NVIDIA Jetson™ Nano

General

CPU	Cortex-A57 @ 1.43 GHz
Cores	4
RAM	4GB
USB2.0 Interfaces	1
USB3.0 Interfaces	4
Ethernet	10/100/1000 MBit
PCIe	x1/x2/x4 Gen 2.0

Note

The above table describes the specification of the NVIDIA Jetson™ Nano Developer Kit.

Additional Settings

Increasing the USB kernel buffer memory

Please refer to Increasing Kernel memory to learn about how to calculate a suitable USB kernel buffer memory for your application. Normally, for a single-camera application, a value of 256 MB should be adequate. To increase the value permanently on NVIDIA Jetson™ platforms, append the calculated usbcore.usbfs_memory_mb as boot parameter in /boot/extlinux/extlinux.conf to "APPEND", e.g.:

APPEND ${cbootargs} usbcore.usbfs_memory_mb=256 

Disabling USB autosuspend mode

In some cases the operating system might auto-suspend the USB device to save power. This might result in data loss during the acquisition sometimes. To permanently disable this power saving mode, append usbcore.autosuspend=-1 as boot parameter in /boot/extlinux/extlinux.conf to "APPEND", e.g.:

APPEND ${cbootargs} usbcore.autosuspend=-1

See also

• Disabling The Auto-Suspend Mode

Benchmarks

The following tests have been performed using different de-Bayering scenarios to achieve the max. FPS while maintaining 0 lost images. The CPU load during the acquisition is also documented below.

Scenarios that have been tested are listed as follows:

1. When de-Bayering is carried out on the camera: The camera delivers RGB8 image data to the host system. This setting results in a lower CPU load but a lower frame rate.
2. When de-Bayering is carried out on the host system: The camera delivers Bayer8 image data to the host system. The Bayer8 image data then get de-Bayered to RGB8 format on the host system. This setting results in a higher frame rate but a higher CPU load as well.
3. When no de-Bayering is performed: The camera delivers Bayer8 image data to the host system. No de-Bayering is performed. This settings results in a lower CPU load and a higher frame rate. The behavior is identical to monochrome cameras.

Camera	Resolution	Pixel Format	Frame Rate [Frames/s]	Bandwidth [MB/s]	CPU Load
mvBlueFOX3-2032C	2064 x 1544	RGB8 (on camera) → RGB8 (on host)	25.20	240.92	40%
Camera	Resolution	Pixel Format	Frame Rate [Frames/s]	Bandwidth [MB/s]	CPU Load
mvBlueFOX3-2032C	2064 x 1544	BayerRG8 (on camera) → RGB8 (on host)	97.99	312.26	~90%
Camera	Resolution	Pixel Format	Frame Rate [Frames/s]	Bandwidth [MB/s]	CPU Load
mvBlueFOX3-2032C	2064 x 1544	BayerRG8 (on camera) → BayerRG8/Raw (on host)	119.19	379.86	50%

Remarks

Choose the right power supply

The Jetson Nano has 2 power supply possibilities: via the micro-USB connection or via the Barrel Jack connection.

The power (by default 10W) via the micro-USB connector is not sufficient if peripherals (e.g. keyboard, mouse, cameras, etc...) are connected. To avoid the system from throttling due to over-current, please supply the board with power through the Barrel Jack connector (4A@5V), when powering the USB/USB3 camera through the USB bus.