I have decided to continue my more reflective pieces in video form as it’s been on my bucket list for ages! Hope you enjoy.
First one goes into the core philosophy I’ve written about before, & a lot more. Let me know your thoughts and share the video with someone you think might find this interesting.
I added the transcript below for those who prefer reading.
Until next time.
Transcript:
I spent 5 years trying to turn my drawings into films in a way that feels like directing in real life.
If that sounds exciting, you’re in the right place.
The struggle begins:
I started drawing in military service about seven years ago. We weren’t allowed to use our phones, so inspired by my friend, I started drawing in this little sketchbook.
Our superiors thought we were taking notes, but in reality, I was drawing whatever came to mind.
The sketchbook became my way of escaping the pain of mandatory military service. But because I always loved filmmaking, at some point I started to wonder: could my drawings become films?
After finishing my service, I came to London to study, and I started working as a freelance filmmaker besides Uni. Music videos, some fashion stuff, random side projects.
Besides Uni and freelancing, I was drawing more and more, and eventually decided to leave traditional filmmaking behind, and really try to turn my drawings into films.
I experimented with traditional 2D animation, and if you’ve ever drawn 24 frames of the same character, just for one second of animation, you know the pain. It’s not for me.
Then I tried stylised 3D animation, which honestly was even worse. It was too technical for me, and it lacked a sense of flow.
I was used to having a real camera, exploring a scene, angles, shots, before ever hitting record. I wanted to draw films in a more organic and surprising way, almost like having real actors and directing with a sense of unpredictability about what the actual performance is really going to give. I was seeking freedom, the current pipeline’s just couldn’t give me.
When I discovered Dall-E, just before it’s release to the public in 2021, the potential of generative AI completely hijacked my life.
It was simple. If you could generate an image, you could generate 24. And if you put those 24 images into a sequence, you have a film.
However, instead of treating generative AI like a magic button that takes away all the work, I went deep to really study the tech and find a way to make it work in a way that made sense to me. I even put in an extra year at uni just to study machine learning.
What I found was both depressing, but also exciting. On their own, generative tools are basically slop machines. Without human control and input, they just create garbage.
But I became convinced there’s a way to use these tools responsibly, not by stealing other people’s art, but by inputting your own style and keeping full control. There is genuinely so much potential. It just takes a shit ton of work.
The journey was brutal, confusing, and full of late night sessions and dead ends. But after literally half a decade of trial and error, I finally reached something I could actually call a minimum viable product.
A new philosophy:
Here’s where it gets interesting. I didn’t just approach this way of working from a purely technical point of view. I wanted to build a philosophy around it.
There are two art movements that really shaped my experimental filmmaking journey early on and allowed me to make sense of everything:
Dada & Surrealism.
Surrealism taught me how to tap into the subconscious. I’d sit down with my sketchbook and draw without judgment. Animals, people, trees, cars, fragments without cohesion. Those fragments represented raw experience and emotions. They formed the human foundation of my work.
Dada, Surrealism’s daddy born in 1916 in Zurich, not too far from where I grew up, came about in the middle of World War I when Europe was in ruins. Artists like Hugo Ball, Tristan Zara, and Marcel Duchamp rejected order completely. They used absurdity, collage, found objects and nonsense poetry to rebel against the very systems that led to devastation.
Dada connected with me on a lot of levels. I was patching together drawings, almost like collages, to make these absurd films while feeling so left field with all the coding, dataset trainings, and just trying to get a hang of all these tools.
For me, Surrealism gave me the input data, the subconscious sketches. Dada gave me the method, permission to break rules and basically rebuild filmmaking from scratch. Together, they formed a flawed but somewhat working method to make simple films.
More issues:
Here I was taking my subconscious fragments and turning them into motion, hoping to make great and authentic films. But I quickly realised something.
Randomness as input without refinement just leads to randomness in the output. If my goal was to create hundreds of random clips. I could have very well proceeded. But I want my films to eventually tell coherent stories. So I had to take it further.
The breakthrough:
This is pretty much where I was at at the end of this summer when things finally shifted. Google recently released a tool called Nano Banana which is basically if ChatGPT and Photoshop had a baby. The release led to a much needed breakthrough and everything started making sense.
Tools like Nano Banana let me sample my drawing. That means I can change poses, angles, and even do camera moves all while keeping the style consistent. That is a huge degree of freedom added to my workflow.
The raw sketches still proved to be problematic though as the lack of intent and design carried into the final films leading to semi-great results. I tried sampling keyframes using Nano Banana by creating an asset library of sketchbook drawings but as they were sketches the films looked sketchy too. That’s where I understood that refinement has to happen at the drawing level before ever thinking about films.
And that’s pretty much where I’m at right now making actual illustrations and not just sketchbook drawings you really make films from.
So for now, here’s the workflow simplified:
01 - Sketchbook:
I connect with the subconscious, draw freely and ideate from within.
02 - Illustration:
I refine the sketchbook drawings into actual illustrations with considered composition, colour, and most importantly, a narrative in mind.
03 - Sampling:
I use tools like Nano Banana to sample the refined illustrations into character sheets, poses, environments, camera angles, basically all the keyframes I need to then actually make a film.
04 - Film:
And then finally the filmmaking stage. At the moment, I’m mostly using image to video tools, but I have a strong feeling that might change into the future. This stage then also includes putting all the clips together on the timeline, colour grading, editing, sound design, and basically everything needed to fully finish a film.
I call this whole process: Metadada.
Metadada is subconscious data refined into moving images.
It’s a messy, but evolving system that is finally somewhat working. And that’s why I decided to start making videos. I’ve reached a point where I finally feel comfortable sharing the process and refining it publicly. And honestly, I’m really, really excited.
From here, I’m planning to dive deeper into everything:
how I draw,
how I actually craft illustrations with narrative in mind,
how I use all these generative AI tools to sample keyframes and animate and all that,
and then how I actually bring it all together to make films.
I know this was a lot, but if you’re curious to see where this goes, subscribe and follow me on this journey.
Until next time.