Arm announces new CPU and GPU chipsets designs, Mali-G77 GPU, Cortex-A77 CPU, and much more!

Cortex-A77 CPU

With every new generation of Arm CPUs, the Cortex A77 promises to be more power efficient and provide better raw processing performance. Cortex-A77 has been built to fit in smartphone power budgets and for improving performance. It is the second generation design that brings in substantial performance upgrade over Cortex-A76. Cortex A77 has been built for next-generation laptops and smartphones and for supporting upcoming use cases like advanced ML. It will also support the range of 5G-ready devices that are set to come to the market following the 5G rollout in 2019.

Due to the combination of hardware and software optimizations, the Cortex-A77 now brings better machine learning performance. It comes with more than 20 percent integer performance, more than 35 percent FP performance and more than 15 percent more memory bandwidth improvements.

Mali-G77 GPU

The company brings the new Mail-G77 GPU architecture, which is the first one to be based on the company’s Valhall GPU design. It offers around 1.4x performance improvement over the G76.

Mail-G77 GPU is also 30 percent more energy efficient and 60% faster at running machine learning inference and neural net workloads. Mali-G77 provides uncompromising graphics performance and brings performance improvements to complex AR and ML for driving future use cases.

https://twitter.com/Arm/status/1132992854282915841

Machine learning processor

Arm already offers Project Trillium, its heterogeneous machine learning compute platform for the machine learning processor. Arm has improved the energy efficiency by 2x and scaled performance up to 8 cores and 32 TOP/s since the announcement of Trillium last year.

The machine learning processor is based on a new architecture that targets connected devices such as augmented and virtual reality (AR/VR) devices, smartphones, smart cameras, and drones, as well as medical and consumer electronics. This processor processes a variety of neural networks such as convolutional (CNNs) and recurrent (RNNs), for image enhancements, classification, object detection, speech recognition, and natural language understanding. It also minimizes system memory bandwidth through various compression technologies.

Read Also: Snips open sources Snips NLU, its Natural Language Understanding engine

The company announced, “Every new smartphone experience begins with more hardware performance and features to enable developers to unleash further software innovation.”

The company further added, “For developers, the CPU is more critical than ever as it not only handles general-compute tasks, as well as much of the device’s ML, compute which must scale beyond today’s limits. The same holds true for more immersive untethered AR/VR applications, and HD gaming on the go.”

To know more about this news, check out Arm community’s post.

Intel discloses four new vulnerabilities labeled MDS attacks affecting Intel chips

The Linux Foundation announces the CHIPS Alliance project for deeper open source hardware integration

AI chipmaking startup ‘Graphcore’ raises $200m from BMW, Microsoft, Bosch, Dell