Skip to main content

The Rise of NVIDIA, Part 15: When Data Center Ate Gaming

When Data Center Ate Gaming

🎮 The Rise of NVIDIA — a 20-part series. See all parts »  |  « Part 14: The $40 Billion That Got Away

For thirty years, one number defined NVIDIA. It was the number of gamers who bought a graphics card to make Crysis run a little smoother, or to squeeze another few frames out of Call of Duty. Gaming was the identity, the origin story, the reason the company existed. Jensen Huang had founded NVIDIA in 1993 to render triangles faster than anyone else, and for three decades gaming paid the bills, funded the research, and defined the brand.

Then, in the spring of 2022, without a keynote or a product launch to mark the moment, that number stopped being the biggest one on the page.

A quiet line in an earnings report

On May 25, 2022, NVIDIA reported results for the first quarter of its fiscal 2023 — the three months ending May 1. Total revenue was a record $8.29 billion. Buried in the segment breakdown was a sentence that, in hindsight, marked a turning point in the history of computing. Data Center revenue had reached $3.75 billion, up 83% from a year earlier. Gaming, even after a record quarter of its own, came in at $3.62 billion, up 31%.

For the first time ever, the chips NVIDIA sold to cloud providers, research labs, and AI startups had out-earned the chips it sold to gamers. "Data Center has become our largest platform, even as Gaming achieved a record quarter," Jensen Huang noted. It was said almost in passing — a milestone delivered in the register of a footnote.

The crossover you see above wasn't a single event so much as a scissor motion. One blade rose steadily; the other, having climbed to a record, suddenly fell away. Understanding why both things happened at once is the whole story of NVIDIA in 2022.

Why data center kept climbing

The rise of the blue line had been years in the making. It started with CUDA in 2006 — the decision to let a graphics chip run general-purpose code — and accelerated when researchers discovered around 2012 that GPUs were spectacularly good at training neural networks. By the early 2020s, the world's largest companies were racing to build AI infrastructure, and there was really only one place to buy the shovels.

Amazon, Microsoft, Google, and Meta were filling data centers with NVIDIA's A100 accelerators. Every recommendation engine, every translation model, every fraud detector, every early large language model ran on the same silicon. This was recurring, structural demand from customers with effectively unlimited capital budgets. It didn't care about console cycles or holiday shopping. It just grew, quarter after quarter: $3.26 billion, $3.75 billion, $3.81 billion, $3.83 billion. Steady, relentless, indifferent to the weather.

Why gaming fell off a cliff

The green line's collapse had a more dramatic cause. Through 2020 and 2021, gaming GPUs had been almost impossible to find at retail — not only because pandemic-bound gamers wanted them, but because cryptocurrency miners were buying every card they could to mine Ethereum. That demand was real revenue, but it was borrowed against the future. It inflated the gaming numbers with sales that had nothing to do with gaming.

In 2022, the bill came due. Crypto prices crashed through the year, and in September Ethereum completed its long-planned move to proof-of-stake — "The Merge" — which eliminated GPU mining almost overnight. The channel was suddenly flooded with used cards and unsold inventory. NVIDIA's gaming revenue, riding a record $3.62 billion in the spring, cratered to $2.04 billion by summer (down 33% year over year) and $1.57 billion by autumn — a stunning 51% collapse from the prior year.

NVIDIA knew the risk of crypto-muddied numbers better than most. Back in 2022 the SEC had charged the company $5.5 million for inadequately disclosing how much of its earlier gaming growth had come from mining — a reminder that the line labeled "Gaming" had, for a while, been telling investors a story that wasn't entirely about games.

So the crossover was really two forces meeting: a durable new business rising into the vacuum left by a speculative old one deflating. Had crypto never happened, data center would still have overtaken gaming eventually — the trend lines guaranteed it. The mining bust just made the moment arrive with a bang instead of a whisper.

