Why Nvidia wins
How Jensen Huang won AI despite several near failures.
“Our company is thirty days from going out of business.” According to CEO Jensen Huang, this is Nvidia’s unofficial corporate motto. Nvidia’s history is fascinating. What looks like an overnight success was decades in the making, with lots of hard-won lessons along the way.
We’re breaking down the story behind Nvidia’s unofficial motto and lessons from three Nvidia stories Jensen shared in an inspiring commencement speech last year. There are some wise words here. Ready? Let’s dive in!
Lesson one: Be humble and ask for help
One of Jensen’s early jobs was as a dishwasher at Denny’s. He often starts his story with these humble beginnings and hard work rather than mentioning that he was also a brilliant engineer and Stanford graduate. He said he considered himself successful until he started Nvidia.
“I was also successful until I started Nvidia. At Nvidia I experienced failures, great big ones, all humiliating and embarrassing. Many nearly doomed us. Let me tell you three Nvidia stories that define us today.”
— Jensen Huang (Source: NTU commencement speech)
That’s from Jensen’s commencement speech at National Taiwan University last year. And here’s the first of the three stories he shared:
In the 1990s, Nvidia won a contract with SEGA to build their next-generation console, the SEGA Dreamcast. At the time, Nvidia used quadrilateral-based graphics, shapes with four points, because it was cheaper to build and looked better. Their competitors were using triangles, shapes with three points, which required more expensive hardware but made developing software and rendering cheaper and faster.
Two-dimensional triangles can easily be combined to make three-dimensional shapes. That’s why everything looks like a Tesla Cybertruck in classic video games. Modern devices can support millions of triangles on a single screen, but each triangle has three simple points, floating points, with coordinates that are quick and easy to change. This would become very important for AI years later, but we’ll get to that.
The trouble with quadrilaterals, Nvidia’s approach, is that four points can exist on multiple planes, creating complex shapes like a pyramid. Now, imagine moving millions of those shapes onto a screen, which uses pixels, and your brain breaking is what’s called rendering. 🤯
Quadrilateral-based shapes could not be rendered as fast as triangles could. In the end, faster was better than better. At the time of this realization, Microsoft announced it would only support triangle-based development in Windows 95, a death sentence for Nvidia’s entire business. Nvidia’s new NV1 chip and their next-generation NV2 chip, which was still in development, were obsolete before they could even hit store shelves. RIP, NV2.
Jensen would need to abandon quadrilateral graphics, do a 180, and create an entirely new chip based on triangular graphics, but the company didn’t have enough money to start all over in that new direction. They were also one year into the development of the SEGA Dreamcast and really needed that money, but if they continued developing their inferior technology, their core chip business would go belly up. Both roads led to bankruptcy.
Here’s what Jensen did: He called SEGA’s CEO and explained that they couldn’t continue to build inferior technology, but if they were not paid, Nvidia would go bankrupt. To his shock, SEGA paid, and when Nvidia launched a new chip, they were only 30 days from going out of business.
The lesson: Be humble, ask for help, and always work like you’re 30 days from going out of business.
Triangles nearly destroyed Nvidia. It was painful; they had to lay off more than 50% of their workforce and almost failed. Yet this is what Nvidia’s Silicon Valley headquarters looks like. Notice anything? 🤔
Lesson two: Endure pain and suffering with glee
Jensen recently appeared on the Acquired podcast and was asked what company he would start today if he were magically 30 years old again and starting from scratch. His answer was shocking: He wouldn’t do it!
“The reason why I wouldn’t do it, and it goes back to why it’s so hard, is building a company and building Nvidia turned out to have been a million times harder than I expected it to be, any of us expected it to be. And at that time if we realized the pain and suffering and just how vulnerable you’re going to feel and the challenges that you’re going to endure, the embarrassment and the shame and the list of all the things that go wrong, I don’t think anybody would start a company. Nobody in their right mind would do it. And I think that that’s kind of the superpower of an entrepreneur, they don’t know how hard it is and they only ask themselves ‘How hard can it be?’ and to this day I trick my brain into thinking ‘How hard can it be?’”
— Jensen Huang (source: Acquired)
This reminds me of the most popular quote from Jessica Livingston’s book Founders at Work: “Determination is the single most important quality in a startup founder.” Jensen was no exception to this rule. People think of startup success as glamorous, but Jensen says you should have low expectations and embrace pain and suffering.
