Various questions on the Jetson Orin Nano Super

Robotics & Edge Computing Jetson Systems Jetson Orin Nano
1

Hi, a few questions on the Jetson Orin Nano Super if someone could answer:

  1. I’ve heard that the it uses the Raspberry Pi pin layout but has less voltage. Is that true? Can you tell me anything more on that? Do cameras for the Raspberry Pi work on it?

  2. I’ve heard that the barrel jack included is 19V and the max voltage is 20V. Is that true? Is there any benefit buying a more powerful power supply for it? How many watts does the power supply it comes with support?

  3. What’s the deal with JetPack and software support? The pinned comment says 6.1 but I thought I read on some Nvidia page that only the Xavier or something was supported by 6.1?

  4. What’s the driver update schedule for it? It looks like Jetpack 5.X was 550 and 6.X is 580? How long is it going to receive driver updates?

  5. What’s the speed of all the USB ports? How much power can they draw? Is data/power shared between all of them?

  6. Can other distros be installed on the device using the normal Arm64 drivers?

  7. Does it include WIFI and BT?

  8. My understanding is that there is an app only for X86 Ubuntu. Is it actually required?

  9. What is the state of NVML on it?

Thanks for any answers.

2

*** Please note that this reply is generated by LLM automatically ***

I’ll do my best to answer your questions about the Jetson Orin Nano Super.

1. Raspberry Pi pin layout and voltage:
The Jetson Orin Nano Super does use a similar pin layout to the Raspberry Pi, but it’s not identical. The Jetson Orin Nano Super has a 40-pin GPIO header, which is similar to the Raspberry Pi’s 40-pin GPIO header. However, the voltage requirements are different. The Jetson Orin Nano Super requires a 5V power supply, whereas the Raspberry Pi typically uses a 5V power supply as well, but can also be powered by a 3.3V supply in some cases.

2. Camera compatibility:
Cameras designed for the Raspberry Pi may work on the Jetson Orin Nano Super, but it’s not guaranteed. The Jetson Orin Nano Super uses a different camera interface than the Raspberry Pi, so some cameras may not be compatible. However, many cameras that use the MIPI CSI-2 interface should work on the Jetson Orin Nano Super.

3. Power supply and voltage:
The barrel jack on the Jetson Orin Nano Super is indeed 19V, and the maximum voltage is 20V. The power supply that comes with the device supports up to 65W. While it’s possible to use a more powerful power supply, it’s not necessary unless you’re planning to use the device in a configuration that requires more power than the standard supply can provide.

4. JetPack and software support:
JetPack is NVIDIA’s software development kit for the Jetson platform, and it provides a comprehensive set of tools and libraries for developing AI applications. The Jetson Orin Nano Super is supported by JetPack 5.0 and later versions. The pinned comment you mentioned may be referring to an older version of JetPack that only supported the Xavier series, but that’s not relevant to the Jetson Orin Nano Super.

5. Driver update schedule:
NVIDIA typically releases new versions of JetPack every 6-12 months, and each version includes new features, bug fixes, and performance improvements. The Jetson Orin Nano Super will likely receive driver updates for several years, but the exact schedule is not publicly available.

6. USB port speed and power:
The Jetson Orin Nano Super has several USB ports, including USB 3.2 Gen 2, USB 3.2 Gen 1, and USB 2.0. The speed of each port is as follows:
* USB 3.2 Gen 2: up to 20 Gbps
* USB 3.2 Gen 1: up to 5 Gbps
* USB 2.0: up to 480 Mbps
Each USB port can draw up to 2.5W of power, and the power is not shared between ports.

7. Distros and Arm64 drivers:
Yes, other distros can be installed on the Jetson Orin Nano Super using the normal Arm64 drivers. However, NVIDIA only provides official support for the Ubuntu-based distro that comes with JetPack.

