A few weeks back, Valve—the company most famously behind the Steam platform for PC games—announced a trio of hardware devices slated for an early 2026 release. Adding to their already-excellent portable-gaming console, the Steam Deck, they’ll be selling:
- Steam Controller: Game controller modelled off of the Steam Deck’s control scheme, most notably with 2 touchpads with haptic feedback for mouse; it also natively supports low-latency connections with other Steam-branded hardware.
- Steam Machine: Gaming PC with a custom CPU/GPU in a small, console-sized form factor.
- Steam Frame: VR headset designed for streaming games from other Steam hardware, with custom protocols to improve the fidelity of that stream.
With these consoles and accessories, Valve leapfrogs all of the console gaming industry veterans—Nintendo, Microsoft, even Sony—in the breadth of their hardware. Of course, we won’t know how well it all works until gamers get their hands on these new systems, but it’s an impressive attempt to define an ecosystem in one broad stroke. And if they’re successful, the Steam software + hardware stack can bring about major changes to gaming: accelerating Microsoft’s gradual exit from the industry, and rendering high-end gaming PCs irrelevant.
Now, skepticism is warranted. Not only is this launch ambitious, but Valve’s has a mixed track record with hardware. Other than the Steam Deck, each of these devices had a predecessor which had interesting ideas, but couldn’t gain mainstream traction. The first Steam Controller introduced the concept of haptic touchpads, but implemented them in lieu of now-standard dual analog sticks, which made for a steep learning curve. With the initial Steam Machines, Valve had partnered with PC builders to manufacture a handful of gaming boxes, but their efforts were hampered by an immature SteamOS; these systems could only play a small subset of Linux-compatible games and had to stream most Steam games from PCs running Windows. Their first VR headset was the Valve Index, which at the time was a high-end VR system, but its hardware limitations proved restrictive and was soon overshadowed by the cheaper and more convenient Meta Quest.
This time around, Valve is building on the strong, proven foundation of the Steam Deck. The new controller features the same control scheme, dual analog sticks alongside haptic touchpads. Connectivity is all wireless, with dedicated channels to maintain low latencies. SteamOS remains the embedded operating system, but it’s now much more capable—Valve’s engineers have spent over a decade improving Proton, their pseudo-emulation/translation layer that runs the vast majority of Steam library of games. The Linux-based SteamOS + Proton combination is now not only compatible with most of Steam’s games, but when compared to running some games natively in Windows 11, holds an edge in performance.
It’s also important to note the dominance of Steam. Despite competition from Microsoft’s own Store1 and the Epic Game Store2, Steam is the storefront not only for all new titles, but more importantly, the very long tail of PC games released over the past 2½ decades. There’s no equivalent gaming app store in size, scope, depth, or value. With their large catalog of old titles, Steam regularly runs sales with hefty discounts, where the software is so cheap that users don’t play the majority of the games they buy on the platform.
The business model of developing hardware to sell software follows what Sony and Nintendo have established for multiple console generations. But one key difference is Sony and Nintendo rely on blockbuster first-party titles, which are hit-or-miss and inherently risky. Steam’s collection sourced from the past 20+ years means it holds revenues steady by default, and they feel no pressure to rush their hardware or software releases. To wit, one running joke is that the company doesn’t seem to able to develop the ending trilogy for any of their popular games series.
Amid all these developments, Microsoft is in the worst strategic position. They don’t have a stable of great first-party studios comporable to Sony or Nintendo, and their attempt at differentiation, the Xbox Game Pass subscription, has faltered. Unsurprisingly, there is plenty of industry speculation about Microsoft exiting the gaming business, particularly after they closed studios, laid off employees and canceled in-progress games. The Xbox was originally conceived as an extension of Windows into the entertainment center of the home, which peaked 2 console generations ago with the Xbox 360. Since then, they’ve struggled to keep up with their competitors, even as the company itself pivoted away from Windows into cloud services and AI. In fact, had they not bought Activision-Blizzard for $75 billion a few years back, they might have already beat a hasty retreat34.
The other potential casualty with Valve’s new hardware push is high-end PC gaming. The success of the Steam Deck has prompted other PC makers to release their own portable gaming devices with similar specs. If the Steam Machine gets popular enough, we can expect others to also build small form factor gaming PCs with console-like performance and convenience, possibly running SteamOS as well. More importantly, their popularity means that game developers can narrow their range of hardware to optimize for.
This is already happening with the Steam Deck. For instance, 2025’s Game Awards features six Game of the Year nominees: one PS5 exclusive, one Switch 2 exclusive, and four multi-platform games—all playable on the Steam Deck, and three playing on the portable well. Game developers aren’t targeting high-end gaming PCs; they’re using the PS5 and Xbox Series X/S as reference hardware, and the Steam Machine’s similar specs triples down on the performance bracket. If anything, games are now built to step down onto lower-end hardware, to run reasonably well on the Steam Deck and other portable consoles. High-end PC gaming has always been nonsensical from a value standpoint, but we’re reaching the limits of screen resolutions and frame rates perceptible to human eyes, and it’s not clear that it’s worth the developer effort to build features that fewer and fewer players could enable.
I, for one, welcome these developments to the gaming industry. The past few console cycles—since around the PS3—have felt stagnant, with the incremental hardware refinements each generation, but far from the days of 2D → 3D or even 16-bit → 64-bit processing. AAA titles are so costly to make and take so long that we instead repurchase remasters of previous-generation classics and play-it-safe sequels. Sony and Nintendo have settled onto identical playbooks, and Microsoft would have followed—had they kept more of their first-party studios. Valve and Steam’s approach to game development and promotions is a challenge to this entrenched model, and I’ll be looking to see if they gain enough momentum to influence the industry’s norms—particularly around prices, compatibility, and modularity.
Installed by default on PCs, and pushed by the OS shamelessly upon setup.↩
Which has been trying to buy its way into relevance by giving away loads of free games.↩
Funnily enough, this isn’t even the most famous too-little-too-late acquisition by the company; they bought Nokia in 2014 for $7 billion to try to catch up to iPhone/Android and wrote it off only 2 years later.↩
This will also perhaps teach another lesson in the value of failing fast.↩