By Benson Tao
What is IoT?
The movement towards smart, connected, autonomous, and contextually aware devices built around the Internet of Things (IoT) creates a revolution that will change the world for the better. Devices can now communicate intelligently to each other (M2M, M2I, M2E) or to us, in real time, processing data in the cloud or at the node, deciding autonomously or presenting the most up-to-date information to us so we can make the best decisions. (Note: M2M is for machine-to-machine communications, M2I is machine-to-infrastructure, and M2E is machine-to-environment). As you can see, the term IoT is so broad that it pretty much covers all device segments we see today including wearables, home automation, appliances, energy saving devices, healthcare/medical, sensors, automobiles, consumer devices, and everything in between.
According to a Cisco, there will be at least 25 Billion IoT devices by 2015 and doubling to 50 Billion by 2020. In contrast, the 2015 global shipments of smartphones is estimated to be around 1.24 Billion devices according to IDC, which is a much small number. The possibilities and growth potential of IoT devices is vast and the next key area of high growth opportunities that companies are looking at. It also has the potential to create breakthrough products or the next large silicon vendor designing for the next billion smart devices.
These opportunities and growth are what makes it exciting for our partners embedding GPUs into their designs as part of their product rollout plans. The cool part about using GPUs in IoT is the seamless, eye catching, feature-rich interface that users expect from their devices that enable a ubiquitous user experience across all consumer, automotive, embedded and IoT user interfaces. This changes the product landscape in an important way, morphing the impersonal devices of today into personal assistants that “learn” and “understand” our preferences.
Examples of IoT
Below are a few examples of IoT possibilities that are within reach or starting to be available to consumers:
- Our home automation (HA) system sends a message to our smartphones or wearable device to tell us the garage door is open. Connecting to our HA central system at home, we can view the home security cameras in the house to make sure things are clear, then send a signal through our mobile device so the garage door closes. The user interface on the smartphone and HA system requires a GPU and composition for viewing. The security camera can use OpenCL (GPGPU) to perform analytics and image processing (ex. event triggers) to determine if an alarm or warning needs to be sounded.
- Sensors placed in our clothes can detect our vitals and ECG/EKG characteristics and adjust temperature or send a warning signal to the doctor (with relevant data) if a vital reading is abnormal. Our medical device or wearables require a GPU and composition to display relevant information in a usable, informative manner.
- Road sensors on a bridge inform an approaching car about icy conditions, and the car adjusts it’s stability control and warns the driver to slow down. The data is also used to tell other cars about the road conditions through real time M2M so nearby cars can use useful information to adjust settings automatically.
- Retail stores can use strategic device placement that communicates with our wearables or smartphones to show our taste preferences (style, color, brand name, mood, social media profile, etc.) on nearby screens.
These are all realistic examples of what you might find today or in the near future. The key point is that our surrounding devices will be able to display or process more data, communicate to each other, and show only relevant information.
Vivante IoT Ecosystem
There are many IoT markets that GPUs play an important role in. Here are a few examples where partners are already designing products:
All these areas have interesting applications and many require a consumer friendly display interface that needs to be rendered by a 3D GPU or composition processing core (CPC).
Android as the Backbone of IoT
Android is synonymous with smartphones, tablets, smart TVs, and embedded platforms inside the home, office, automobile, and embedded world, and it can be found pretty much everywhere. In a technology article by Bloomberg Businessweek Technology appropriately titled “Behind the ‘Internet of Things’ Is Android – and It’s Everywhere” it states:
“Android is becoming the standard operating system for the “Internet of Things”—Silicon Valley’s voguish term for the expanding interconnectedness of smart devices, ranging from sensors in your shoe to jet engine monitors…Every screen variant, mobile chip, and sensor known to man has been tuned to work with Android.”
What makes Android a solid platform goes beyond its existing market penetration and its rapidly growing marketshare. It has all the qualities, next generation features, and roadmap needed to support the vast amount of web interconnected products imaginable, including security, upgraded communications protocols, interoperability, and more. From a GPU perspective, the key with using Android to its fullest potential is allowing the visual centric nature of the OS to take center stage. This requires a tiny, IoT based GPU that is full featured and capable of driving the device screen, and a product with a fancier interface will be more enticing to buyers than those without.
As an example, (and assuming costs of both products are similar, which will be the case since a Vivante GPU takes up minimal die area) which device would you like to have? Pre-Android or the Android IoT devices?
About Vivante IoT GPU Products
Optimized for Google Android, Windows Embedded, and other operating systems, Vivante’s IoT product portfolio includes performance-leading technologies in 3D graphics, CPC composition processors, vector graphics, and optional GPGPU cores. Vivante IoT cores leverage a unified driver architecture that is compatible with industry-standard application programming interfaces like OpenGL® ES 2.0, desktop OpenGL®, OpenCL®, OpenVG®, Microsoft® DirectX® 11, WebGL, Google Renderscript / FilterScript Compute, and other standard APIs.
Robust features built into Vivante IoT GPUs include:
- Option 1 (GC400 Core): World’s smallest licensable OpenGL ES 2.0 GPU Core at less than 1 mm2 total silicon area (TSMC 28nm HPM process technology)
- Option 2 (GC880 Core): World’s smallest licensable OpenGL ES 3.0 GPU Core at 2 mm2 total silicon area (TSMC 28nm HPM process technology)
- Supports up to 720p @ 60 FPS with high quality 32-bit color formats (GC400)
- Supports up to 1080p @ 30/60 FPS with high quality 32-bit color formats (GC880)
- Supports all major IoT operating systems, APIs and middleware
- Accelerated composition processing for butter smooth UIs
- Ultra-low power consumption to conserve battery power on-the-go
- Tiny software driver footprint for DDR constrained and DDR-less configurations
- Real-time sensor fusion processing to reduce bandwidth and increase device intelligence
- Industrial temperature support for -40C to 85C
Please visit here for more information on IoT products.