MediaTek’s new flagship mobile processor ‘Helio P90’ offers a new level of AI experiences and high-resolution smartphone photography in an energy-efficient design
MediaTek has launched the Helio P90 chip which features APU 2.0, the company’s new-generation architecture for AI processing boost and performance of 1,127GMACs (2.25TOPs).
The company claimed this AI architecture offers 4X performance than the Helio P70 and Helio P60 chipsets. It helps users perform intensive AI tasks effortlessly with faster and more accurate results while at the same time delivering longer battery life.
“The Helio P90 gives device makers the ability to offer outstanding camera features, battery life, performance and top of the line AI capabilities, while making devices accessible to mass-market consumers,” said TL Lee, General Manager of MediaTek’s Wireless Communication business unit. “The Helio P90 is an AI processing powerhouse with ultra-fast performance and a host of AI-powered imaging improvements. This chip will reset consumer expectations of what a smartphone can do, and usher in a new era of super high-resolution smartphone photography to the market.”
Powerful AI engine and advanced camera features
The octa-core Helio P90 is designed with two Arm A75 processors operating at up to 2.2GHz and six A55 processors operating at up to 2.0GHz. It packs an Imagination Technologies PowerVR GM 9446 GPU.
The company said its newest CorePilot technology also ensures processing is done in the most efficient way across its cores. That means MediaTek Helio P90 offers high performance with low power operation delivering improved battery life.
With its powerful AI engine, the chip supports faster, complex and more dynamic AI experiences such as human pose detection which can track and analyze body movements.
It also supports Google Lens, deep-learning facial detection, real-time beautification, object and scene identification, artificial reality (AR) and mixed reality (MR) acceleration, plus other real-time enhancements for photos and videos.
Developers and device makers can create innovative AI applications with the Helio P90 using common frameworks such as TensorFlow, TF Lite, Caffe and Caffe2, based on MediaTek’s NeuroPilot SDK which is fully compliant to Google Android Neural Networks API (Android NNAPI).
Users can capture pictures with support for a supersized 48MP camera or 24+16 MP dual cameras.
Users can capture at 48MP with up to 30 frames per second (FPS) with zero shutter delay, or enjoy super slow-motion at 480FPS in HD to capture every moment.
The upgraded triple image signal processors (ISPs) are capable of 14-bit RAW and 10-bit YUV processing. This, the company says, will allow photography enthusiasts to have even more flexibility to capture and process stunning photos.
The new ISP AI engine can accurately detect faces and scenes in real-time under low light and motion conditions, making a great shot easier than ever for any user, the company noted.
The Helio P90 supports dual SIM dual VoLTE 4G LTE WorldMode with a fast Cat-12 (DL) / Cat-13 (UL) 4G LTE modem. 4×4 MIMO and 256QAM provided better data throughput in highly populated areas. It also supports 2×2 802.11ac and Bluetooth 5.0
The Helio P90 is expected to be available globally in consumer devices starting in Q1 2019.
Click here for more details on MediaTek Helio P90.