ARM “notebook-class performance” new processor

ARM announced the ARM Cortex-A77 processor for mobile, the Mali-G77 GPU, and the ARM ML processor, an NPU for machine learning.

Cortex-A77 is a CPU using 7nm manufacturing process. More than 20% integer arithmetic performance, more than 35% floating point performance and 15% more memory bandwidth than previous Cortex-A76 models. In addition, instructions per clock (IPC) increased the number of execution instructions per clock by 20%. The machine learning performance is 35 times more than the Cortex-A75.

According to ARM, the Cortex-A77 is comparable to current mainstream notebooks within the smartphone power range. In addition, it can be used for functions such as AI camera, 3D scan, face recognition, voice recognition, etc. that require machine learning ability, and 8K resolution streaming and mobile games using 5G.

Mali-G77, a mobile GPU, has adopted a new architecture called Valhall. The performance of the Mali-G77 is 30% higher than its predecessor, the G76, and the performance per size is 30% higher. Overall performance is improved by 40%. Machine learning performance is also expected to increase by 60%.

ARM explains that the performance of Mali-G77 can be used to enjoy mobile games such as battlegrounds and port nights for a long time and comfortably.

The ARM ML processor, which is a machine learning processor, is 5TOPS more energy efficient than the NPU on the Project Trillium announced in 2018, and it has triple the memory compression technology and 32TOPS maximum performance. For more information, please click here .



Through the monthly AHC PC and HowPC magazine era, he has watched 'technology age' in online IT media such as ZDNet, electronic newspaper Internet manager, editor of Consumer Journal Ivers, TechHolic publisher, and editor of Venture Square. I am curious about this market that is still full of vitality.

Add comment

Follow us

Don't be shy, get in touch. We love meeting interesting people and making new friends.

Most discussed

%d 블로거가 이것을 좋아합니다: