It's difficult to imagine how games would look today without the quiet yet transformative change that Microsoft brought about in the year 2000. Twenty-five years ago, to this day, the company introduced DirectX 8. The release was accompanied by little fanfare, no generation-defining tech demo, but it carried one major breakthrough with it — programmable shaders — which would forever revolutionize the way GPUs render graphics.
Before DirectX 8, graphics cards worked on a fixed-function pipeline, meaning that almost everything was predefined, baked into the silicon itself. Lighting equations, texture blending, transformations; it was all at the mercy of the GPU's support. For instance, in lieu of real-time reflections, you'd have environment maps because the GPU itself wasn't able to calculate them dynamically. You were bound by the logic of the hardware, which wasn't very flexible.
Think of it as adjusting knobs on a console. You can tweak the parameters slightly, but what if you wanted to change the very knobs themselves? Enter, DirectX 8.
DirectX 8's programmable shaders
Microsoft added Shader Model 1.0 to DirectX, which came with Vertex Shader 1.0 that would allow devs to manipulate each vertex, and Pixel Shader 1.0 that would allow devs to control the final color of each pixel. Previously, none of this was really accessible, but DirectX 8 gave control to the people making the games, enabling them to write code to program the GPU to render in a certain way.
Devs could define the very math behind lighting, for example, or control material behavior, specular highlights, (basic) tessellation, and so on for the first time ever. It was no longer about just accepting what the GPU provided by default, but rather what the GPU could do, full stop. It turned the silicon from a locked-down machine to an actual, programmable processor, true to its name.
Piggybacking off the possibilities of DirectX 8, games like Half-Life 2 would arrive with groundbreaking graphics for the era. They were built on the ideas of authorship, giving autonomy (even if minor at the time) to devs to build real-time shadows, refraction and water shaders, post-processing effects and so much more. All because they could now write custom code, telling the GPU how to calculate something.
It was up to the creatives to shape how light interacted with each object, something that may seem primitive by today's standards. Even in 2025, some things are pre-baked, like global illumination, for efficiency's sake, while others operate in real-time. And it's shaders that decide how both interact with each other, merging them seamlessly to give you "better graphics."
Stay On the Cutting Edge: Get the Tom's Hardware Newsletter Get Tom's Hardware's best news and in-depth reviews, straight to your inbox. Contact me with news and offers from other Future brands Receive email from us on behalf of our trusted partners or sponsors
Modern-day GPUs (Image credit: Tom's Hardware)
... continue reading