Recommended Free Tools
Some links on this page are affiliate links: if you buy through them we may earn a commission, at no extra cost to you.
NVIDIA’s early survival, as co-founders Jensen Huang and Chris Malachowsky later described it, depended less on following a fixed plan than on recognizing when the company was wrong and changing course. Their recollections, shared at NVIDIA’s 2018 GPU Technology Conference (GTC), trace two sharp reversals: a first graphics chip that reportedly barely worked, and a successful product whose momentum faltered after Intel entered the picture.
The setting was a startup competition, not a corporate-history lecture. At GTC in San Jose, Huang and Malachowsky spoke to companies competing in NVIDIA’s Inception Awards, an event with about $1 million in prize money at stake. Huang said the moment took him back roughly 25 years, when NVIDIA was itself a young company. Their stories offered the audience a vivid account of startup risk—but also a founder’s retrospective, not a fully documented corporate history.
| # | Preview | Product | Price | |
|---|---|---|---|---|
| 1 |
|
ASUS Dual GeForce RTX 5060 Ti 16GB GDDR7 OC Edition Gaming Graphics Card | $794.37 | Buy on Amazon |
| 2 |
|
ASUS TUF Gaming GeForce RTX™ 5080 16GB GDDR7 OC Edition Graphics Card | $1,770.00 | Buy on Amazon |
A broad vision, a badly flawed first product
The founders’ original thesis was that computer graphics would help drive technological progress, with video games supplying demand and economic fuel. That broad bet proved important in retrospect. It did not mean they had accurately forecast the gaming market’s eventual scale or timing, or that their first attempt to serve it was sound.
Huang recalled that about a year and a half into the effort, NVIDIA was close to going out of business. He attributed the danger to a combination of wrong technology, market strategy, architecture and execution—not to one isolated manufacturing mishap.
#1 Best Overall
- AI Performance: 767 AI TOPS
- OC mode: 2632 MHz (OC mode)/ 2602 MHz (Default mode)
- Powered by the NVIDIA Blackwell architecture and DLSS 4
- Axial-tech fan design features a smaller fan hub that facilitates longer blades and a barrier ring that increases downward air pressure
- A 2.5-slot design maximizes compatibility and cooling efficiency for superior performance in small chassis
The first graphics processor, NV1, had been taped out and returned from fabrication. Huang said about 90 percent did not work. That figure is his retrospective account as reported by IEEE Spectrum, not an independently audited yield statistic. The account does not specify what “did not work” meant technically, identify the failure mechanism, or quantify the cost and schedule effects.
Even so, the team was excited enough to demonstrate NV1 at Computex. The demo impressed observers, but a compelling public presentation was not the same thing as a viable product. The following year, NVIDIA returned to the same conference with essentially the same demo. Huang later treated that repetition as a warning sign: the company had not made the progress its situation required, and it nearly failed.
The episode suggests a broader startup risk. A polished demonstration can buy attention, but it cannot by itself establish that a product works reliably, can be made at scale, or has a durable market. In this account, the repeated demo is best read as a symptom of stalled progress, not as proof that the presentation itself caused NVIDIA’s crisis.
Quick wins for a faster PC:
Fix the driver behind crashes, sound loss and screen glitchesFind Drivers →Repair Windows errors before they cause bigger problemsFix Now →“No business plan” did not mean no planning
Huang also recalled being asked for a business plan in an early venture-capital meeting. He said he never completed one, arguing that no plan could have predicted NVIDIA’s next 25 years. That is a statement about the limits of long-range prediction, not persuasive evidence that planning is useless.
A young technology company still needs working assumptions, financial discipline, product priorities and a way to revise them. The distinction is between treating an early forecast as a script and using a plan as a tool that changes when evidence changes. Huang’s point was that a document could not have mapped every technology and market shift ahead—not that founders should stop budgeting, forecasting or deciding what to build next.
