ARM Announces New Chip Designs for Automotive and Industrial Use

0
42
A photographer takes pictures in front of a logo showing the ARM, one of the main chip designer in the world from the UK, in Taipei on May 27, 2019. (Photo by Sam YEH / AFP) (Photo credit should read SAM YEH/AFP/Getty Images)

INTRO: Ahead of its impending $40 billion acquisition by US chip giant, Nvidia, British semiconductor firm, ARM, has announced new ‘safety capable’ computing solutions aimed toward a spread of autonomous systems, including vehicles and industrial machinery. These include the Cortex-A78AE CPU, the Mali-G78AE GPU and therefore the Mali-C71AE ISP. during a handout , the corporate claimed that its new chips would “accelerate autonomous decision-making with safety capability across automotive and industrial applications”.

ARM Cortex-A78AE CPU

Described by ARM as its “highest performance CPU with safety”, the Cortex-A78AE comes with the company’s new hybrid technology which allows the cores to run independently in either split mode or in lock mode. While the previous features a low safety requirement, the latter has higher security but slows down performance. However, the corporate says that the chip offers a 30% increase in performance compared to its predecessor.

ARM Mali-G78AE GPU

The Mali-G78AE GPU, described by ARM as its “first safety capable GPU”, comes with four fully independent partitions for workload separation for safety use cases. consistent with the corporate , it might enable GPU resources to be utilized for safety-enabled human machine interfaces or for the heterogeneous computing needed in autonomous systems.

ARM Mali-C71AE ISP

Like its CPU and GPU twins, the Mali-C71AE image signal processor also comes with attention on safety. consistent with the corporate , the chip offers the pliability needed to support both human and machine vision applications like , assembly line monitoring and ADAS camera systems. It also comes with enhanced safety features, making devices compliant with ASIL B / SIL2 safety standards. It offers support for four real-time cameras, or 16 buffered cameras, delivering a 1.2 giga pixel per second throughput.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.