News for hackers. Stuff that matters.
Re:Nvidia DGX Spark encrypted firmware slows inference speed... by 999900000999 on Wednesday November 04, 2026 @08:30AM (#50365264)
>Oddly, the direction of the problem flips depending on power source. Unplugged, Linux hit around 61 to 67 tokens per second on LM Studio, comparable to Windows. Plugged in, that same test dropped to about 15 tokens per second.

Absolutely hilarious.

This is just incompetence. No reason Nvidia with its TRILLIONS of market cap couldn't of tested and fixed this on Linux before shipping.

Or a day one patch.

I have no doubt that'll it'll work right eventually. Long after RTX Spark laptops flop and can be purchased for half off.

Last night I pulled out my 5070TI laptop and installed CachyOS. I was running various models within about 20 minutes.

That 12GB of VRAM is a hard limitation, Unsloth can't split jobs between vram and system memory.

Re:Nvidia DGX Spark encrypted firmware slows inference speed... by throw1234567891 on Wednesday November 04, 2026 @08:36AM (#50365333)
> couldn't of tested

Could have tested.

Re:Nvidia DGX Spark encrypted firmware slows inference speed... by anonym29 on Wednesday November 04, 2026 @08:41AM (#50365386)
Thanks, I had absolutely no idea what GP was saying, it was completely incomprehensible until you published this Rosetta Stone here.
Re:Nvidia DGX Spark encrypted firmware slows inference speed... by throw1234567891 on Wednesday November 04, 2026 @08:55AM (#50365557)
Enjoy.
Re:Nvidia DGX Spark encrypted firmware slows inference speed... by crest on Wednesday November 04, 2026 @08:40AM (#50365367)
What makes you think this was an oversight?
Re:Nvidia DGX Spark encrypted firmware slows inference speed... by 999900000999 on Wednesday November 04, 2026 @09:31AM (#50366044)
Because they also could of just locked the bootloader fully if they didn't want Linux to work.

Even from the article I'm guessing you could write a script to automatically extract the encryption keys while on Windows and the load those in Linux.

This is only mildly inconvenient for anyone who'd use this in a data center cluster.