Hacker News Reader

Comment by cogman10 | Hacker News Reader

I strongly considered it, but the one thing that scares me away from wanting to do the spark is you basically have to use nvidia's linux (from what I've read) and it doesn't appear the nvidia is interested in upstreaming their kernel changes.

I'm avoiding where possible buying electronics where support is controlled by the manufacturer and not me.

View on HN

Many other distros almost work out of the box (as in they boot and run without any modifications). The custom kernel patches you mentioned address mainly non-critical bugs such as a bug where the Realtek r8127 stops working after a reboot (but it works if you turn it off and on again) [1] [2]. I'd consider it in a way better state than trying to run other Linux distros on certain device tree-based devices like, say, Qualcomm Snapdragon machines. The regular NVIDIA drivers with the open source kernel modules work just fine. Talos Linux supports DGX Spark since version 1.12 [3]. I also know of people using Fedora and nixOS successfully.

[1] https://github.com/NVIDIA/NV-Kernels/compare/ea55925ab430f1e...

[2] https://forums.developer.nvidia.com/t/realtek-r8127-ethernet...

[3] https://github.com/siderolabs/talos/issues/12170

Why are those patches not in mainline? They appear to be generic enough not to give out any Nvidia secret sauce.

iirc the bug that the R8127 patch was addressing was fixed in the R8169 driver upstream in kernel 6.18 and later. Not super sure what other patches there are and what the status of those are...

If I understand your comment correctly, I think he addresses that in the first few paragraphs:

> The DGX Spark runs “DGX OS” but it is in fact just plain old Ubuntu 24.04 with some additions. If you want, you can just install another Linux distribution easily (Fedora works well), although there may be a couple of weird bugs with the Realtek Ethernet driver so the NVIDIA version of the Linux kernel has a couple of patches. Unlike some other ARM devices, the DGX Spark is all ACPI rather than device tree based, so regular Linux builds for arm64 work just fine.

Well that is better than what I gleened.

I thought I'd read that the GPU needed extra kernel patches to properly work. If it's just the Ethernet driver that seems a lot more appealing.

I thought it was device tree as well, so great that it's actual ACPI.

A good instinct. There are a lot of things a $500 AMD GPU can do in linux that a $5000 DGX cannot.

Although one of the ones it cannot do is have/address 128GB of video memory, so it depends on what you want to achieve.

AMD Strix Halo was half the price at launch, same 128GB unified memory and nearly identical memory bandwidth. It is x86, so run whichever OS your heart desires.

You give up 200GB networking, but anyone 'daily driving' one of these things are probably not needing to cluster multiples of them at high speed..

I'm seeing it for 4k and up for similar specs as the spark?

Yes, due to the huge increase in memory price and also exploiting the demand for local LLM inference, the price for a mini-PC with Strix Halo has become more than double in comparison with the first months after launch (when you could pay something like $1800 for a system with 128 GB).

For those who are interested mainly in GPU performance, the NVIDIA system may be preferable. Otherwise, the Ryzen CPU is significantly stronger in multi-threaded performance, much stronger when running applications that are aware of AVX-512, so Strix Halo is preferable when the CPU performance matters.

Also for things like video decoding or transcoding, the DGX Spark GPU is likely to be weak.

>You give up 200GB networking, but anyone 'daily driving' one of these things are probably not needing to cluster multiples of them at high speed..

I mean that's not as true as you'd think: in fact its the whole benefit IMO of the spark-alikes, and if you're not running TP=2 or above you're sort of wasting it's potential

DGX does parallel inference with llamacpp/vLLM much better than AMD GPU at 128GB VRAM.

I know nothing about this topic but your comment piqued my curiosity. What would a $500 AMD GPU do better than DGX?

AMD has Mesa drivers for graphics, which are better-optimized than Nvidia's proprietary Linux Vulkan drivers. It can be fixed in software, but Nvidia's only barely started to catch up.

The focus for Nvidia's GPU stack on Linux is getting CUDA working, which means that some traditional raster features get neglected.

NVIDIA also cares about Hollywood and CAD/CAM shops.

I learnt three times that AMD isn't necessarily better in Linux as advertised, now I don't care any longer.

I'm not aware of any commercial CAD/CAM software running Linux that Nvidia would have a money interest in supporting. Which current CAD/CAM is running on Linux with graphics? The closest I'm aware of it FEA packages allowing remote calculation on Linux machines while the GUI runs on Windows. Is that what you're referring to?

BricsCAD, ARES Commander and Platicity come to mind.

A lot of Hollywood studios use renderers that jump straight to CUDA as their backend. Cycles is a popular one, as well as Renderman and Octane.

Linux is part of the VFX reference platform, and you forgot Hyperion, for example.

Fair. Honest caveat - I keep seeing "a good instinct" everywhere now. Is this humans acquiring new phrases from Claude? Is there a name for this phenomenon yet?

If not I propose HLRF.

Nice

Reply on news.ycombinator.com