Jetson TX2
NVIDIA's Jetson TX2 is an embedded system-on-module (SoM) with dual-core NVIDIA Denver2 + quad-core ARM Cortex-A57, 8GB 128-bit LPDDR4 and integrated 256-core Pascal GPU.
Useful for deploying computer vision and deep learning, Jetson TX2 runs Linux and provides greater than 1TFLOPS of FP16 compute performance in less than 7.5 watts of power.
Jetson TX2 is available as the module, developer kit, and in compatible ecosystem products. See wiki of previous Jetson's here.
Parallel ForAll — NVIDIA Jetson TX2 Delivers Twice the Intelligence to the Edge |
Contents
Jetson TX2 Module
The Jetson TX2 module contains all the active processing components. The ports are broken out through a carrier board.
Below is a partial list of the module's features. Please see the Jetson TX2 Module Datasheet for the complete specifications.
Processing Components
- dual-core NVIDIA Denver2 + quad-core ARM Cortex-A57
- 256-core Pascal GPU
- 8GB LPDDR4, 128-bit interface
- 32GB eMMC
- 4kp60 H.264/H.265 encoder & decoder
- Dual ISPs (Image Signal Processors)
- 1.4 gigapixel/sec MIPI CSI camera ingest
Ports & Peripherals
- HDMI 2.0
- 802.11a/b/g/n/ac 2×2 867Mbps WiFi
- Bluetooth 4.1
- USB3, USB2
- 10/100/1000 BASE-T Ethernet
- 12 lanes MIPI CSI 2.0, 2.5 Gb/sec per lane
- PCIe gen 2.0, 1×4 + 1×1 or 2×1 + 1×2
- SATA, SDcard
- dual CAN bus
- UART, SPI, I2C, I2S, GPIOs
Form-Factor
- 400-pin Samtec board-to-board connector
- dimensions: 50x87mm (1.96" x 3.42")
- Thermal Transfer Plate (TTP), -25C to 80C operating temperature
- mass: 85 grams, including TTP
- 5.5-19.6VDC input power (consuming 7.5W under typical load)
Software Support
- JetPack 3.1
- Linux4Tegra R28.1 (L4T)
- Linux kernel 4.4
- Ubuntu 16.04 aarch64
- CUDA Toolkit 8
- cuDNN v6.0
- TensorRT 2.1
- VisionWorks 1.6
- OpenCV4Tegra 2.4.13-17
- OpenGL 4.5 / OpenGL ES 3.1
- V4L2 media controller support
- GStreamer 1.8.2
- Tegra System Profiler 3.7
- Tegra Graphics Debugger 2.3
Parallel ForAll — JetPack 3.1 Doubles Jetson's Low-Latency Inference Performance |
Jetson TX2 Developer Kit
The Jetson TX2 Developer Kit bundles together all the parts to get started, including:
What's Included
- mini-ITX Reference carrier board
- Jetson TX2 Module
- fan and heatsink (pre-assembled)
- 5MP CSI camera module (with Omnivision OV5693)
- WiFi/BT antennas
- USB OTG adapter
- 19VDC Power brick
- AC Power cable
The design files for the reference carrier board and camera module are freely available for download.
Getting Started
- Get the latest development software for PC and TX1 by using JetPack.
- Plug in an HDMI display into Jetson, attach the antennas and USB keyboard & mouse, and apply power to boot it up. (User Guide)
- Visit the Embedded Developer Zone and Jetson TX2 Developer Forum to access the latest documentation & downloads.
Availability
- The devkit is available through NVIDIA's Jetson TX2 Developer Kit webpage.
- The individual module is available through NVIDIA's Jetson TX2 Module webpage.
- Alternatively, use the Region Selector to find distributors of the devkit in your region.
- There's also an Academic Discount available for those affiliated with an educational organization.
Platform Documentation
NVIDIA has released comprehensive documentation and reference designs for the Jetson TX1 module and devkit.
- Module Datasheet the official module features, ports, signal pin-out, and package specifications.
- Design Guide detailed technical design and layout information for creating OEM products.
- DevKit Carrier Spec design info about the reference carrier board from the devkit.
- DevKit Design Files schematics, layout, and design files for the devkit reference carrier board.
- DevKit CAD Models 3D STEP file for reference carrier board, heatsink, camera board, and module.
- Camera Design Files schematics, layout, and design files for the devkit MIPI CSI-2 camera module.
- Thermal Design Guide mechanical specifications for designing active and passive cooling solutions.
- TX1/TX2 Migration guide to porting applications and hardware between Jetson TX1 and TX2
- Battery Charger Guide document for the design of battery charger
- Tegra X2 (Parker) TRM Technical Reference Manual for NVIDIA TX2 system-on-chip and register data.
- L4T Kernel Docs documentation for L4T kernel developers (including V4L2/camera drivers).
- Multimedia API Reference documentation to Argus camera API and V4L2 media codecs
- Accelerated GStreamer Guide example gstreamer pipelines for accessing H.264/H.265 hardware video codec.
Above is a partial list of documents.
Please visit the Downloads Center at Embedded Developer Zone for the full list that's currently available.
USB 3.0 webcams known to be working
e-con Systems' See3CAM_CU135 was tested on Jetson TX2 with HD (1280X720) @ 46fps and FullHD (1920x1080) @ 36fps in MJPEG (compressed) format, as well as other settings.
Guides and Tutorials
This section contains recipes for following along on Jetson.
System Tools
Please see Jetson TX1 Wiki for similar entries that also apply to TX2.
