NVIDIA has unveiled Jetson TX2, a credit-card sized board computer that has the size of Raspberry Pi, but is more powerful and faster. The AI enabled platform is designed to help make robots, drones, smart cameras, portable medical devices and other devices that rely on computer vision applications to operate.
NVIDIA claims Jetson TX2 to be two times faster than its predecessor, Jetson TX1, or two times power efficient, while drawing less than 7.5 watts of power. This allows it to run larger, deeper neural networks (on robot, drones), resulting in smarter devices with higher accuracy and faster response times for tasks like image classification, navigation and speech recognition.
The Jetson TX2 can deliver 1.5 teraflops of performance, according to NVIDIA, which is unmatched on developer boards, and the GPU on-board is based o n NVIDIAs Pascal architecture, which is the companys fastest GPU, like the Tesla P100.
The TX2 has two Denver 2 CPU and four Cortex-A57 CPUs and measures 50mm x 87mm (2" x 3.4"). It features 8GB LPDDR4 RAM, 32GB eMMC storage, up to six cameras, Gigabit Ethernet, Bluetooth, and WiFi 802.11ac. The GPU can deliver 4K videos at 30fps or 2K video at 60fps. It can run Ubuntu and ROS (Robot Operating System), which is built on top of Linux OS.
The module will be bundled with JetPack 3.0, a SDK (software de velopment kit), which will make it easy for developers to integrate AI into a wide variety of application. It supports TensorRT, cuDNN 5.1, VisionWorks, CUDA 8 and the latest graphics drivers and API.
The NVIDIA Jetson TX2 Developer kit, which includes the carrier board and TX2 module is priced at $599 (Rs. 40,000) and is available for pre-order in the US and Europe. The shipping will begin March 14. It will also be available in other regions in the coming week. The Jetson TX 2 module will be available in Q2 for $399 (RS. 26,590) (in quantities of 1,000 or more) from NVIDIA and its distributors around the world. Meanwhile NVIDIA is also reducing the price of the Jetson TX1 Developer Kit and now costs $499.
Source | Via
0 komentar:
Posting Komentar