Valve Debunks Speculation on RDNA4 Steam Console: Regularly Tests Unused Hardware

According to a recent report by GamingOnLinux, rumors about a new Steam Console being in the works were debunked. Valve developer Pierre-Loup Griffais confirmed on BlueSky that the speculation was unfounded. Griffais explained that the work being done with pre-release Mesa Vulkan on AMD’s RDNA 4 architecture is consistent with the company’s ongoing efforts since the AMD Vega architecture. Although Valve is indeed testing software updates for an upcoming GPU architecture, it doesn’t mean they’re resurrecting the Steam Machine project anytime soon.

So, for those hoping for a new Steam Console, it’s a bit of a letdown for now. But it’s worthwhile to delve deeper into Valve’s past ventures into the console space, and explore what’s possible with a revived Steam Machine or even the anticipated Steam Deck 2.

What We Know About Valve’s Hardware Strategy

Detecting any signs of a forthcoming Steam console based solely on AMD’s RDNA 4 GPU work wouldn’t be fruitful. Typically, consoles, including Valve’s Steam Deck, are equipped with custom hardware derived from established architectures. When a console hits the market, its components are usually one or two generations behind the cutting-edge technology. This stems from the extensive planning required for such platform launches.

When the Steam Deck debuted in 2022, it utilized AMD’s RDNA 2 architecture, originally introduced in November 2020. This created a gap of approximately a year and a half from the Deck’s release to its GPU architecture and over two years between the Deck and the Zen 2 CPU architecture. AMD’s RDNA 4 is gearing up for desktops, and incorporating it into a feasible handheld APU would require significant resources.

Today’s leading integrated GPUs are based on AMD’s RDNA 3 and RDNA 3.5 architectures, though they haven’t shown remarkable performance improvements while on battery power. While enhanced plugged-in performance is achievable, battery restrictions are a limiting factor for handhelds. The Ryzen AI HX 300 Series APUs with RDNA 3.5 iGPUs are a recent development.

Another clear indicator that a Steam Console isn’t a priority for Valve is the confirmed timeline for the Steam Deck 2, which is expected to be released in at least two to three years. Valve wants to wait for substantial advancements in hardware. Since consoles aim for a significant generational leap between models, an RDNA 4 iGPU for the Steam Deck 2 is conceivable but unlikely anytime soon. Launching three different SteamOS hardware platforms in just under seven years seems improbable.

Despite the allure of RDNA 4, much about it remains unknown—its power consumption, performance, and pricing are crucial details yet to be uncovered. It’ll be a while before any handheld PCs with this technology come to fruition.

Forecasting Valve’s Future Moves

As we venture into the world of speculations while staying grounded in the known aspects of PC hardware, we can try to envision the direction of the handheld console market. Towards the end of last year, credible leaks surfaced depicting Valve’s new designs, including a revamped Steam Controller with a design inspired by the Steam Deck and next-gen VR controllers. Does this suggest a possible revival of the Steam Link set-top box or the Steam Machine initiative? Perhaps, but it might just as likely be aimed at enhancing parity between the Steam Deck’s handheld and docked play modes, catering to users keen on Steam Input’s advanced features like touch-sensitive gyro support.

Could a Steam Console be entirely out of the question? Maybe not. It’s plausible that with the broader rollout of Steam OS 3, more manufacturers might embrace it for mini PCs, laptops, handhelds, etc. A larger device format could facilitate the incorporation of features like real-time ray-tracing. With current Deck hardware, achieving 30 FPS in games with ray-tracing is viable only through mostly rasterized graphics, with complete RT or path-traced games being off the table due to outdated hardware and low power limits.

For Valve to achieve a significant boost in handheld performance, more than RDNA 4’s improved RT capabilities is needed. A GPU architecture capable of delivering those RT results at a 25W or 15W TDP is essential. Current estimates suggest full discrete RDNA 4 mobile GPUs operate at 80-175W TDP, which shows AMD isn’t there just yet.

However, AMD’s TSMC N4 process node, dated two years back, offers hope. The Steam Deck’s Van Gogh APU is based on what’s now seen as an older N7 node. Transitioning to N5 or N4 could offer significant improvements, and within a couple of years, N3 and N2 nodes should become more accessible. Smaller nodes optimized for efficiency could be what’s needed for a compelling future handheld chip—provided that die size and pricing are reasonable.

Do we even need a standalone Steam Console? The Steam Deck paired with a compatible docking station already surpasses the Nintendo Switch in power. If Nintendo has shifted away from traditional home consoles, a company like Valve might find more value in a hybrid-handheld approach, steering clear of direct competition with conventional consoles and gaming PCs.