- Cloning & Restore
- Factory Images
- Setting the DTB
- Enabling the SPI Port
- Building Kernel and Modules
- Enabling USB on Custom Carriers
- Maximizing RootFS Partition on eMMC
- nvpmodel - dynamic performance profiles
- TX2 GPU support in Docker - script to enable GPU sharing from within Docker container
- Tegra-Docker
- Ubuntu Base Minimal footprint (500Mb with OS only) using Ubuntu Base
- Boot from SD card
- Display Driver Debugging
- Getting Swift to Run on TX2
Robotics
- NVIDIA Jetson GitHub (open-source robotics projects with deep learning)
- NVIDIA Redtail GitHub (end-to-end deep learning drone for ROS)
- Jetson Reference Platforms (off-the-shelf robots with TX1/TX2)
- FIRST FRC Configuration (setup guide for FIRST Robotics)
- FIRST FRC Neural Networks (Zebracorns team #900 object tracking)
- ROS for FRC Whitepaper (Zebracorns team #900 Vision GitHub)
- Installing ROS Kinetic (TX2) (JetsonHacks guide)
- Fast SLAM Particle Filter (Accelerated Localization using Raycasting)
Computer Vision
- NVIDIA OpenCV 101 - screencast tutorials
- OpenCV-3.4.0 for TX2 building script
- Build OpenCV for TX2 (JetsonHacks)
- Building OpenCV 3.2 with CUDA for Tegra
- VisionWorks training
- gstreamer Pipelines for TX2
Deep Learning
- NVIDIA Two Days to a Demo (DIGITS/TensorRT)
- Caffe (BVLC Model Zoo)
- NVcaffe FP16 (Install Guide)
- Caffe Installation (JetsonHacks)
- Caffe2 (github.com/caffe2)
- Deep Visualization Toolbox install script
- TensorFlow install for JetPack 3.1
- TensorFlow post for JetPack 3.0 (JetsonHacks)
- TensorFlow install procedure (pip wheel)
- TensorFlow script and pip wheel for JetPack 3.2 DP
- Torch7 install script
- pyTorch install script
- dusty-nv's Jetson GitHub jetson-inference jetson-reinforcement
- jetson-inference-gv GigEVision / RTP streaming video (Ross Newman)
- jetson-inference-cards (playing card recognition by S4WRXTTCS)
- face-recognition (face detection with TensorRT plugin API by AastaNV)
- ChatBot (TensorFlow→TensorRT inferencing workflow by AastaNV)
- NVIDIA GitHub (open-source robotics/DL projects)
- NVIDIA Redtail (end-to-end deep learning drone for ROS)
- Training a Fish Detector with DetectNet part 1 part 2 (jkjung)
Multiple Cameras
There are several ways to handle multiple cameras on Jetson TX2 at the same time:
6 MIPI CSI-2 Cameras support for Jetson TX2 from e-con systems
Ecosystem Products
The following are 3rd-party carriers, enclosures, drones, and accessories available for Jetson TX2.
Please see additional backwards-compatible Ecosystem Products for TX1.
- Abaco GVC1000 enclosure
- Aetina N620 nano-ITX carrier
- Auvidea J20 6-camera module
- Auvidea J100 carrier
- Auvidea J106 carrier (6 camera)
- Auvidea J120 carrier
- Auvidea J130 carrier (4K input)
- Auvidea J140 dual-GbE carrier
- Auvidea J150 OpenGear blade
- Auvidea J200 carrier
- black.ai perception platform
- ConnectTech Sprocket carrier
- ConnectTech Orbitty carrier
- ConnectTech Spacely carrier
- ConnectTech Cogswell carrier
- ConnectTech Rosie enclosure
- ConnectTech Elroy carrier
- ConnectTech Rudi enclosure
- ConnectTech Astro carrier
- ConnectTech 3U VPX card
- Curtiss-Wright Parvus DuraCor-312 rugged enclosure
- DCDZ(冬虫电子) XCB-Lite carrier with eDP, CSI, DSI, MicroHDMI etc
- DCDZ(冬虫电子) XCB-OV5640 OV5640 M12 lens camera
- DCDZ(冬虫电子) XCB-OV5693 OV5693 camera
- DCDZ(冬虫电子) XCB-H2C HDMI to CSI expansion board
- DCDZ(冬虫电子) XCB-SDI SDI to CSI expansion board
- DCDZ(冬虫电子) XCB-8eyes 4 or 8 cameras ADAS expansion board
- DCDZ(冬虫电子) XCB-AUDIO Sound card expansion board
- RidgeRun GStreamer and Multimedia Solutions
- e-con Systems 2MP SONY IMX290 MIPI Jetson TX2 Camera
- e-con Systems 13MP AR1335 MIPI Jetson TX2 Camera
- e-con Systems 3.4 MP AF AR0330 MIPI Jetson TX2 Camera
- e-con Systems 13MP AR1820 MIPI Jetson TX1 Camera
- e-con Systems 3.4 MP AR0330 MIPI Jetson TX1 Camera
- Silverstone PT13 mini-ITX system
- Rebotnix GUSTAV enclosure
- Vision4CE CHARM-100 enclosure
Getting Help
If you have a technical question or bug report, please visit the DevTalk Developer Forums and search or start a topic.
We summarize some useful topics in http://elinux.org/Jetson_TX2/TX2_Issue page.
See the official Support page on Embedded Developer Zone for warranty and RMA information: https://developer.nvidia.com/embedded/support
For NVIDIA webstore Customer Service, please see the My Account page or contact 1-800-797-6530.