Chip designer Arm launched a new generation of products for smartphones: the Arm Cortex-A77 CPU, Arm Mali-G77 GPU, and Arm Machine Learning processor. The Cortex-A77 delivers advanced machine learning and AR/VR experiences thanks to a 20 percent IPC performance improvement over Cortex-A76 devices, the company said. The Mali-G77 features the new 'Valhall' architecture, delivering a nearly 40 percent performance improvement over the previous Mali-G76 used in devices today.
Through a combination of hardware and software optimisations, the past two generations of Cortex-A7x series processors (Cortex-A76 and Cortex-A77) increased overall ML performance 35-fold, the company said.
Mali-G77 boasts key micro-architecture enhancements including engine, texture pipes, and load store caches, which achieve 30 percent better energy efficiency and 30 percent more performance density. On top of that, the Mali-G77 also brings a 60 percent improvement to machine learning performance. These generational enhancements provide developers with more performance functionality to design more immersive games for the mobile app ecosystem, the company said.
Arm Machine Learning processor
The company's already announced Project Trillium, a heterogeneous ML compute platform, includes the Arm ML processor and the open-source Arm NN software framework, which is now shipping in more than 250 million Android devices. Since Arm announced Project Trillium in 2018, it has made enhancements to its ML processor including increasing energy efficiency by more than 2x up to 5 TOPs/W, improving memory compression techniques up to 3x, and scaling to peak next-generation performance up to 8-cores for up to 32 TOP/s.
