NVIDIA's CEO sat down with senior members of the GeForce team to celebrate the 25th anniversary of the launch of the GeForce 3 GPU, which, according to Jensen, marked the beginning of the AI revolution.
NVIDIA's GeForce 3 Marked a Massive Shift From Fixed-Function Accelerators To Programmable Shaders
NVIDIA's CEO hasn't talked much about gaming in recent days, given how 'busy' he and his company are with the AI revolution, so it was nice to see him sit down with the GeForce team to talk about how gaming was basically what started the modern-day AI revolution. While talking with GeForce members, Jensen recalled that with GeForce 3, NVIDIA made the transition from fixed-function accelerators to programmable shaders, with the core idea being to infuse the "artistic" touch of developers into each game.
In the late 90s, Jensen says that every game started looking "the same" because of the fixed-function accelerators in models like the Riva 128 and TNT, which offered no flexibility in how each GPU would function. This led NVIDIA to leap, giving developers more control over how their games turn out through programmable vertex and pixel shader architecture, as we witnessed with the launch of GeForce 3. Jensen says that the shift towards this newer programming approach eventually paved the way for CUDA, which added parallelism to GPU computation.
Game was a medium for artistic expression. And that if you look at all these different games, we wanted them all to look different. And if you wanted them all to look different, you need the ability to express the artistry in some form of program.
It can't be pre-coded. And our company really didn't have a whole lot of compiler technology capability. And so as we were transitioning from fixed hardware accelerated pipeline to a also programmable fixed hardware pipeline, we realized that we had to move into and become a computing company.
- NVIDIA's Jensen Huang
Another major leap Jensen talks about was NVIDIA being one of the first to take a chance on ray tracing, since it was a computationally expensive process that required Team Green to rely on something other than raw compute power. Yet again, RTX paved the way for upscaling technologies, like DLSS, which leverages neural rendering to bring "generative capability to computer graphics". NVIDIA's CEO believes that consistent advancements in compute and rendering capabilities paved the way for technologies like generative AI.
So it's been a great joy serving all of you. Thank you for everything you've done for GeForce. Without GeForce, there would be no CUDA. Without CUDA, there would be no AI. Without AI, there would be no today. And so all of you have made it possible.
There's no doubt that advancements in computer graphics made NVIDIA realize that GPUs are much more than just for rendering workloads, and the gradual improvements in CUDA and related technologies paved the way for the modern-day AI computation we see today. At the same time, when we talk about the future of gaming and NVIDIA, the prospects look gloomy at the moment, but it appears that Team Green is eager to break traditional compute barriers by developing upscaling technologies that leverage AI to "generate" frames without the need for hefty hardware.
About the author: Muhammad Zuhair is a hardware and technology reporter for Wccftech, specializing in the semiconductor industry and the complex interplay between technology, manufacturing, and geopolitics. His coverage focuses on the corporate strategies and technological roadmaps of industry giants like TSMC, NVIDIA, Samsung, and Intel. Zuhair's expertise lies in deconstructing complex topics such as fabrication nodes (e.g., 2nm process), the economic impact of policies like the CHIPS Act, and the strategic development of AI infrastructure from NVIDIA, AMD and Intel.
Follow Wccftech on Google to get more of our news coverage in your feeds.
