New Jetson software, modules and pricing

Thanks for the updated!

Hi all, DeepStream 4.0 was released yesterday (including support for Nano), see here for more info: [url]https://developer.nvidia.com/deepstream-sdk[/url]

Must I install DeepStream 4.0 from a Linux host running SDKManager? I am using Jetson Nano and install system image from a Micro-SD image.

You can install it on top of your SD card image from the tarball or Debian package found on the DeepStream Downloads page.

See here for the install documentation: https://docs.nvidia.com/metropolis/deepstream/4.0/dev-guide/DeepStream_Development_Guide/deepstream_quick_start.html#wwpID0E01E0HA

Hi,

When will the Jetson Nano Module (i.e. not part of developer kit) be available for purchase? The website ([url]https://developer.nvidia.com/embedded/jetson-nano[/url]) still says June.

Thanks.

Hi acsmith, thanks for pointing that out - we’ll update the page. The latest timeframe for Nano module availability is mid-to-late August - thanks for your patience.

Thanks dusty. Those modules will be following the current documentation and schematics (version B01)?

Hi acsmith, yes they will be, that is correct.

They should allow to first boot and auth on TTL like with Raspberry Pi or Orange Pi.
I personally don’t have hdmi screen (only VGA with hdmi adaptator) so I can’t first boot my JetPack.

@mrtom31 If you use the NVIDIA NANO DLI course image you can use Jupyter to interact with the Nano with a networked browser. You don’t need a monitor, mouse or keyboard for the Nano.

You can sign up for the (free) course here https://news.developer.nvidia.com/get-started-with-ai-on-the-nvidia-jetson-nano-dli-course/

@romilly.cocking Yeah already did ty. But I prefer to use the JetPack and either connect through ssh or use my current screen after I installed drivers. I mean with my raspberry Pi I could either connect through TTLtoUSB, ssh or VGAtoHDMI screen on first boot.

Here the only thing I can do is using DLI flash. and go through Jupyter browser.

@mtrom31 I understand your frustration. OTOH the Pi has had 7 years to get its act together, and you still can’t use its GPU easily :)

@romilly.cocking I thought for rpi3 it was more like a chipset graphic something like a VPU only used to run the OS. I think it has never been intended to be used as a GPU.

@mrtom31 You’re right, but many of us would have loved to use it for neural network stuff. Of course we’d have needed more RAM on the Pi as well.

Hi all, the commercial Nano module is now available through Arrow for custom designs:

[url]https://www.arrow.com/en/products/900-13448-0020-000/nvidia[/url]

Other distributors should begin stocking the Nano module in coming weeks.