Nvidia’s Jetson TX1 Module Can Make Machines Smarter

It appears that Nvidia is looking to give a helping hand to hardware developers who want to make their devices smarter, as the company has recently announced its Jetson TX1 developer module, which focuses on machine learning. What makes this particular module stand out is the fact that it could be used to dramatically shorten the lead time required to create “smart” machines such as drones, as well as complex software used for facial recognition and behavioral analysis. Acting as an official successor for Nvidia’s TK1, the TX1 development kit boasts a performance of one teraflop, requires just 10 watts of power and can connect to the internet using 802.11 ac and ethernet.

Available with Linux pre-installed, the NVIDIA Jetson TX1 Developer Kit is set to become available for pre-order on November 12 for $599. As for the standalone module, it costs $299 and is scheduled for shipping during 2016’s first quarter. Jetson TX1 is the very first embedded computer that was designed specifically to process complex software that can recognize objects or interpret pieces of information. Machine learning could be used to enable devices such as drones to process speech, recognize images, or navigate through obstacles. Deepu Talla, the vice president and general manager of Nvidia’s Tegra, has stated the following:

“Jetson TX1 will enable a new generation of incredibly capable autonomous devices. They will navigate on their own, recognize objects and faces, and become increasingly intelligent through machine learning. It will enable developers to create industry-changing products.”

Below you will find a list of key features for this new Nvidia product.

• GPU: 1 teraflops, 256-core Maxwell architecture-based GPU offering best-in-class performance.
• CPU: 64-bit ARM A57 CPUs.
• Video: 4K video encode and decode.
• Camera: Support for 1400 megapixels/second.
• Memory: 4GB LPDDR4; 25.6 gigabits/second.
• Storage: 16GB eMMC.
• Wi-Fi/Bluetooth: 802.11ac 2×2 Bluetooth ready.
• Networking: 1GB Ethernet.
• OS Support: Linux for Tegra.
• Size: 50mm x 87mm, slightly smaller than a credit card.