The company nobody had noticed changing

By the close of fiscal 2023 in January 2023, the transformation was written into the full-year books. Total revenue was flat at $27.0 billion, but the mix had inverted. Data Center rose 41% to a record $15.01 billion. Gaming fell 27% to $9.07 billion. The company that everyone still thought of as a gaming-hardware maker now earned more from the cloud than from the living room — and the gap was widening every quarter.

Wall Street, fixated on the plunging gaming line and a brutal 2022 for tech stocks, largely read the quarter as a story of decline. What the numbers were actually announcing was a change of identity. NVIDIA was no longer primarily a company that made things go fast on your screen. It was becoming the company that made intelligence go fast in the cloud — and almost nobody outside the earnings call had noticed yet.

They were about to. Just weeks before that fiscal year closed, a chatbot called ChatGPT had been released to the public, and the demand it was about to unleash would make even the blue line on this chart look flat. To feed it, NVIDIA had already built something new — a chip designed not to draw pixels, but to think.

Next time: Hopper and the H100 — the architecture NVIDIA built specifically for the transformer, and the engine of the AI boom.


🔗 Explore more from Syncster

Comments

Popular posts from this blog

Cursor AI Review: Is the AI Code Editor Worth It?

I've been using Cursor as my main code editor for a while now, and enough people have asked whether it's worth switching to that a proper review felt overdue. Short version: for me, yes — but with caveats. What is Cursor? Cursor is an AI-first code editor built as a fork of VS Code. That means every extension, theme, and keybinding you already use in VS Code works here, but with AI woven directly into the editing experience instead of bolted on as a plugin. It's made by Anysphere and can run models from OpenAI and Anthropic under the hood. What I like Tab completion is uncanny. Cursor predicts your next edit — not just the rest of the line, but the next change across the file. Once you get used to hitting Tab, going back to a plain editor feels slow. The Composer / Agent mode. You describe a change in plain language and it edits multiple files at once, showing you a diff to accept or reject. For refactors and boilerplate, this saves real time. It unde...

How I used Google Sheets and Apps Script

Google Sheet is one of the most powerful spreadsheet application that exists online, rivaling with Microsoft's Excel. One of the main strengths is its strong support for collaboration with other users, much easier and popular than collaboration tools with Microsoft Office. Aside from plain spreadsheet, it also supports extensions such as macro. If you are familiar with macros on other office tools, they work almost the same. However, the most extension I use and tinker with is the Apps Scipt . Apps Script Extension One of the challenges I faced recently is how do I track or monitor reports in our department if they are submitted on time or worst, forgotten due to lack of better monitoring tools. So I thought if there can be simple applications that can be deployed or use by a more general user to allow reminding periodically what reports are approaching due dates or those that are past dues. Then I looked for a way, instead of creating a full blown app from scratc...

MacBook Pro M5 vs M5 Pro: Which One Should You Actually Buy?

Apple's latest 14-inch MacBook Pro comes in two very different flavors: the base M5 and the step-up M5 Pro . On paper they look similar — same gorgeous Liquid Retina XDR display, same design — but under the hood the gap is bigger than the names suggest. Here's a clear, no-hype breakdown, with concrete use cases so you can match the chip to your work. Quick spec comparison Spec M5 M5 Pro CPU 10-core (4 performance + 6 efficiency) Up to 18-core (6 performance + 12 efficiency) GPU 10-core Up to 20-core Neural Engine 16-core 16-core Memory bandwidth 153 GB/s 307 GB/s (roughly double) Unified memory 16 / 24 / 32 GB 24 / 48 / 64 GB Max storage Up to 4 TB SSD Up to 8 TB SSD Battery (video playback) Up to 24 hours Up to 22 hours Media engines Single encode/ProRes engine More encode/ProRes engines (higher configs) What actually changes between them More cores — the M5 Pro nearly doubles CPU cores and adds GPU cores, so sustained, multi-threaded work finishe...