RIVA 128: success brought a different danger
NVIDIA’s later RIVA 128 graphics chip was, in Huang’s words, a “raging success.” But commercial momentum did not eliminate risk. Huang recalled that while NVIDIA was preparing to go public, Intel announced it would make a graphics chip. Customers pulled back, sales fell, and NVIDIA moved, in his description, almost overnight from being close to an IPO to being close to insolvency.
That is a striking account of how quickly expectations can turn in a competitive market. It should not be reduced to the claim that Intel alone nearly killed NVIDIA: the IEEE Spectrum report gives Huang’s compressed recollection, but no customer names, sales figures, cash balances, precise IPO timetable or independent account of why buyers retreated. The story shows how the founders remembered the shock; it does not establish the complete financial or competitive chain of events.
Rank #2
- Powered by the NVIDIA Blackwell architecture and DLSS 4. System Requirements: Minimum 850W PSU with 16-pin 12V-2x6 (12VHPWR) connector required. Verify before purchasing.
- Military-grade components deliver rock-solid power and longer lifespan for ultimate durability. Compatibility: 348mm (13.7") length, 3.6 slots, 4.3 lbs. Confirm case clearance and slot spacing. GPU bracket included.
- Protective PCB coating helps protect against short circuits caused by moisture, dust, or debris
- 3.6-slot design with massive fin array optimized for airflow from three Axial-tech fans
- Phase-change GPU thermal pad helps ensure optimal thermal performance and longevity, outlasting traditional thermal paste for graphics cards under heavy loads
The founders’ stated rule: protect the goal, not the plan
Across both crises, Huang and Malachowsky emphasized intellectual honesty: identify what is failing, admit it plainly and move on instead of defending a project out of pride. Malachowsky stressed surrounding a company with people willing to abandon an approach when needed. Their underlying advice was to put the long-term outcome ahead of attachment to a particular product or strategy.
That principle has a trade-off. Pivoting can rescue a company when its assumptions are wrong, but constant changes can also scatter effort and weaken focus. The useful distinction in the founders’ story is between a durable ambition—making computer graphics matter—and the methods used to reach it. They describe changing technology, architecture, products or positioning when the existing approach was not working. The account does not detail exactly which decisions followed NV1, when they were made, or how the company financed them.
What this recollection can—and cannot—tell us
The founders’ memories make the stakes clear, but they are not a substitute for a complete company history. The 2018 account does not explain NV1’s technical architecture or failure mechanism, provide manufacturing-yield data, document RIVA 128 sales, or include customers’ and Intel’s perspectives. It also offers no financial records to test the descriptions of near-insolvency or near-IPO status.
Retrospective stories can also make a difficult path seem more coherent after success. The founders’ emphasis on judgment and adaptability is valuable, but their account does not independently establish that those qualities alone explain NVIDIA’s survival; employees, investors, partners, customers and competitive conditions are not examined in comparable detail. Nor does a story about a graphics-and-gaming startup establish that NVIDIA’s later prominence in artificial intelligence was inevitable from the outset.
Do these 3 things before closing this tab:
1Repair Windows errors before they cause bigger problems2Fix the driver behind crashes, sound loss and screen glitches3Clear out junk files and repair common Windows errorsWhy tell the story to startup founders?
The Inception Awards put NVIDIA’s early uncertainty beside that of newer companies. The 2018 winners named in the account were Subtle Medical, focused on AI for medical imaging; AiFi, focused on checkout-free retail; and Kinema Systems, focused on industrial robotics and pallet handling. The parallel was about exposure to uncertainty, not a prediction that winning the awards guaranteed lasting success.
NVIDIA’s founders did not describe a straight line from a brilliant idea to an industry giant. Their account is of a broad vision surviving repeated failures of execution and strategy, followed by a new kind of threat after product success. The most defensible lesson is not to imitate a particular pivot or discard planning. It is to hold ambitions firmly enough to keep working toward them, while remaining willing to abandon the methods that reality has disproved.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
