←back to thread

Nvidia's Project Digits is a 'personal AI supercomputer'

(techcrunch.com)
623 points magicalhippo | 6 comments | 07 Jan 25 04:14 UTC | HN request time: 0s | source | bottom
Show context
ryao ◴[07 Jan 25 06:44 UTC] No.42619962[source]▶
This looks like a successor to the Nvidia Jetson AGX Orin 64GB Developer Kit:

https://www.okdo.com/wp-content/uploads/2023/03/jetson-agx-o...

I wonder what the specifications are in terms of memory bandwidth and computational capability.

replies(3): >>42619979 #>>42620718 #>>42622001 #
kcb ◴[07 Jan 25 06:47 UTC] No.42619979[source]▶
Hopefully, the OS support isn't as awful as the Jetson platforms usually are. Unless they change, you'll get 1 or 2 major kernel updates ever and have to do bizarre stuff like install a 6 year old Ubuntu on your x86 PC to run the utility to flash the OS.
replies(1): >>42620030 #
1. ryao ◴[07 Jan 25 06:56 UTC] No.42620030[source]▶
The community likely will make instructions for installing mainstream Linux distributions on it.
replies(1): >>42620055 #
2. kcb ◴[07 Jan 25 07:01 UTC] No.42620055[source]▶
Doesn't really help though if it requires an nvidia kernel.
replies(2): >>42620082 #>>42620257 #
3. ryao ◴[07 Jan 25 07:05 UTC] No.42620082[source]▶
The Linux kernel license requires Nvidia to disclose their Linux kernel sources and Nvidia open sourced their kernel driver.

That said, you can probably boot a Debian or Gentoo system using the Nvidia provided kernel if need be.

replies(1): >>42620677 #
4. snerbles ◴[07 Jan 25 07:39 UTC] No.42620257[source]▶
The official Linux kernel driver for Blackwell is GPL/MIT licensed: https://developer.nvidia.com/blog/nvidia-transitions-fully-t...
replies(1): >>42620462 #
5. sliken ◴[07 Jan 25 08:21 UTC] No.42620462{3}[source]▶
Keep in mind that a kernel module != driver. It's just doing initialization and passing data to/from the driver, which is closed source and in user space.
6. bionade24 ◴[07 Jan 25 09:03 UTC] No.42620677{3}[source]▶
It always has been the userspace of the Jetsons which was closed source and tied to Nvidia's custom kernel. I have not heard from people running Jetpack on a different userland than the one provided by Nvidia. Companies/Labs that update the OS don't care about CUDA, Nvidia contributes to Mesa support of the Jetsons and some only need a bit more GPU power than a RasPi.