For about a quarter of a century, Nvidia has led the revolution in computer graphics and become a beloved brand of gamers on the road.
Nvidia dominates the graphics processing unit (GPU) market, which it entered in 1999 with the GeForce 256. Gaming brought in over $9 billion in revenue for Nvidia last year despite a recent downturn.
But Nvidia’s last winning streak points to a new phenomenon in the GPU industry. The technology is now at the center of the artificial intelligence boom.
“We had the good wisdom to put the whole company behind it,” CEO Jensen Huang told CNBC in an interview last month. “We saw early on, about a decade or so ago, that this way of doing software could change everything. And we changed the company from the bottom all the way to the top and sideways. Every chip we made was focused on artificial intelligence.”
As the engine behind large language models (LLM) like ChatGPT, Nvidia is finally reaping the rewards of its early investment in AI. It has helped soften the blow wider the semiconductor industry struggles tied to Trade tensions between the US and China and a global chip shortage.
Not that Nvidia is immune to geopolitical issues. In October, the United States introduced sweeping new rules which banned the export of leading AI chips to China. Nvidia counts on China for about a quarter of its revenue, including sales of its popular AI chip, the A100.
“It was a turbulent month or so as the company went up and down to redesign all of our products to be compliant with the regulation and still be able to serve the commercial customers that we have in China,” Huang said. “We can serve our customers in China with the regulated parts and delightfully support them.”
AI will be a big focus at Nvidia’s annual GTC Developer Conference runs March 20-23. Ahead of the conference, CNBC sat down with Huang at Nvidia’s headquarters in Santa Clara, California, to discuss the company’s role at the heart of the explosion in generative AI.
“We just thought that one day something new would happen, and the rest of it requires a bit of serendipity,” Huang said when asked if Nvidia’s fortunes are the result of luck or foresight. “It wasn’t foresight. The foresight was accelerated computing.”
GPUs are Nvidia’s primary business and account for more than 80% of revenue. Usually sold as cards that plug into a computer’s motherboard, they add computing power to central processing units (CPUs) built by companies such as AMD and Intel.
Now there are tech companies trying to compete with ChatGPT public boasting about how many by Nvidia’s roughly $10,000 A100s they have. Microsoft said the supercomputer developed for OpenAI use 10,000 of them.
Nvidia founder and CEO Jensen Huang shows CNBC’s Katie Tarasov a Hopper H100 SXM module in Santa Clara, California, on February 9, 2023.
Andrew Evers
“It’s very easy to use their products and add more computing power,” said Vivek Arya, semiconductor analyst for Bank of America Securities. “Computing capacity is basically the currency of the valley right now.”
Huang showed us the company’s next-generation system called the H100, which has already started shipping. The H stands for Hopper.
“What makes the Hopper really amazing is this new type of processing called a transformer motor,” Huang said while holding a 50-pound server card. “The transformer engine is T for GPT, generative pretrained transformer. This is the world’s first computer designed to process transformers at massive scale. So large language models will be much, much faster and much more cost-effective.”
Huang said he “hand delivered” to ChatGPT maker OpenAI “the world’s very first AI supercomputer.”
Not afraid to bet everything
Today, Nvidia is among the world’s 10 most valuable technology companies, with a market capitalization of close to $600 billion. It has 26,000 employees and a newly built polygon-themed headquarters. It’s also one of the few Silicon Valley giants with a 30-year-old founder still at the helm.
Huang, 60, immigrated to the United States from Taiwan as a child and studied engineering at Oregon State University and Stanford. In the early 1990s, Huang and fellow engineers Chris Malachowsky and Curtis Priem used to meet at a Denny’s and talk about dreams of enabling computers with 3D graphics.
The trio launched Nvidia from an apartment in Fremont, California, in 1993. The name was inspired by NV for “next version” and Invidia, the Latin word for envy. They hoped to speed up computing so much that everyone would be green with envy – so they chose the jealous green eye as the company’s logo.
Nvidia founders Curtis Priem, Jensen Huang and Chris Malachowsky pose at the company’s headquarters in Santa Clara, California, in 2020.
Nvidia
“They were one of dozens of GPU manufacturers at the time,” Arya said. “They’re the only ones, they and AMD actually, that really survived because Nvidia worked very well with the software community, with the developers.”
Huang’s ambitions and preferences for seemingly impossible ventures have pushed the company to the brink of bankruptcy a number of times.
