens he was using was your standard EF 18-35mm kit lens from Canon (he’s trying to save money, it was the cheapest lens he could get), and apparently there were almost no resources online for the lens protocol.

What he found was a 100-page PDF file that was written by an author who Joshua describes must have been “on the exact same wavelength” as himself because he enjoyed nerding out on the same kind of intricate technical details of photography as him. I absolutely love this part of the story by the way, but I haven’t even revealed the big hurdle—the document was in French.

That meant Joshua (who I guess doesn’t speak French) had to rely on Google Translate to make sense of that 100-page document. Now, I’ve never used Google Translate to do something like this before, but the last French-English translation I saw made zero sense. Still, that didn’t stop him because he got it to work, and the rest is history.

Except, it still wasn’t. Now that his camera was working, he could go out and shoot these 16-second clips of his memories, friends and spaces. But now came the tedious part—scanning everything into digital so he could edit and share this video. To do that, he set up a DSLR scanning setup (which meant he had to go out and buy his first digital camera, RIP saving money), but because the frames were so tiny, he had to use a 50mm macro lens with a 50mm extension tube to scan the frames before inverting the negatives.

Then, he had to write an OpenCV Python script to align all the frames so that they wouldn’t “be a jumpy mess”, and finally stitch all the frames together.

I honestly love everything about this story, and I would wholeheartedly recommend that you go read his full blog post for the entire story. It’s such a cool showcase of what sheer determination fueled by the refusal to pay exorbitant fees to the capitalist machine could achieve. Is there no way? Well, we will BUILD A WAY. It’s awesome.

But what was the result of the project? I mean, you’ve seen the resulting film in the beginning, but what about the main problem statement that led down this complicated and wonderful path? How much was it to shoot film with his 3D-printed camera?

After doing the math, Joshua found that using his camera would cost him about USD600 (~RM2,660) for 10 minutes of footage. That was over double what it would cost him to use motion picture film. And that was before accounting for all the reliability issues he faced using the camera—issues that were so severe that it was “unbearably stressful” for him to use it on trips.

Look, it’s the spirit that counts. It’s the fact that he stared straight into the eyes of Capitalism and said, not today. And for that he will have my eternal respect.

[VIA]