TOKYO – GPUs shine no longer in just supercomputers, PCs and game machines. Nvidia, the graphics chip giant, is eagerly plowing inroads into the embedded market by leveraging Jetson TX1, its AI computing platform.
As Nvidia zeroes in on the AI market, it sees an opportunity to pursue diverse product segments ranging from gaming and AR/VR to data centers and self-driving cars.
In particular, the company is exploring “billions of intelligent devices” – drones, robots, anything connected – designed to exploit deep neural networks (DNN), according to Lazaar Louis, Nvidia’s marketing and business development leader. The goal is to extend AI from the cloud to the edge, he told us.
While other chip vendors may opt to design development boards tailored for a specific product category (i.e. Qualcomm’s S820A Automotive Development Platform, NXP’s Blubox, both for automotive), Nvidia is taking a universal approach in its common visual and AI computing platform – called Jetson. Nvidia hopes to meet diversified system needs by providing designers a variety of SDKs ranging from deep learning and computer vision to GPU compute and multimedia.
Nvidia is pitching Jetson TX1 to system designers interested in adding AI capabilities or high-end multimedia features to embedded systems.
Packed in Jetson TX1 are an SoC featuring 1 TFLP/s 256-core Maxwell GPU, 64-bit ARM A57 CPUs and 4k H.265 encode and decode IP blocks, camera inputs, memory, storage, wireless connectivity and 1-Gigabit Ethernet. Nvidia claims that Jetson TX1 is credit card-size with power consumption under 10W.
Cisco to sign up
Jetson TX1 Spec (Source: Nvidia)
The latest to sign up for Jetson TX1 is Cisco. It will use Jetson TX1 for its newest enterprise collaboration tool, Cisco Spark Board. The 55-inch and 70-inch 4K Spark Board allows users to share a screen, use interactive whiteboard and video-conference.
Cisco picked Jetson TX1 for several reasons. One is a “DirectStylus” technology Nvidia originally developed for tablets, with “extremely low pen-to-ink latency” to replicate the natural whiteboarding experience. Jetson TX1 also offers “advanced graphics,” important in driving the interactive content-sharing experience. Cisco finds TX1’s “image processors and GPUs” critical, as they deliver high-resolution video from the Spark Board’s built-in camera to “create intelligent views from remote participants.”
Cisco Spark Board (Source: Nvidia)
Asked to compare Nvidia’s Jetson with what others are offering in AI computing, Kevin Krewell, principal analyst at Tirias Research, told us, “Most of the other silicon vendors prefer to build development boards like this Qualcomm Snapdragon 820 development board.” He added, “then, they rely on the Tier 1 automotive suppliers to build automotive class boards or partners like Arrow to make them production worthy.”
Next page: 'Under 10W' low power enough?