The discussion highlights that the Steam Deck 2 is expected to feature a significant generational hardware leap, likely leveraging AMD’s next-gen RDNA5 architecture and maintaining a 15-watt power envelope to balance performance and portability. Additionally, advancements like 1080p displays and machine learning technologies such as DLSS are anticipated to enhance the handheld’s gaming experience, though overcoming memory bandwidth limitations remains a key challenge.
The discussion centers around the anticipated hardware advancements for a potential Steam Deck 2, especially in light of recent handheld announcements like the Strix Halo projects. The consensus is that while current handhelds are seeing iterative performance improvements at similar power levels, Valve aims for a significant generational leap rather than modest upgrades. Rumors about Sony’s upcoming handheld, possibly featuring RDNA5 or a similar next-gen architecture, suggest that such advancements could influence Valve’s approach to the Steam Deck 2.
Oliver highlights that the existing RDNA2 and RDNA3 architectures, along with their incremental updates, do not present a compelling upgrade path for Valve at the 15-watt power envelope typical for handheld devices. He believes Valve is likely to wait for next-generation architectures, such as RDNA5, to deliver a meaningful performance boost while maintaining power efficiency. The Strix Halo hardware, although powerful, requires higher wattages, making it unsuitable for a Steam Deck 2 without major design changes that Valve seems reluctant to undertake.
Tom adds to the conversation by emphasizing that the Steam Deck 2 will probably align with the Switch 2 in terms of display resolution, aiming for a 1080p screen, which is becoming the standard for handhelds. He agrees that maintaining a 15-watt power target is crucial to avoid extreme design compromises that could impact user comfort and device practicality. Tom also concurs that Valve will likely wait for AMD’s next fabrication process and chipset innovations to define the next Steam Deck iteration.
The conversation also touches on the broader context of generational improvements in handheld gaming hardware. Despite advancements in semiconductor fabrication—from 7nm to 4nm processes—there hasn’t been a dramatic performance leap in handheld APUs, largely due to limitations such as memory bandwidth. This bottleneck has constrained the potential gains from process shrinks and increased compute power, indicating that future improvements may rely on solving these architectural challenges.
Finally, the speakers note the growing importance of machine learning features like DLSS (Deep Learning Super Sampling) seen in devices like the Switch 2. Such technology could become a significant factor in enhancing performance and visual quality on PC handhelds like the Steam Deck 2. Overall, while concrete details remain scarce, the outlook for the next generation of Steam Deck hardware appears promising, with expectations for meaningful advancements once next-gen architectures and supporting technologies mature.