【AI】NVIDIA Jetson series summary

1. Introduction to Jetson series

NVIDIA defines Jetson as an embedded system for a new generation of autonomous machines. Each Jetson system is a complete system on module (SOM) with CPU, GPU, PMIC, DRAM and flash memory, and is scalable.

2. Jetson family tree

2.1 The first generation: TK1

Jetson's earliest product is the TK1, which is an antique and was launched in March 2014

TK1 is NVIDIA's first attempt in the field of edge computing. It is a low-cost development board based on Tegra K1, which unleashes the potential of GPU for embedded system applications. It's built on the revolutionary NVIDIA Tegra K1 SoC and uses the same NVIDIA Kepler computing core designed for supercomputers around the world. TK1 provides a full-featured NVIDIA CUDA platform with pre-installed Linux4Tegra OS (Ubuntu 14.04 with pre-configured drivers). According to the official website, NVIDIA can provide the entire BSP and software package, including CUDA, OpenGL4.4 and OpenCV accelerated by Tegra. Can be used to rapidly develop and deploy compute-intensive systems for computer vision, robotics, medical and more.

2.2 The second generation: TX1

Jetson TX1 was launched in March 2016, TX1 has been discontinued

The Jetson TX1 is the world's first module-based supercomputer capable of delivering the performance and energy efficiency required for the latest visual computing applications. It is built on the revolutionary NVIDIA Maxwell architecture with 256 CUDA cores, delivering over 1 TeraFLOP of performance. 64-bit CPU, 4K video encoding and decoding capabilities, and 1400 MPix/s camera interface make it an embedded deep learning, computer vision

Guess you like

Origin blog.csdn.net/u010168781/article/details/130284758