People consume rich media content of all flavors wherever they go now. And user expectations in terms of unique experiences and immersive quality are continuously trending upward. Consumers have proven an insatiable hunger for sophisticated image and sound. This generation’s eyes and ears are trained to notice fine degrees of qualitative difference.
Evolving processing capabilities are about to change how users interact with both technology and content in revolutionary ways. The combination of powerful serial (CPU) and parallel (GPU) processing—also known as heterogeneous computing—delivers critical multimedia features to power next-generation interface and experience breakthroughs, including applications that use artificial intelligence (AI). What once required a server with multiple processing cores and a professional-class discrete graphics card can now be delivered in a personal device.
NXP’s new i.MX 8M family of applications processors introduces groundbreaking functionality to a wide spectrum of next-generation devices. The i.MX 8M key features include high-performance compute, power efficiency and embedded security that will drive rapid proliferation of edge node computing, rich streaming multimedia and machine learning (ML).
Why Multimedia will be Critical to Next-Gen Devices
Rich, experiential multimedia (4K HDR video, immersive sound, voice control, computer vision) requires significant compute power. The i.MX 8M family is based on quad-core Arm® Cortex®-A53 cores running up to 2GHz. It also includes a high performance 3D graphics GPU capable of processing OpenCL machine learning inference on chip along with the Arm® Cortex®-M4 to support voice wake word detection and low power audio playback.
Cloud connectivity on the i.MX 8M expands real-time processing for edge compute and machine learning applications. This enables devices to take advantage of AI applications that collect and analyze data in support of algorithm training on cloud servers. In essence, it gives developers the ability to create devices that are capable of self-learning and intelligent, automated responses.
Within the series, the i.MX 8M Mini model introduces NXP’s first embedded, multi-core, heterogeneous applications processor built using advanced 14nm FinFET process technology – supporting high-performance processing with power efficiency for battery-driven devices in small enclosures.
What’s Now Possible
We’re bringing new capabilities to your next-gen smart designs, as well as other compute-intensive use cases, such as industrial controls, building automation, machine vision, imaging and scanning.
The i.MX 8M family capabilities fall into four key solution areas:
Voice: Humans speaking to computers and receiving verbal replies from an application is both the present and future of human machine interface (HMI). It’s estimated that users will initiate 50% of all Internet searches via voice command by 2020.[1]
Audio: The migration of DSP functionality into the Arm core is driving a system architecture change that enables immersive audio in compact form factors. With TV panels getting slimmer, audio capabilities are being separated into detached sound bar devices. Bluetooth speakers are giving way to new networked Wi-Fi speakers which include voice control capabilities. These products are always connected and ready to spontaneously answer an owner’s request.
Video: The i.MX 8M family’s HDR (high-dynamic range) video capabilities were designed in-house—which makes them a unique and original NXP feature that no other company can offer. By 2018, IP video will represent 79% percent of all global traffic.[2] The cord-cutting momentum away from cable subscription is shifting markets from traditional STBs (set-top boxes) to OTT (over-the-top) IP-based video on demand. Customer desire for better viewing experiences are driving the need for updated equipment that supports higher video quality standards and lower power consumption.
Artificial Intelligence and Machine Learning: Connected IoT devices featuring voice control are already revolutionizing the smart home. Many home appliances are currently adding voice, face recognition and other smart controls in current or upcoming models. AI and machine learning (ML) capabilities will combine with new applications that can be trained to learn and respond intuitively to drive this market even higher.
Take Your New Designs Further, Faster with NXP
The i.MX 8M family was designed to accelerate production of devices that will introduce radical new capabilities and experiences to your customers. Not only does it offer breakthrough performance and features in an applications processor, but it does so with a price point and product availability that makes i.MX 8M products accessible to a larger customer base.
NXP is committed to providing outstanding experiences to our tens of thousands of customers around the world –and we have a proven track record for excellence. Building on the decade-long success of the i.MX series of general purpose application processors, the i.MX 8M family has the same outstanding technical support, training resources and participation in our longevity program.
To learn more about i.MX 8M family products visit: iMX8M.
[1] “Why You Need To Prepare For A Voice Search Revolution,” Forbes, January 9, 2018.
[2]“Cisco Visual Networking Index: Forecast and Methodology, 2016–2021,” Cisco, September 15, 2017.
Tony Masterson is CTO of AI and Media for Edge Processing at NXP. He founded the startup Zenverge, maker of video transcoder chips, that was successfully acquired by Freescale. Prior to Zenverge, he was VP of Engineering at multiple companies that were acquired including GlobespanVirata and iCompression. Tony Masterson received his MSEE and BSEE from MIT and holds 30 U.S. patents in the areas of digital video, computer graphics and architecture.
