Insomniac Games senior environment artist, Ryan Benno, has put together a super informative Twitter thread that looks at the history of 3D texturing in video games, and it’s well worth a look if you’re even vaguely interested in how games are put together.
Charting the evolution of texturing from the ’90s onward, Benno, who most recently worked on Marvel’s Spider-Man, starts by digging into the differences between real-time rendering and pre-rendered textures, looking at how each type affects hardware performance and delivers different results.
For instance, he explains how the emergence of the first true 3D consoles like the Nintendo 64 and PlayStation exposed some of the limitations of real-time rendering, with developers at the time unable to bake lighting and shadows into a scene or use bump mapping.
The resulted in some rather creative workarounds, and artists would often paint lighting information (shadows, highlights, and depth) onto the vertices of textures themselves to make areas appear lighter or darker.
Player shadows were also usually comprised of a single texture that followed the character around, as there was no way to cast actual shadows at the time. Yep, that means shadows were just lonely textured glued to their counterparts, doomed to follow their host to the very ends of the Earth. How melancholy.
If those brief tidbits scratched an itch, you’ll definitely want to check out the full thread, which delves into everything from special maps and real-time shadow acting to ambient occlusion. It’s a veritable treasure trove of game development delights.