Blog Post #22 A New Approach
W. 4/2/2025
I’ve been looking into what Eli recommended after my last Milestone presentation— about how to fix the error that I made in working on alla prima, and about some options for effectively developing the mechanic. He recommended:
Computer Vision. A field of AI that allows computers to “see” images and analyze them. This could hypothetically work, but it would require a lot of work, and skills that I don’t have. It’s also not directly what alla prima is; it would have to be implemented to create the mechanic from other bases.
Image-to-Mesh/Geometry. This is also AI, and is typically used for stuff that I would never do— create 3D assets from 2D images. I’d be fine with using AI to achieve alla prima, because it’s not creating assets— it’s just altering the structure of pre-existing ones. There are websites and programs that achieve this, but for Unreal Engine 5, the best option is the Geometry Tools plugin. (One program that does this outside of UE5 is Rodin, which I learned about from this video.) I skipped around a bit in this video from Unreal about Geometry Tools. It was from 2022, so I’m certain that there’s more advanced, more recent, aspects that could help me more. At 1:20:50, they demonstrated “Displace from Texture”— which could, potentially, work, if I could generate a unique texture based on the refraction weapon trail each time that it’s used. The example didn’t have color, though, which is key to alla prima.
Vertex Shader Distorts Geometry.
F. 4/4/2025
I’ve been researching Vertex Shaders. This video was the first one that I watched. It had only a brief introduction to vertex shaders, but it seems like this might be the way to go for alla prima. This next video, also by Ben Cloward, is about vertex displacement. In Unreal, vertex displacement can be used with a texture map to displace vertices relative to the texture. This is all in Materials, so I’m unsure how to make a function in Blueprints that would displace vertices based on the weapon trail. What I do know is that the weapon trail uses a texture map, as well, so it seems reasonable that I’d be able to utilize the same brush texture as the displacement map.
I watched this video on, apparently, all of the ways to use Displacement in UE5. It was a long-ish video, and was completely unhelpful.
I reached out to some people for help. The issue is that this vertex displacement/vertex painting method is completely new to me, so I don’t even know what to search when looking for help online. So I reached out to Ulm, Shivansh, and Griff. Hopefully they will be able to give me a kickstart. Griff already responded, saying that vertex painting could work— it would just change the appearance, only— and that vertex displacement could be relative to the weapon trail, with “parameters”. He said that it’d be “complex but doable probably.” He brought up an important point, though— collisions might not automatically change (they probably would not) with the displacement. So for my example of creating handholds in an unscalable cliff— well, I don’t know enough about reverse kinematics to determine how this would affect climbing.
We had our introduction to Praxis & Thesis meeting after class on Wednesday. I was able to talk with Amanda and Ulm after regarding my idea for Thesis. See, this program is “Interactive Media & Design”, so it makes sense to make the thesis interactive, as well. My idea was to research how to optimize creative productivity for neurodivergent people, and then to convey that research through a gamebook (like a Choose Your Own Adventure book, but with game mechanics)— with the narrative telling the research. Ulm had a good idea— she said that I definitely need to write a research paper, Semester 1, but that I could make the gamebook, Semester 2, and have that be my thesis presentation. I’m unsure how that presentation would go, because I want to have a self-bound physical book— but Ulm reassured me, saying that I have over a year to figure that out. And I get to do my idea! Which is great.
I also inquired whether it would be inefficient to start on the research paper/gamebook over the summer, or whether I would gain crucial information next year that would make such eagerness inefficient. Ulm and Amanda said that yes, definitely start on it as soon as possible— especially since I relayed that I want my gamebook to be at least 500 pages long (so that it looks appealing on my shelf).
For the premise of the gamebook, I think that I am going to go with the San Diego-based narrative surrounding The Mouldy Underpass, a hidden-world consignment store that sells both archaic artefacts and secret information. I had a slight realization: our praxes are supposed to be related to our theses, so the non-quantitative immersion aspect of Pareidolia: Unbound should transfer to the gamebook. How do I design gamebook mechanics that don’t involve numerical values? I have one idea that actually fits with the narrative premise quite well.
In Pareidolia: Unbound, the items that the player stuffs into the lidded cauldron strapped to their back relay buffs based on a qualitative “tag” system. In The Mouldy Underpass, there are canon multitudes of strange artefacts that would only naturally impart boons to the possessor. So maybe Halloway (the protagonist) gets to keep specific artefacts as rewards for good clerkmanship— or maybe he nabs them. Either way, those items that he acquires will allow for different abilities, that will allow for different outcomes.