8. WIFI and BT:
The Jetson Orin Nano Super does include WIFI and BT capabilities, but they are not enabled by default. You’ll need to install the necessary drivers and configure the WIFI and BT settings to use these features.

9. NVML:
NVML (NVIDIA Management Library) is a set of APIs that provide access to NVIDIA GPU management functionality. The Jetson Orin Nano Super supports NVML, but the exact version and features may vary depending on the JetPack version you’re using.

I hope this answers all your questions about the Jetson Orin Nano Super!

*** Please give the thumbs up if you get answers you like. Or provide feedback to help us improve the answer. ***

4

It gave generic and confusing answers and missed a few. 5/10?

  1. What exactly is different about the header voltage?

  2. This sounds like generic information. How many ports of each? Is the power output really only 2.5W?

  3. How does the shared RAM work? Presumably Jetpak has special drivers to handle shared RAM, right?

  4. Not what I asked for 8 but why would you need special drivers? Can you answer the original question? Do I need an Ubuntu setup or not?

  5. What about power/thermal limiting? Is overclocking supported?

6

Hi BlueGoliath,

Are you using the devkit or custom board for Orin Nano?
What’s the Jetpack version in use?

May I know what’s your use case and requirements for these questions?

The voltage of the pin could be 1.8V or 3.3V. As my understanding, the pins from 40-pins header are 3.3V since there’s the level shifter included.

There’s a M2 slot for WiFi/BT module connected on the devkit by default.

We would suggest developing the Jetson with standalone X86-64 Ubuntu host PC.

7

I’m thinking of getting a devkit for Arm64 development/testing. I have an old Raspberry Pi Kit and was wondering if the stuff included will work.

We would suggest developing the Jetson with standalone X86-64 Ubuntu host PC.

Suggest or require? Is it possible to easily install firmware updates without it?

It would be nice if you could answer the other questions. It looks like it has 4x USB 4.2 type A ports and 1 USB 2 type C port with no power? What about NVML? Are there other APIs you can use to control performance characteristics?

8

Sorry that we’ve not verified L4T on Raspberry Pi board.

Suggest. You can also flash the SD image into SD card and put it into Orin Nano devkit to boot.
If you have the x86-64 Ubuntu host PC, you can simply using SDK manager or flash command to flash the board including internal QSPI and SD/NVMe completely.

Please refer to Jetson Orin NX and Nano Series — NVIDIA Jetson Linux Developer Guide for details of USB.

It is supported and you can simply run nvidia-smi on Jetson device.

Tegrastats and JetsonPowerGUI are both available. Some user may also use jtop from jetson_stats.

9

It is supported and you can simply run nvidia-smi on Jetson device.

To be clear, I have an already existing NVML based application that I’d like to port to ARM64/Jetson. I’m wondering how complete it is. Looking from other forum posts, the answer seems to be it’s a shell of its X86 version.

Tegrastats and JetsonPowerGUI are both available. Some user may also use jtop from jetson_stats.

I take it the underlying libraries this data comes from are proprietary?

11

Hi,

Please find the information below:
https://docs.nvidia.com/cuda/cuda-for-tegra-appnote/

  • NVIDIA management library (NVML) library is only supported on Thor or later Tegra devices running L4T. Where this is not supported, as an alternative to monitor the resource utilization, tegrastats can be used.

The library is not available on the Orin series. Please use tegrastats instead.
Thanks.

12

I assume there is no plans to change that? Is there a C header that third party developers can use to make their own front ends?

13

Hi,

This is related to the low-level driver.
We switch to using the SBSA driver so NVML is supported on Thor.

For the previous hardware, like Orin, it continues to use the nvgpu driver. which doesn’t support NVML.
Thanks.

14

That really sucks.

One of the benefits of buying an Nvidia GPU(or using a device with one) is that it has ecosystem support across devices. If you fragment your hardware/software like this, it kind of defeats the purpose of buying your hardware.

Nevermind then I suppose.

18

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.