institutional access

You are connecting from
Lake Geneva Public Library,
please login or register to take advantage of your institution's Ground News Plan.

Published loading...Updated

Arm juices mobile GPUs with neural tech for better graphics

Summary by The Register
: Designs scheduled for launch in 2026, developer kit for programmers out today

14 Articles

A little more than a year ago, in July 2024, ARM presented the ASR, the acronym for Accuracy Super Resolution. It was a transposition of the FSR 2 of AMD to mobile chips...

Read Full Article

Arm introduces a major step forward by integrating a Neural Unit dedicated to mobile GPUs, accompanied by an IA technology called Neural Super Sampling. This innovation promises to double the resolution of images while drastically reducing the work of GPUs, with a strong impact on mobile games and beyond.

Historically positioned as a provider of processor architectures in the mobile, Arm extends its influence to demanding environments requiring high graphic performance. With the integration of neuronal accelerators in its GPUs, the company is taking a strategic step that opens up concrete perspectives for robotics, edge computing and critical systems. From graphic power to local embedded intelligence, [...] The post Arm is heading towards neurona…

During the SIGGRAPH, Arm unveiled a major innovation for the mobile world: the integration of neural Accelerators dedicated directly to its GPUs from 2026 onwards. This technology, called Arm Neural Technology, promises up to 50% GPU load reduction for the most demanding mobile content, especially video game, while preserving autonomy and fluidity. [...] Read more

Think freely.Subscribe and get full access to Ground NewsSubscriptions start at $9.99/yearSubscribe

Bias Distribution

  • 100% of the sources are Center
100% Center

Factuality 

To view factuality data please Upgrade to Premium

Ownership

To view ownership data please Upgrade to Vantage

HotHardware broke the news in on Tuesday, August 12, 2025.
Sources are mostly out of (0)