When he was recently asked to give advice to Stanford University students about building a successful company, his answer was: You need resilience, and that doesn’t come from intelligence; it comes from pain and suffering.
“To this day I use [the phrase] ‘pain and suffering’ inside our company with great glee. … I wish upon [you] ample doses of pain and suffering.”
— Jensen Huang (source: Stanford SIEPR keynote)
This brings us to the second Nvidia story from Jensen’s NTU commencement speech. Nvidia nearly failed because they bet big on CUDA, a platform they developed in 2006 as a revolutionary way to program on a GPU instead of a CPU. GPUs are capable of parallel computing, which would make them especially useful for breakthroughs in machine learning, data, and AI, but they needed CUDA to unlock their future potential in AI.
Focusing on CUDA was a costly gamble that didn’t pay off for years. It took the stock market nearly two decades to realize the potential of CUDA and AI. At a time when tech stocks were soaring, Nvidia’s stock performed terribly.
The lesson: Overnight successes often take years of pain and suffering.
In the background of all those years before Nvidia stock took off, innovation was happening, and it was as if the company would go bankrupt in 30 days. They were doing things that don’t scale, helping researchers write code, and even giving away AI supercomputers to non-profits. In 2016, Jensen hand-delivered the world’s first AI supercomputer in a box to a non-profit called OpenAI, and we all know how that turned out. 📈
Lesson three: Pursue your vision and retreat if you go off course
The third and final Nvidia story from Jensen’s commencement speech was about following your vision.
In 2010, Nvidia partnered with Android to build chips for mobile phones. They were an instant success. However, competition quickly grew and Nvidia would need to dedicate significant resources to succeed, but the mobile market did not fulfill their mission to build computers to solve problems that ordinary computers could not. So, they strategically retreated from a multi-billion dollar industry to work on robotics chips in a zero-billion dollar industry.
“We now have billions of dollars of automotive and robotics business and started a new industry. Retreat does not come easily to the brightest and most successful people like yourself, yet strategic retreat, sacrifice, deciding what to give up, is at the core, the very core, of success.”
— Jensen Huang (source: NTU commencement speech)
The lesson: Follow your vision and strategically retreat when you go off course.
Deciding what to give up is at the very core of success. To be strategic, you must learn to select the very few products or features that can fulfill your company’s vision and retreat from all other projects.
Nvidia’s journey is one of the most amazing success stories. Check out the full references linked below to dive deeper.
You got this!
If you’re looking for more product management lessons, here’s how I can help!
📚 Product Management Foundations course. This self-paced training program will set a strong foundation for your product career and offers optional one-on-one sessions with me.
📺 PM Crash Course (free) is a fun, free playlist of videos that introduces the basics of product management.
📖 Product Influence course. Learn simple influence principles and frameworks to take your product career to the next level.
❤️ Develop Products People Love course on Skillshare—a beginner class on product development for entrepreneurs.
🙋🏻♀️ 1-on-1 coaching. For personalized coaching, book a one-on-one call with me. You can also book time with me on Skillshare!
Referenced in this post & video
[YouTube] Commencement speech at National Taiwan University, May 2023, by TTV News
[The New Yorker] How Jensen Huang’s Nvidia Is Powering the A.I. Revolution by Stephen Witt (November 2023)
[The Wall Street Journal] He Built a Trillion-Dollar Company. He Wouldn’t Do It Again. by Ben Cohen (December 2023)
[YouTube] NVIDIA CEO Jensen Huang interview by Acquired
[YouTube] Nvidia CEO Jensen Huang and the $2 trillion company powering today's AI by 60 Minutes
[YouTube] Keynote by NVIDIA CEO Jensen Huang at 2024 SIEPR Economic Summit by Stanford Institute for Economic Policy Research
[YouTube] Nvidia CEO Jensen Huang full keynote at GTC 2024 by Yahoo Finance
[YouTube] E167: Google's Woke AI disaster, Nvidia smashes earnings (again), Groq's LPU breakthrough & more by All-In Podcast
[YouTube] How do Video Game Graphics Work? by Branch Education
[YouTube] Why video games are made of tiny triangles by Vox
[YouTube] Nvidia CUDA in 100 Seconds by Fireship
[WebWire] NVIDIA CEO Delivers World’s First AI Supercomputer in a Box to OpenAI (August 2016)
[Electronic Design] Nvidia’s RIVA 128
[Nvidia.Developer] Life of a triangle - NVIDIA's logical pipeline
[Nvidia] NVIDIA History and About Us