Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Nvidia announces Jetson Nano 2GB, a single board computer

medium.com

Post reply on HN
Print view View original

51–60 of 205 posts

Re: Nvidia announces Jetson Nano 2GB, a single board computer

#51

Post by barkingcat »

varispeed wrote: post #29

What is exactly the point of such devices? If you create something worthwhile with it, you cannot exactly include it in your product. I get that you can create something like a one off garage opener based on face recognition, but you won't learn any real world knowledge of developing a product and if you later on decide that this is something for you, you'll have to learn a whole new ecosystem. These devices remind m…

The Jetson Nano is the "small" SDK for the Jetson Xavier series which is Nvidia's embedded ai platform for volume (ie they will sell you thousands of units of them for volume manufacturing in cars, etc) - that's always been the draw for the Jetson Nano - you use the hardware for prototyping and when you are ready, buy the larger and more expensive Xavier and that goes into your hardware product.

https://www.nvidia.com/en-us/autonomous-machines/embedded-sy...

Re: Nvidia announces Jetson Nano 2GB, a single board computer

#53

Post by talkingtab »

I have the 4GB version of the Nano and the 4GB Raspberry Pi. I like the Nano, but I use it mostly as a development machine and for that I would rank the 8GB Raspberry Pi at about $89 US the best deal. Substantially below that is the Nano, and almost tied is 4GB Pi. There may be cases where you absolutely need an Nvidia machine in which case I would argue for the 4GB version, but otherwise the 8GB Pi appears to be the winner by a long shot.

Re: Nvidia announces Jetson Nano 2GB, a single board computer

#54

Post by my123 »

ageitgey wrote: post #46
chunkyks wrote: post #42

I have a Jetson nano. It's a nice thing for what it is, and the price. But, the software misses the mark by a lot. It's still based on Ubuntu 18.04. Want python later than 3.6? Not in the box. A lot of python modules need compiling, and the cpu is what you'd expect. Good for what it is, bad for compiling large numbers or packages. They run a fancy modern desktop based on the Ubuntu one. Sure, it's Nvidia, gotta be fl…

FWIW, I got a review unit and it ships with a much lighter desktop environment (LXDE) this time and is has a swap file enabled by default. I guess that was needed to ship a 2GB version, but overall it seemed a little better thought out. They also include a wifi dongle in the box. It still only has Python 3.6, but OpenCV and numpy are pre-installed correctly this time so you don't have to compile them which is an impr…

You can update it to Ubuntu 20.04 pretty easily. (using a Xavier as a desktop with that)

If you don't need CUDA, you can outright use regular Fedora, which uses nouveau. (no reclocking issues or such on Tegra)

Re: Nvidia announces Jetson Nano 2GB, a single board computer

#55

Post by craftinator »

catsdanxe wrote: post #33
Ftuuky wrote: post #24

No WiFi kills it for me

Buy a chinese usb wifi dongle. They are very cheap.

But don't buy the WRONG dongle. I just spent 15+ hours trying to convince the Jetson Nano that the rtl8822bu driver ought to compile on an arm64 system. It's definitely not like the Pi, where 99% of the software is open source and you don't have to deal with Nvidia's nasty hackish tooling to download special Jetson headers and such.

Re: Nvidia announces Jetson Nano 2GB, a single board computer

#56

Post by slezyr »

fgblanch wrote: post #39
slezyr wrote: post #31

I booted mine Radxa Rock (rk3188) with BareBox, ArchLinux ARM and a mainline kernel. It just took me few hundred hours to figure out all the problems on my way there. And it's one of the most open source friendly ARM SOCs. I wish luck to everyone buying these things. You'll need it to run any modern distro in a few years.

Very recently I recovered my rock too. I'm about 20 hours in. Any repository or helpful documentation to save part of the 80 remaining hours? pleaaaasee

1.) Forget about EMMC, it's gone and not supported by anyone.

2.) Boot is only possible from microSD card. USB is not supported.

3.) Build BareBox with https://www.barebox.org/doc/latest/boards/rockchip.html

Use `rk-splitboot` to split FlashData from the original boot loader, other option doesn't work. Build it, it will create barebox-radxa-rock.img. It's .img, not the .bin.

Write an SD card as in the instruction.

4.) Follow these instructions, but use the generic image.

https://archlinuxarm.org/platforms/armv7/amlogic/odroid-c1

http://os.archlinuxarm.org/os/ArchLinuxARM-armv7-latest.tar....

5.) Build kernel as in this tutorial, ignore the `dts` stuff. The `dts` comes with the kernel sources.

https://wiki.radxa.com/Rock/Linux_Mainline

https://github.com/torvalds/linux/blob/master/arch/arm/boot/...

6.) Copy the zImage, rk3188-radxarock.dts to the /boot and plug in the UART.

7.) sudo picocom /dev/ttyUSB1 -b 115200 -e w

8.) Boot in the bootloader and execute the following commands:

global.bootm.appendroot=true

global.bootm.image=/mnt/mshc1.0/boot/zImage

global.bootm.oftree=/mnt/mshc1.0/boot/dtbs/rk3188-radxarock.dtb

bootm

9.) If it boots, you can make it permanent by following the BareBox docs.

Re: Nvidia announces Jetson Nano 2GB, a single board computer

#57

Post by gnarbarian »

CyberDildonics wrote: post #17
nl wrote: post #13

Earlier quoted context omitted.

Things like video analysis for security cameras for people who don't want to stream everything to the cloud. It's possible on a single board computer CPU but only with lots of compromises - you can't do multiple streams etc.

Why wouldn't someone just use a PC for that since it doesn't need to be portable?

Convenience. If you have a security camera up 50 ft on a tower 40 miles away connected to the internet via cell modem where do you put the PC? And is the PC able to fit in the power envelope and put up with environmental factors? An sbc is much easier to harden environmentally

Re: Nvidia announces Jetson Nano 2GB, a single board computer

#59

Post by ageitgey »

varispeed wrote: post #29

What is exactly the point of such devices? If you create something worthwhile with it, you cannot exactly include it in your product. I get that you can create something like a one off garage opener based on face recognition, but you won't learn any real world knowledge of developing a product and if you later on decide that this is something for you, you'll have to learn a whole new ecosystem. These devices remind m…

These get a lot of use in limited-run industrial applications where you need something custom and smaller/lower power than a desktop computer but you aren't manufacturing thousands of devices and you don't have millions of dollars to develop custom hardware. You can take pretty much any model you trained on a desktop GPU and run it on the board without any changes which is nice for the right kind of project. And they sell a range of modules with a spectrum of price/performance/power draw depending on how much processing power you need.

If you are manufacturing something in larger volume, they will sell you production modules without the dev boards that you can mount on your own boards. But those are still aimed at more expensive products. So this isn't a solution for low-cost high volume consumer products, but there is definitely a market for it.

Post reply on HN