MediaTek’s Helio P70 SoC Suits AI Apps

October 25, 2018 Rich Nass

MediaTek recently announced the launch of the Helio P70 system-on-chip (SoC), with an enhanced AI engine combined with CPU and GPU upgrades for more powerful AI processing. Helio P70 also comes with upgraded imaging and camera support, a gaming performance boost and advanced connectivity features.

Built with TSMC’s 12-nm FinFET technology, the new device features a multi-core APU operating at up to 525 MHz. To handle AI applications, the chipset includes four Arm Cortex-A73 2.1-GHz processors and four Arm Cortex-A53 2.0-GHz processors in an octa-core big.LITTLE configuration. In addition, the chipset has an improved Arm Mali-G72 MP3 GPU operating at up to 900 MHz.

Helio P70 is built on MediaTek’s NeuroPilot platform, the company’s comprehensive hardware and software ecosystem to support edge AI. NeuroPilot supports common AI frameworks, including TensorFlow, TF Lite, Caffe, Caffe2 and custom third-party offerings. Helio P70’s enhanced AI engine can support complex AI applications such as real-time human pose recognition and AI based video encoding. Helio P70 GPU enhancements also deliver a better overall gaming experience as it’s optimized to reduce frame-rate jitter and improve latency for touch controls and display visuals for a smooth, fluid gaming experience.

Helio P70 supports a variety of AI-infused photo and video experiences including real-time beautification, scene detection and artificial reality (AR) capabilities. The chipset features improved deep-learning facial detection with up to 90 percent accuracy. With support for a super-sized 32MP single camera, or 24+16MP dual cameras, the chipset gives device makers more design flexibility. Its three ISPs are tuned to provide significant power savings, reducing consumption by 18 percent compared to previous Helio generation dual-camera setups.

The SoC comes with a 4G LTE modem and 300MBit/s of download performance. Dual 4G VoLTE support provides a seamless user experience across two different SIMs for higher quality calls and faster connection times. The part is now in volume production.

About the Author

Rich Nass

Richard Nass is the Executive Vice-President of OpenSystems Media. His key responsibilities include setting the direction for all aspects of OpenSystems Media’s Embedded and IoT product portfolios, including web sites, e-newsletters, print and digital magazines, and various other digital and print activities. He was instrumental in developing the company's on-line educational portal, Embedded University. Previously, Nass was the Brand Director for UBM’s award-winning Design News property. Prior to that, he led the content team for UBM Canon’s Medical Devices Group, as well all custom properties and events in the U.S., Europe, and Asia. Nass has been in the engineering OEM industry for more than 25 years. In prior stints, he led the Content Team at EE Times, handling the Embedded and Custom groups and the TechOnline DesignLine network of design engineering web sites. Nass holds a BSEE degree from the New Jersey Institute of Technology.

Follow on Twitter Follow on Linkedin Visit Website More Content by Rich Nass
Previous Article
It’s Time to Vote!

The winner of the Make the World a Better Place Design Challenge stand to claim more than $10,000 in prize ...

Next Article
RTI Announces Connext 6, Connectivity Software for Highly Autonomous Systems

Connext 6, built on Data Distribution Service (DDS), includes the first connectivity software designed to a...