History of NVIDIA: Company and Stock

NVIDIA makes computer processors. The company started out, primarily, in the video game and graphic design market. In the late 1990s and early 2000s, it was known mainly as a leading company in the relatively niche high-performance gaming industry. 

Today, it is one of the biggest tech companies in the world with a market cap of $2.25 trillion ($NVDA), and it is a main hardware supplier for major emerging technologies such as AI and cryptography. Here’s what you need to know.

NVIDIA’s core business model is the production of specialized, high-end computer processors. These chips are called “graphics processor units,” or GPUs. 

A GPU is a specialized form of processor. It supports the central processing unit, or “CPU,” by taking on specific, high-demand tasks as they emerge. This allows your computer to operate normally while also running demanding software because, instead of having to put all other processes on hold, the CPU can hand off single, complicated calculations to the GPU. In essence, it’s a difference of breadth vs. depth.

A central processing chip is built to handle many different tasks at the same time, and its purpose is to run your computer overall. However, as a result, it has fewer resources for any given task. So, for example, the CPU can run the relatively simple calculations of your operating system, web browser and word processor all at once. But it will take a long time to process very complicated demands, which can slow down your computer overall. To do this, the CPU uses a physical architecture known as “serial processing.”

A graphics processor unit, on the other hand, is built to do one specialized task at a time, and . It can dedicate all of its resources to a single calculation by breaking the task down into smaller, repetitive units, but it cannot handle simultaneous demands. So, for example, a GPU can handle the enormously complicated math of rendering three dimensional images very quickly (historically the most common use case for a GPU), but it would move very slowly if tasked to run four different programs. This is a physical architecture known as “parallel processing.”

The name “graphics processing unit” comes from the fact that, originally, GPUs were mainly used for graphics rendering in video games. Originally these chips were known as “ASICs,” or “Application Specific Integrated Circuits. During the 1990s and 2000s, video games overwhelmingly drove the demand for high end consumer-level hardware, and GPUs found an early niche in meeting this demand.

However, that is simply an accident of market demand, not a function of hardware limitations. NVIDIA makes sophisticated computer processors designed to supplement core processing and solve specific, high-demand needs. In recent years, demand for specialized computing has grown significantly. These chips are used, among other industries, for blockchain mines, artificial intelligence processing, and data storage and distribution. As a result, demand for NVIDIA’s chipsets has grown, and today industry-specific uses such as data centers and AI have surpassed consumer graphics as the core of the company’s business model.

NVIDIA does make a range of additional products beyond the GPU, including laptops, cloud computing, and networking hardware. However the company is overwhelmingly known for their application specific chips, which make up the bulk of their brand and revenue.