First screen, captured 9 days ago. Open the video.

Nvidia RTX Spark encrypted firmware slows inference speed to 35% on Linux

What HN thinks

Half the thread sees the dawn of AI DRM serfdom; the other half keeps begging everyone to read the article about a laptop driver bug.

Nvidia is pushing lock-in and DRM
Many see encrypted power firmware as another step in Nvidia's long history of crippling hardware, market segmentation and control, possibly toward licensed or cloud-dependent AI.
It's just missing Linux drivers
Others argue this is a new Windows-only laptop platform where Linux support simply hasn't shipped yet, as with Dynamic Boost, and the outrage is confirmation bias.
Run local inference at home instead
A side debate holds that laptops are a poor place for inference; put a DGX Spark or workstation at home and reach it over Tailscale.

Consensus: No real consensus on intent, but most agree encrypting the power table is hostile to Linux users and Nvidia's closed firmware approach invites suspicion.

Recommended in the comments

Tailscale (service): Lets you use home inference hardware from a lightweight laptop anywhere.
NVIDIA DGX Spark (product): Same chip with native Linux support, suited as a home inference box.
AWS Trainium (service): Offered as existing competition to Nvidia, albeit locked to AWS.
gpu-unlock ↗ (tool): Cited as a way around Nvidia's half-speed tensor core cap on GeForce cards.

Summarised by Claude from the comment thread 9 days ago.

Points: 586
Comments: 271
Best rank: #1
Posted: 2026-11-04 by novaRom
Category: Hardware

nvidia-rtx-spark linux gpu-throttling dual-boot wsl llm-inference msi-prestige

All captures:

Full page

Captured 9 days ago. Click for the whole screenshot.

Looks similar

Screenshot of 9to5mac.com: Apple has new Mac app coming to make running local AI models easy

Apple has new Mac app coming to make running local AI models easy (9to5mac.com)

466 points by linolevan 3 days ago | 481 comments

9to5Mac article reporting a leaked Apple Mac app called Harbor that runs open-weight AI models locally and clusters multiple Macs over Thunderbolt 5.

apple-harbor local-llm apple-silicon macos

Screenshot of linebender.org: JIT bitmap text rendering for microcontrollers

JIT bitmap text rendering for microcontrollers (linebender.org)

102 points by dmit 5 days ago | 22 comments

Raph Levien's Linebender blog post describing a JIT technique for fast proportional bitmap text rendering on the RP2350 microcontroller in his vic-2350 project.

rp2350 microcontrollers jit bitmap-fonts

JIT bitmap text rendering for microcontrollers (linebender.org)

5 points by raphlinus 6 days ago | 1 comments

Raph Levien's Linebender blog post describing a JIT technique for fast proportional bitmap text rendering on the RP2350 microcontroller for his vic-2350 project.

jit-compilation bitmap-fonts text-rendering rp2350