Transitioning from 3d artist for film/animation to videogame industry.
Since I started to study to be a 3d artist, I always oriented it towards the film and animation industry, and all my past jobs were as 3D artist for animated films, so I all the technical stuff that I learned so far (workflows, polycount, topology, texture resolution...) was aiming to this field.
The thing is that I would like to work in videogames at some point, and Im quite lost about where should I start. Im not talking about artistis or quiality stuff, but more about the technical limitations and workflows to follow for videogames.
Does anyone knows where to find resources and info about this?
The thing is that I would like to work in videogames at some point, and Im quite lost about where should I start. Im not talking about artistis or quiality stuff, but more about the technical limitations and workflows to follow for videogames.
Does anyone knows where to find resources and info about this?
Replies
We've all been there, and many here want to help others with their journey.
What kind of game content are you interested in making?
Well, I dont have a game content in mind right now, I was mostly looking for the differences in modeling, UVing, texturing of sets and props... between both fields.
Im just trying to figure out what would change from working in a movie production to work in a game production.
Subdivision surfaces mostly aren't used, except as a way to generate textures ahead of time.
UDIMs mostly aren't used. UVs are often straightened (for better LODs, and easier texture reuse). UV mirroring and overlapping is used a lot, and texture sizes are constrained to smallest possible (and always powers-of-two).
Modularity is big, mostly to reduce number of textures. Trim sheets. Channel packing. UV atlases. Optimized shaders. Hacked lighting.
Wiki has most of this stuff.
The best thing really, is to dive in and start making content in a game engine. Learn by example.
What we do for games or animation is practically the same nowadays. One example is Arch Viz, Unreal engine is being used a lot, and the assets used in realtime are the same of the 3ds Max scene without optimizing anything (Evermotion assets).
Actual videogame character models are like a subdiv model with 1-2 iterations applied with some mesh parts edited. And we also use multiple 8k textures aswell. The environments use an amount of mesh and details we usually don't use/view in DCC, and all in realtime. Just look Unreal engine 5, in a few years all the things we know will serve for nothing, like all the things i did back in 2005.
You won't need to optimize meshes of 1000-4500 triangles, and you won't need to do hand painted textures. And less, you won't need to worry about using just one single texture of 512 or 1024 and crazy UV optimizations that take time and too much effort. All is very flexible nowadays, we have almost no limits.
Right now, you need to know a lot about normal maps, trims, decals, and a few other things. So texturing is practically the same, the same pbr materials but less textures of 4-8k.
Don't forget to read the sticky threads. There's a good read about normal map usage and its issues.
However for mobile & web (a significant market) optimization is still essential, and will remain so for a long time.
I'd love to write all that off, but we're not there yet. And we can't use artists who don't understand how to excel within limits.