Nvidia could learn a thing or two from Nintendo on how to have a good launch

The Nintendo Switch 2 launched just last week, and it seems, by all metrics, to be an initial success. While we’ve seen this in the past with the Nintendo 3DS, famously having a very strong launch only for sales figures to drop off within months, the follow-up to the wildly successful Switch has been phenomenal so far. It’s actually had me thinking about past launches, most notably Nvidia’s RTX 50 series, and Nvidia should be taking notes. Love or hate Nintendo, getting consoles into the hands of practically every gamer who wanted one on launch day is an impressive feat.

As it turns out, and this will be very relevant later in the article, the Nintendo Switch 2 potentially leaked years ago, and theories have circulated that the purported “Switch Pro” with its bumped-up specs was actually the Switch 2. This has been supported by Digital Foundry, who have been right on the money for practically every Switch 2 leak over the last year, right down to the T239 chip powering it. It’s a significant spec bump over the original Switch, and on a personal level, I’ve been absolutely loving ​​​​​it.

So, how did Nintendo manage to get so many consoles in the hands of consumers and effectively “beat” scalpers? If I were to purchase a Switch 2 here in Ireland, I’d receive it in two days, and Best Buy in the U.S. lists it as being in stock right now ​​​​​for in-store pick-up. Shortages have affected every major PC launch of the last few years, and even the PlayStation 5 went through years of difficulties and stock shortages.

Unsurprisingly, the answer is nuanced, but the crux of the matter is that at least some of what Nintendo did could also be done by Nvidia.

Nintendo chose to build up stock instead of avoiding leaks

Though there were other advantages Nintendo had, too

Eight months ago, we started to see our first CAD renders of the console, stated to be made from “factory prototypes.” These renders turned out to be pretty accurate, which suggests that production began around the same time. It would make sense, too; practically every detail about the console dripped out over the next few months following that initial leak, with detailed console specifications shortly following as well. Given that we know Nintendo stockpiled a huge number of consoles (Bloomberg reports 750,000 imports from Vietnam to the U.S. in February 2025 alone), stock was clearly a priority.

With that said, leaks are a major downside of production at this scale. Thousands of individuals were likely building these devices, and at that stage, it was impossible to stop leaks from flowing. This was then worsened by the distribution of units to stores early in anticipation of launch; great for retailers and consumers, but not so much for the company that had most of its marketing material taken from its grasp. You can’t “tease” the design of a product or even its launch when everything about it has already been known for months ahead of time, especially when all of the credible and established leakers in the space corroborated what was now in the public eye well ahead of its expected schedule.

However, I’d argue this was a choice by Nintendo, and one I’m curious to see the results of in the wider industry. The day-one update, required to go online and play original Switch titles, felt like a choice by Nintendo to prevent too much getting out. The device could be photographed, the box could be photographed, but the games, which ultimately are what sells a console, were (for the most part) not playable at all, on account of the requirement for its day one update. In a way, even Nvidia appears to be playing with this idea, where RTX 5060 drivers were only selectively provided to certain members of the media ahead of its release.

There’s one other key piece of the puzzle, though, that gave Nintendo an advantage. Production of the console itself isn’t where Nintendo began. In fact, the Nvidia Tegra T239, which powers the console, was reportedly produced in 2021, according to Digital Foundry. This timeline certainly checks out, as known leaker kopite7kimi stated in June 2021 that Nintendo had ordered a modified version of the T234, dubbed T239. That’s a four-year head start on the silicon powering it, and wholly unrealistic for the likes of Nvidia to achieve when it comes to cutting-edge technology.

Yet, there’s some merit in saying that it’s not totally impossible to have more stock at launch for the biggest releases of the year. YouTuber Moore’s Law is Dead stated that fewer than 1,000 RTX 5090 cards had been allocated to the entire U.S. market, which certainly sounds plausible given how quickly the entire stock of the RTX 50 series evaporated in minutes. Nintendo benefits from a multi-year head start and a less sophisticated fabrication node, but Nvidia also admitted in October 2024 that it had made a mistake when it came to Blackwell production, which saw shipments delayed at the time. It was an unusually short turnaround, but then again, stock issues have been plaguing high-end tech for a long time now, with causes attributed to chip shortages and shipping difficulties.

However, it all boils down to one simple fact: Nintendo had a four-year head start… and Nvidia had approximately three months. Sure, it’s not exactly feasible to expect Nvidia to stockpile silicon for four years, but a longer lead-in time seems only natural in order to build up appropriate stock levels for consumers… though let’s face it, every silicon wafer turned into a GB202 for the RTX 5090 could have been used to build a high-end enterprise GPU that sells for significantly more. Plus, partners that produce Nvidia GPUs make other products too, whereas Nintendo’s only high-tech production, really, is the Switch 2.

With that said, all of these roadblocks for the likes of Nvidia undoubtedly have a positive effect on preventing leaks and gives Nvidia a quick turnaround from announcement to launch, regardless of stock, and the same could be said for Apple, which reportedly begins producing iPhones right before they’re shipped to consumers… so, after they’ve already been announced. Sure, these practices cut down on leaks (something Apple seems particularly keen on preventing), but they enable scalpers and predatory pricing practices that ultimately harm consumers in the end. Scalpers aren’t so much a problem when it comes to Apple, but some of its launches can see consumers waiting weeks to get their devices.

Surely there’s a middle ground here?

Stockpiling tech seems to be the way to go

And I hope companies learn from it

There are two ways to avoid shortages: produce fast enough to meet demand, or stock up in anticipation of demand. The most cynical of consumers will likely make the argument that shortages benefit Nvidia and GPU partners as a result, as they can charge higher prices, which is true to a point, but those shortages have PC enthusiasts looking to alternatives. It’s also completely true to say that Nvidia still holds a significant majority of the desktop GPU market, but even titans lose their footing on pebbles. These seemingly “small” inconveniences can grow and fester, and for a modern example, Intel’s decline is proof of that.

One of the most anticipated consoles of the last few years, and one that appears to be the best-selling console ever in week one sales, managed to completely avoid stock issues and enabled the vast majority of consumers to get one on day one. No matter how Nintendo did it, and no matter how much of the process is seemingly infeasible for competitors to pull off in the same way, there are certainly lessons to be learned from Nintendo’s approach to stock that enabled paying customers and fans to simply go out and buy a console in a way that has seemed impossible for any modern tech in over a decade.

Personally, I can’t remember the last time I was able to just… buy a highly-anticipated product at launch and have it arrive on the release day. Cutting-edge tech will obviously be more difficult to achieve a similar level of production ahead of launch, but you also don’t need millions of RTX 5090s for a launch. Just having more than 1,000 for a country of 340 million people would be a great start.