“Every company makes mistakes and I make a lot of them,” said Huang, who was one of Time Magazine’s 2021 Most Influential People. “Some of them put the company at risk, especially in the beginning, because we were small and we We face very, very large companies and we’re trying to invent this whole new technology.”
In the early 2010s, for example, Nvidia made a failure move to smartphones with its Tegra series of processors. The company then left the space.
In 1999, after laying off the majority of its workforce, Nvidia released what it claims was the world’s first official GPU, the GeForce 256. It was the first programmable graphics card that allowed custom shading and lighting effects. In 2000, Nvidia was the exclusive graphics provider for Microsoft’s first Xbox. In 2006, the company made another big move, releasing a software toolkit called CUDA.
“For 10 years, Wall Street asked Nvidia, ‘Why are you making this investment? No one is using it.’ And they valued it at $0 in our market cap,” said Bryan Catanzaro, vice president of applied deep learning research at Nvidia. He was one of the only employees working on AI when he joined Nvidia in 2008. Now the company has thousands of employees working in the space.
“It wasn’t until around 2016, 10 years after CUDA came out, that people suddenly understood that this is a dramatically different way of writing computer programs,” Catanzaro said. “It has transformational speeds that then produce breakthrough results in artificial intelligence.”
Although AI is growing rapidly, gaming is still Nvidia’s primary business. In 2018, the company used its AI expertise to make its next big step in graphics. The company introduced GeForce RTX based on what it had learned in AI.
“In order to take computer graphics and video games to the next level, we had to reinvent and disrupt ourselves, literally changing what we invented entirely,” Huang said. “We invented this new way of doing computer graphics, ray tracing, basically simulating the paths of light and simulating everything with generative AI. And then we calculate one pixel and we imagine with AI the other seven.”
“Boom-or-bust cycle”
From the start, Huang was determined to make Nvidia a fab chip company, or one that designs the product but contracts out production to others that have chip manufacturing factories, or fabs. Nvidia is keeping capital expenditures down by outsourcing the extraordinary costs of making the chips Taiwan Semiconductor Manufacturing Company.
Taiwan Semiconductor Manufacturing Company’s US office in San Jose, California, 2021.
Katie Tarasov
Investors are right to be concerned about that level of reliance on a Taiwanese company. The US passed the CHIPS Act last summer, which sets aside an additional $52 billion encourage chip companies to manufacture on American soil.
“The biggest risk is really US-China relations and the potential impact of TSMC. If I’m a shareholder in Nvidia, that’s really the only thing that keeps me awake at night,” said CJ Muse, an analyst at Evercore. “This is not just an Nvidia risk, this is a risk for AMD, for Qualcomm, even for Intel.”
TSMC has said it’s expenses 40 billion dollars to build two new chip manufacturing facilities in Arizona. Huang told CNBC that Nvidia will “absolutely” use TSMC’s Arizona fabs to make its chips.
Then there are questions about demand and how many of the new use cases for GPUs will continue to show growth. Nvidia saw a surge in demand as cryptomining took off as GPUs became core to effectively compete in that market. The company even created a simplified GPU just for crypto. But with the cratering of crypto, Nvidia experienced an imbalance in supply and demand.
“It’s created problems because cryptomining has been a boom-or-bust cycle,” Arya said. “Playing cards run out of stock, prices go up, and then when the cryptomining boom collapses, there’s a big crash on the gaming side.”
Nvidia caused big stick shock among some gamers last year by pricing its new 40-series GPUs much higher than the previous generation. Now there is too much supply and in the most recent quarter, gaming revenue was down 46% from a year ago earlier.
Competition is also increasing as more technology giants design their own chips for special purposes. Tesla and Apple do it. So are Amazon and Google.
“The biggest question for them is how do they stay ahead?” Arya said. “Their customers can also be their competitors. Microsoft can try to design these things in-house. Amazon and Google are already designing these things in-house.”
For his part, Huang says such competition is good.
“The amount of power the world needs in the data center will grow,” Huang said. “It’s a real issue for the world. The first thing we should do is: every data center in the world, however you decide to do it, for the good of durable computing, accelerate everything you can.”
In the car market, Nvidia is doing autonomous driving technology for Mercedes-Benz and other. Its systems are also used to power robots in Amazon’s warehouses and to run simulations to optimize the flow of millions of packages every day.
Huang describes it as “omniverse.”
“We have over 700 customers trying it now, from (the auto industry) to logistics warehouses to wind farms,” Huang said. “It probably represents the single largest container of all of Nvidia’s technology: computer graphics, artificial intelligence, robotics and physics simulation, all in one. And I have high hopes for it.”