Arm Details Its Mobile‑First AI Reconstruction Approach, Distinct from DLSS 5’s Black‑Box Method

Arm, the British semiconductor and software design powerhouse, recently staged its Arm Everywhere China showcase in Shanghai, unveiling a suite of fresh AI‑computing offerings aimed at reshaping the mobile graphics landscape. Central to the announcement is a comprehensive AI‑native ecosystem that includes the next‑generation compute platform CSS for Mobile 2 and the brand‑new Arm Mali G2‑Ultra NX GPU.

This GPU is built around neural‑accelerated processing, promising a noticeable lift in visual fidelity while keeping heat and power draw to a minimum. Arm positions its on‑device AI solution as a more transparent alternative to Nvidia’s contentious DLSS 5, aligning more closely with developers’ original artistic intent.

While Arm did not disclose specific product partners for the Mali G2‑Ultra NX, its predecessor, the Mali G1‑Ultra, already powers devices from manufacturers such as Xiaomi, Vivo and OnePlus. Deyan Lazarov, a senior member of Arm’s GPU product team within the Edge AI division, explained that the new chip has been under development for roughly two and a half years.

Its purpose is to give mobile developers access to compute‑heavy techniques—most notably real‑time ray tracing—that have traditionally been the domain of high‑end consoles and PCs. "We’re seeing a steady rise in ray‑tracing adoption on smartphones," Lazarov said. "The technology was initially slow to take off, but today more titles are actively exploring it." GamesIndustry.biz reported earlier this year that the Mali G2‑Ultra NX enables developers to tap into Unreal Engine features like MegaLights.

Arm also collaborated with Sumo Digital to produce a demo title, *Neural Dawn*, which showcases the chip’s capabilities. Lazarov described the target audience as "mobile‑first games that still want desktop‑grade visual quality" as well as cross‑platform titles that aim for a seamless experience across consoles, PCs and phones. He illustrated the use‑case: a player could enjoy a high‑quality session on a home console and then continue the same experience on a train without a noticeable drop in graphics or performance.

Among the upcoming games that will leverage the new GPU are Tencent’s *Arena Breakout Infinite*, NetEase’s *Where Winds Meet* and Infold Games’ *Infinity Nikki*. Lazarov highlighted the unique constraints of smartphones: unlike consoles, phones operate with limited thermal headroom and rely on battery power, meaning they cannot dissipate heat or draw energy at the same scale. Arm’s answer is to fuse neural processing directly into the traditional rendering pipeline. By offloading parts of the workload to dedicated neural accelerators, the GPU’s power consumption drops, which in turn improves overall system efficiency.

"That extra efficiency removes many of the compromises developers usually have to make," Lazarov explained. "It frees them to allocate resources toward higher visual fidelity, smoother frame rates, longer battery life, or any combination thereof." To foster adoption, Arm is taking an open‑software stance, releasing a toolbox of SDKs and plugins. Over the past two years the company has worked closely with the broader ecosystem to make AI‑enhanced graphics developer‑friendly.

The core technologies include Neural Super Sampling (NSS), Neural Super Sampling and Denoising (NSSD) and Neural Frame‑Rate Upscaling (NFRU). In practice, these techniques can upscale lower‑resolution images, generate additional rays for better lighting, and even synthesize new frames through AI reconstruction.

Lazarov illustrated the process with a two‑frame example. In the first frame, the GPU renders only a quarter of the image while the neural accelerator reconstructs the remaining three‑quarters via NSS. In the second frame, the GPU does no traditional rendering at all; every pixel is produced by the neuro‑accelerators, resulting in a fully AI‑reconstructed frame.

Nvidia’s DLSS 5 also uses AI to add detail, but it has attracted criticism for its "black‑box" nature and potential divergence from a developer’s artistic vision. Arm draws a clear line between the two approaches. "Our networks are not generative; they’re convolutional, meaning they don’t hallucinate new content—they simply know how the final image should look based on real data," Lazarov said.

"We train them on authentic content and release the models publicly on GitHub and Hugging Face." Because the models are open, developers can fine‑tune them with data from their own games, preserving unique art styles and protecting intellectual property. In contrast, DLSS’s proprietary models cannot be retrained or inspected, leaving developers with a take‑it‑or‑leave‑it proposition.

Another differentiator is the mobile‑first design philosophy. Arm’s networks are built to respect the tight power, bandwidth and thermal budgets of smartphones. "If you tried to run DLSS on a phone, you’d instantly exceed those limits," Lazarov warned. Power efficiency is a cornerstone of the Mali G2‑Ultra NX.

Arm reports up to a four‑fold boost in performance and efficiency, alongside a 70 % reduction in DRAM traffic—a major source of power draw on mobile devices. Less memory access translates directly into longer battery life.

To illustrate real‑world gains, Lazarov compared the new GPU to its predecessor when running *Neural Dawn*. The older chip could only sustain 40 fps under ideal conditions and dropped to 15 fps in typical use, whereas the G2‑Ultra NX comfortably hits 60 fps. On the tooling side, Arm currently offers plugins for Unreal Engine and plans to expand to other engines, with Unity support on the roadmap. A recent partnership with Unity China signals that broader engine integration is forthcoming.

For studios using custom engines, Arm provides an SDK that streamlines integration without requiring a dedicated plugin. Lazarov emphasized that even legacy titles can benefit quickly. "Porting our AI‑native graphics to an existing shipped game can take as little as three months," he noted.

Finally, he pointed out that *Neural Dawn* was built by a team of fewer than 20 developers in under 18 months, underscoring how Arm’s technology can accelerate development cycles and enable smaller teams to deliver visually impressive, efficient games. This open, mobile‑centric AI reconstruction strategy marks a significant departure from the closed, console‑centric approaches of competitors, promising a new era of high‑quality, power‑aware gaming on smartphones.