How to choose your own Jetson development board？

As Nvidia launched the core boards of Jetson Orin NANO and Jetson Orin NX, many electronic designs or open source hardware companies have created carrier boards based on the official core boards. And put it together, launched the Jetson Orin NANO 4GB/8GB SUB and Jetson Orin NX 8GB/16GB SUB developer kits.Nvidia launched the core boards of Jetson Orin NANO and Jetson Orin NX, many electronic designs or open source hardware companies have created carrier boards based on the official core boards. And put it together, launched the Jetson Orin NANO 4GB/8GB SUB and Jetson Orin NX 8GB/16GB SUB developer kits.

Due to there are more and more development boards in the Jetson series, most customers don’t know how to choose the Jetson development board that suits them, so today’s article will help you solve this question.

1. Jetson Orin NX(2023 New)

Composed of NVIDIA Jetson ORIN NX core board and Yahboom Jetson ORIN carrier board.

RAM: 8GB/16GB 

Storage: No eMMC/No TF card slot, 128G SSD(Yahboom provide for free)

Computing power: Jetson ORIN NX 8GB---70TOPS and Jetson ORIN NX 16GB---100TOPS

The Jetson Orin NX SUB development kit adopts the official original Jetson Orin NXmodule, equipped with Yahboom Jetson Orin NX carrier board, and comes with WiFinetwork card,128GBM.2 SSD, power supply and other accessories. On board resources and size layout are consistent with the official board. We have alsodesigned a more stable power supply scheme specifically for the power consump-tion characteristics of the Orin NX core module, supporting POE power supply tomeet the 25W power consumption requirement, which can achieve larger and morecomplex neural networks.

Jetson Orin NX Hardware Function Distribution

2. Jetson Orin NANO(2023 New)

Composed of NVIDIA Jetson ORIN NANO core board and Yahboom Jetson ORIN carrier board.

RAM: 4GB/8GB 

Storage: No eMMC/No TF card slot, 128G SSD(Yahboom provide for free)

Computing power: Jetson ORIN NANO 4GB---20TOPS and Jetson ORIN NANO 8GB---40TOPS

The Jetson Orin Nano SUB developer kit adopts the official original Jetson rinNano module, equipped with the Yahboom Jetson Orin Nano carrier board, andcomes with WiFi network card, 128GB M.2 SSD, power supply and other accesso-ries. lts performance, on board resources,size, and interface layout are consistentwith the official Jetson Orin Nano board, and adopt a more stable current powersupply circuit that supports POE power supply. lt is an excellent alternative to theofficial Jetson Orin Nano.

Jetson Orin NANO Hardware Function Distribution

!!! Note: The carrier boards of Yahboom Jetson ORIN NANO and Jetson ORIN NX are the same, with only the core board being different.

3. Jetson Xavier NX SUB

Composed of NVIDIA Jetson Xavier NX core board and Yahboom Jetson Xavier NX carrier board.

RAM: 8GB/16GB 

Storage: No TF card slot, 16G eMMC + 128G SSD(Yahboom provide for free)

Computing power: 21TOPS

This version is JETSON XAVIER NX-SUB developer kit(eMMC version),not JETSON XAVIER NX developer kit(Official Demo). It aims to solvethe long-term material shortage problem of the JETSON XAVIER NX developer kit(Official Demo). It is cost-effective kit for all users. And ithas the same performance as the JETSON XAVIER NX developer kit(Official Demo), only slightly different in storage performance.
The SUB developer kit comes with 16G eMMC storage and does not support TF card.In order to meet the needs of Al development, all SUB kit configurations in our store include a 128G NVMe SSD. Its readingspeed is more than 18 times that of TF card, and its writing speed ismore than 5 times that of TF card.

Jetson Xavier NX Hardware Function Distribution

4. Jetson TX2 NX SUB

Composed of NVIDIA Jetson TX2 NX core board and Yahboom Jetson TX2 NX carrier board.

RAM: 4GB 

Storage: No TF card slot, 16G eMMC + 128G SSD(Yahboom provide for free)

Computing power: 1.33TFLOPS (FP16)

The Jetson TX2-NX core module is a cost-effective alternativecore module newly designed by NVIDIA in 2021, and the TX2-NXdeveloper kit is not officially provided. Because it has the coreperformance of TX2 and the size is consistent with Nano/XavierNX, the official name of this core module is "TX2-NX".
According to the market status and customer demand, Yahboom will launch the TX2-NX development kit in 2022. This kit providesdevelopers, engineers, and students with an affordable newoption with greatly improved performance.

Jetson TX2 NX Hardware Function Distribution 

5. Jetson NANO B01  

Composed of NVIDIA Jetson NANO B01 core board and NVIDIA Jetson NANO B01 carrier board.

RAM: 4GB

Storage: 16G eMMC/With TF card slot

Computing power: 0.5T (FP16)

NVIDIA JETSON NANO B01 is a small, powerful computer that lets you run
multiple neural networks in parallel for applications like image classification, object detection, segmentation, and speech processing. The JETSON NANO with a quad-core CORTEX-A57 processor, a 128-coreMAXWELL GPU and 4GB of LPDDR memory with power Al computing power toprovide 472GFLOP and support a range of popular Al frameworks and algorithmssuch as TENSORFLOW, PYTORCH,CAFFE ICAFFE2,KERAS,MXNET, etc.

Summarize

1. If you are a beginner learning artificial intelligence or Python programming, or if you are a student with limited funding. I recommend choosing Jetson NANO 4GB B01, which has sufficient computing power and a suitable price.

2. If you need to build your own ROS robot with equipment such as lidar and depth camera, and want to achieve artificial intelligence algorithms such as map navigation, object recognition, AI visual recognition, etc., we recommend you choose the TX2 NX development board.

3. If you need to build high-performance AI systems suitable for drones, portable medical devices, small commercial robots, intelligent cameras, high-resolution sensors, automatic optical detection, and other IOT embedded systems. Recommend using Xavier NX development board.

4.If you need to carry out some large-scale AI projects with high power consumption, complex algorithms, and sufficient budget. We recommend choosing Orin NX/Orin NANO.