Keeping the Creative Flow Alive (Patreon)
Content
Guys, I’ve seen that the whole AI topic is pretty divisive… and I have to admit that I’ve just run into another problem that I honestly didn’t expect.
When I’m creating a scene in VaM, my brain is completely in creative mode. The story is playing out in my head while I work, so I naturally position the characters, assets, cameras, lighting, etc. according to what I’m imagining at that exact moment.
But as soon as I switch to AI, my brain suddenly has to go into technical mode.
The creativity stops, and now I’m thinking about prompts, seeds, settings, iterations, and trying to force the AI to give me the result I already had in mind.
Then, once I finally get the image I want, I have to go back into VaM and somehow instantly switch my brain back into creative mode to continue the story.
And I’m realizing that my brain really doesn’t like that constant switching. 😂
AI images can be great for illustrations or even for creating a series of images around a specific concept. But for the way I create stories, I’m starting to find it counterproductive.
The problem is that I can’t simply move on to the next shot until I’ve managed to get the AI result I want. Before, I would immediately adapt my ideas to VaM’s limitations, so the story could continue flowing naturally in my head.
So I’m going with a third solution.
It’s actually something I didn’t really want to do because it’s going to require a lot more work, especially considering that I already produce around 14 episodes per month… 😭
Instead of using AI to add extra characters, I’ll create multiple renders of the exact same scene and camera angle in VaM, but with different characters placed in different positions.
For example, I can put two characters near a wall and render the image. Then place two other characters near the door and render again. Then another group somewhere else, and so on.
After that, I’ll bring all those renders into Photoshop and use layer masks to combine them into one final image.
That way, I should be able to create scenes with 10 characters or even more, while keeping everything entirely within VaM.
This is going to be especially useful for the upcoming AMD episode with the White Angels and their emergency meeting. That scene needs a lot of characters, and normally VaM would start begging for its life long before I could display all of them at once. 😂
I’ll show you a quick test I made.
The first image "01" is yesterday’s AI version.
The second image "02" is the version I just created using only VaM renders combined with Photoshop layer masks.
So I think this is the method I’m going to use from now on. It requires more manual work, but I stay in control of everything, I keep the original VaM look, and most importantly, I can stay in creative mode without constantly interrupting the flow of the story to fight with prompts, seeds and AI generations.
This way, everybody wins, and I get to keep my creativity intact.