Arm Details Its Mobile‑Centric AI Reconstruction Approach, Diverging from the "Black‑Box" Model of DLSS 5

Arm, the UK‑based semiconductor and software design firm, staged its Arm Everywhere China showcase in Shanghai to announce a suite of fresh AI‑computing advances aimed squarely at mobile devices. The centerpiece of the announcement is a comprehensive AI‑native ecosystem that includes the second‑generation AI‑native compute platform CSS for Mobile, as well as a brand‑new graphics processor, the Mali G2‑Ultra NX. This GPU is built around neural‑processing blocks and promises to lift visual performance while keeping thermal output and power draw dramatically lower than traditional solutions. Arm says the on‑device AI capabilities of the Mali G2‑Ultra NX will boost graphics through a method it describes as more transparent than Nvidia’s contentious DLSS 5, and more in line with what developers originally envisioned for real‑time rendering.

While the company did not disclose the exact devices that will ship with the new chip, the predecessor Mali G1‑Ultra already powers phones from Xiaomi, Vivo and OnePlus, suggesting a similar market penetration for the next generation. Deyan Lazarov, a senior member of Arm’s GPU product team within the Edge AI division, explained that the Mali G2‑Ultra NX has been in development for roughly two and a half years.

Its primary goal is to give mobile developers access to compute‑heavy features such as ray tracing, which has historically been limited on smartphones. “We’re seeing a growing adoption of ray tracing in mobile titles,” Lazarov noted. “It was slow to get started, but now more games are looking to enable it.” The new GPU integrates tightly with Unreal Engine, allowing developers to leverage technologies like MegaLights.

Arm partnered with Sumo Digital to produce a demo title called *Neural Dawn*, which showcases the chip’s ability to render high‑quality visuals on a phone‑form factor. Lazarov emphasized that the chip targets games that are "mobile‑first but want desktop‑level fidelity," as well as cross‑platform experiences where a player expects a seamless visual quality whether they are on a console at home or a handset on a train. Among the first titles announced to run on the Mali G2‑Ultra NX are Tencent’s *Arena Breakout Infinite*, NetEase’s *Where Winds Meet* and Infold Games’ *Infinity Nikki*. Lazarov highlighted the unique constraints of phones: unlike consoles, they cannot dissipate large amounts of heat and must operate continuously on battery power.

To address this, Arm’s solution blends neural processing with the conventional rendering pipeline, off‑loading a substantial portion of the GPU’s workload to dedicated neuro‑accelerators. This not only cuts the GPU’s power consumption but also improves overall system efficiency, giving developers more headroom to pursue visual fidelity, performance, or battery life without making hard trade‑offs. Arm is also taking an open‑source stance to encourage adoption.

The company has released its AI models—Neural Super Sampling (NSS), Neural Super Sampling and Denoising (NSSD) and Neural Frame Rate Upscaling (NFRU)—on public repositories such as GitHub and Hugging Face. Lazarov explained that these are convolutional networks, not generative ones, meaning they do not hallucinate new content but rather learn how existing content should look. Because the models are open, developers can fine‑tune them with data from their own games, preserving artistic intent and intellectual‑property control. This contrasts sharply with Nvidia’s DLSS 5, which Lazarov described as a closed, black‑box system that cannot be retrained or inspected.

The mobile‑first design of Arm’s approach also means the networks are built to respect the tight power, bandwidth and memory budgets of smartphones. DLSS, by contrast, would quickly exceed those limits on a phone, even if the software were technically compatible. Arm reports that the Mali G2‑Ultra NX delivers up to four times the performance and efficiency of its predecessor, while slashing DRAM traffic by roughly 70 %, a major source of power draw on mobile SoCs.

In practical terms, the performance gains are tangible. When pushed to its limits, the previous‑generation GPU could only run *Neural Dawn* at about 40 fps under ideal conditions and fell to 15 fps in typical use.

The new Mali G2‑Ultra NX can comfortably sustain 60 fps, delivering a smoother experience without sacrificing battery life. From a developer tools perspective, Arm currently offers plug‑ins for Unreal Engine and plans to expand support to other engines, with Unity integration on the roadmap. A recent partnership with Unity China signals that broader engine coverage is forthcoming. For studios using custom engines, Arm provides an SDK that streamlines integration, allowing teams to embed AI‑native graphics without writing low‑level code.

Lazarov stressed that existing games can also benefit from the new technology. Porting a shipped title to incorporate Arm’s AI‑enhanced pipeline can take as little as three months, enabling rapid upgrades in visual quality or efficiency.

The *Neural Dawn* demo itself was built by a sub‑20‑person team in under 18 months, illustrating how Arm’s tools can accelerate development cycles. In summary, Arm’s mobile‑first AI reconstruction strategy offers a transparent, open, and power‑aware alternative to Nvidia’s DLSS 5.

By marrying neural accelerators with conventional rendering, providing publicly available models, and delivering robust developer tooling, Arm aims to empower game creators to bring console‑grade graphics to handheld devices without the compromises that have traditionally hampered mobile gaming.