A surprising new competitor in the GPU upscaling wars has entered the ring

By Kunal Khullar Published July 11, 2024

Bistro rendering benchmark done using Arm's Accuracy Super Resolution technology. — Arm

Arm is stepping into the graphics upscaling arena with its new Accuracy Super Resolution (ASR) technology, promising significant enhancements for GPU performance on mobile devices.

Recommended Videos

ASR is a temporal upscaler based on AMD’s open-source FidelityFX Super Resolution 2 (FSR2). This compatibility allows developers to leverage familiar API and configuration options, streamlining integration. AMD’s FSR and Nvidia’s DLSS have primarily boosted graphical fidelity on gaming PCs, enabling higher frame rates and rendering capabilities. Arm’s focus, however, is on the power savings achievable through reduced GPU use, which is crucial for mobile devices where thermal throttling can degrade performance.

Peter Hodges, Arm’s Director for Ecosystem Strategy, highlighted ASR’s ability to maintain high-quality visuals at lower temperatures, reducing thermal throttling. This results in a more stable gaming experience and prolonged device engagement. ASR’s optimization for lower memory bandwidth devices, like mobile GPUs with integrated graphics, is a key differentiator. Arm claims ASR uses only a third of FSR2’s GPU resources at equivalent resolutions and scaling factors, delivering 20% to 40% better frame rates on Arm devices.

GPU Performance Analysis of FSR2 and Arm ASR — Arm

Arm used a specific GPU at 2x upscaling resolution. ASR boosts frame rates by 53%, compared to FSR2’s 36%. Tests with MediaTek’s Dimensity 9300 chip reveal that 2x ASR reduces power consumption by over 20% versus native 1080p rendering. This results in longer battery life and reduced overheating, addressing key concerns for mobile gamers.

Once the ASR specification is released, it will be up to OS makers and game developers to adopt it. Apple is expected to continue using its MetalFX upscaling technology, which also stems from AMD’s FSR. Microsoft is promoting Automatic Super Resolution for its Qualcomm-based Arm devices and is working on a vendor-agnostic upscaling API called DirectSR. Qualcomm also introduced its Snapdragon Game Super Resolution just over a year ago.

While this hardware-agnostic technology promises superior image quality and performance on a broad range of devices, it is too early to say if it can set a new standard in mobile graphics upscaling.

Please enable Javascript to view this content

Topics

Computing Writer

Kunal Khullar is a computing writer at Digital Trends who contributes to various topics, including CPUs, GPUs, monitors, and…

Computing

This fascinating new GPU completely ditches fans

ASRock Radeon RX 7900 XTX Passive 24GB graphics card.

We may still be in limbo when it comes to next-gen graphics cards, but ASRock just launched a couple of interesting AMD GPUs. The ASRock RX 7900 XTX Passive and the RX 7900 XT Passive are two of AMD's most powerful consumer graphics cards, but this time around, they're completely fanless. That's right -- without an active cooling solution, these cards rely entirely on passive cooling.

At a glance, no one would've said that the cards in question are part of AMD's RX 7000 series. After all, the RX 7900 XTX normally features three fans, but this one comes with zero, and it's also significantly thinner than ASRock's other versions of the RX 7900 XTX. Both cards are now just dual-slot, and they're both shorter and thinner than their air-cooled counterparts.

Computing

Nvidia may be working on a surprising new budget GPU

Nvidia GeForce RTX 3050 graphics card.

This is certainly unexpected. Hints of an upcoming Nvidia graphics card showed up in the PCI-ID database, and it's definitely not what you think. Instead of working on the RTX 50-series or refreshing a 40-series GPU, Nvidia is bringing back the slowest RTX 30-series card -- the RTX 3050 -- but with an Ada Lovelace chip. In theory, that could make this the first RTX 30-series GPU to have access to Deep Learning Super Sampling 3 (DLSS 3).

The GPU in question is a laptop version of the RTX 3050. It's hard to say why Nvidia would choose to release new versions of that particular card in 2024 (or beyond), but the RTX 3050 A exists -- VideoCardz found traces of it in the latest Nvidia drivers. Just the fact that there's an RTX 3050 in the works is surprising, but that it's based on the AD106 chip is what baffles me.

Computing

AMD’s multi-chiplet GPU design might finally come true

RX 7900 XTX installed in a test bench.

An interesting AMD patent has just surfaced, and although it was filed a while back, finding it now is all the more exciting because this tech might be closer to appearing in future graphics cards. The patent describes a multi-chiplet GPU with three separate dies, which is something that could both improve performance and cut back on production costs.

In the patent, AMD refers to a GPU that's partitioned into multiple dies, which it refers to as GPU chiplets. These chiplets, or dies, can either function together as a single GPU or work as multiple GPUs in what AMD refers to as "second mode." The GPU has three modes in total, the first of which makes all the chiplets work together as a single, unified GPU. This enables it to share resources and, as Tom's Hardware says, allows the front-end die to deal with command scheduling for all the shader engine dies. This is similar to what a regular, non-chiplet GPU would do.