Nvidia CEO Jensen Huang Says Nvidia Would Not Exist Without Windows
Nvidia CEO Jensen Huang stated at a Windows and Surface event co-hosted with Microsoft CEO Satya Nadella that Nvidia would not have been founded without Windows.
The event took place on October 7 in San Francisco, where both parties launched laptops equipped with Nvidia RTX Spark chips. Huang recalled that at the end of 1992, during the Windows 3.1 era, he and his co-founders envisioned personal computers with 3D graphics capabilities that could run games and serve as design workstations, leading to the company's official establishment in 1993.
Huang mentioned that prior to this, he mainly encountered workstations, file servers, and supercomputers used for chip design, and had never seen a personal computer. Windows 95 made it possible for graphics chips to connect to personal computers, and DirectX opened up these chip capabilities to applications.
During the DirectX 8 phase, Nvidia and Microsoft co-invented the first programmable GPU, which later evolved into the CUDA software platform. Huang pointed out that without Windows, there would be no GeForce, and without GeForce, there would be no CUDA, making it impossible for early deep learning researchers like Alex Krizhevsky and Ilya Sutskever to train models on relevant hardware.
When Nvidia went public in January 1999, its market value was less than $1 billion, while Microsoft was around $390 billion. At the event, Nadella jokingly expressed a desire for Huang to transfer some market value back, to which Huang responded, "I do that every day."
Market Mechanism: The standardization of operating systems and graphics APIs created a massive demand for personal computer graphics chips. Nvidia achieved product implementation and subsequent software expansion by adapting to the Windows ecosystem; Microsoft gained support for graphics and computing capabilities, while Nvidia transitioned from gaming graphics cards to general-purpose accelerated computing. Their long-term collaboration transformed early hardware demand into a major supply source for today's AI computing power market.
Source: Public Information
ABAB AI Insight
Jensen Huang co-founded Nvidia in 1993 with Chris Malachowsky and Curtis Priem, initially focusing on graphics acceleration chips. After several product failures, the company aligned with Microsoft's DirectX standard, ultimately opening the market with the GeForce series. Nvidia has consistently pursued an accelerated computing path rather than relying solely on general-purpose CPUs.
In terms of capital, Nvidia has continuously invested resources into GPU hardware and the CUDA software stack, expanding from gaming graphics cards to scientific computing and deep learning. Microsoft, through Windows and DirectX, provided standardized interfaces and a large user base, enabling Nvidia to connect chips to personal computers and cultivate a developer ecosystem, with CUDA later becoming a key platform for AI training.
Similar cases include Intel leading the personal computer era through its Wintel alliance with Microsoft, and AMD's long-term competition in graphics and CPU fields. Currently, Nvidia is in a phase of expansion from graphics-specific chips to general AI acceleration platforms, with its market value significantly surpassing that of early partners.
This fundamentally involves technological substitution and industrial chain reconstruction. The operating system and API standards lowered the barriers for graphics hardware to enter personal computers, while programmable GPUs and software layers further transformed specialized graphics capabilities into general computing platforms. The early reliance on the Windows ecosystem laid the foundation for subsequent CUDA and AI computing power, forming an industrial chain extension from consumer graphics to data center acceleration.