Report: Nvidia RTX Spark firmware tools limit GPU speed on native Linux
A MindStudio report says RTX Spark laptops are capped on native Linux without a Windows-only firmware component, cutting GPT-OSS-120B speed from about 65 to 14 tokens per second.
Laptops built around Nvidia's RTX Spark GPU can run native Linux with Nvidia's open kernel driver, CUDA and 4K/120Hz output, according to a MindStudio report. The plugged-in GPU, however, is held to a fraction of its Windows speed unless a Windows-only component is present. Without the Nvidia Platform Controllers and Framework component, the report says the GPU is capped around 780 MHz, against about 2.2 GHz on Windows. On GPT-OSS-120B, that cuts throughput from about 65 tokens per second on Windows or WSL to about 14 on native Linux.
The report says the Windows component manages the firmware's power tables. A Reddit discussion of the report adds that the component decrypts those tables and controls how power is shared between the CPU and the GPU. Without that component, the report says, a native Linux install stays under the lower clock cap described above.
The numbers in the discussion's title do not quite match the throughput figures. The title gives 35 percent, but 14 tokens per second is about 22 percent of 65.
Commenters in the discussion split on why the limit exists. Some suspect Nvidia is deliberately restricting Linux to protect the relevance of Windows and WSL. Others argue that if Windows can decrypt and manage the firmware on the same hardware, the required key handling or protocol behaviour should be observable, and Linux tooling could eventually reproduce it. These are the commenters' views. The report and the discussion do not establish intent or show that a Linux workaround exists.
The report also does not say whether the Windows component can be run or replaced on Linux. That question decides whether the cap is a driver gap or a firmware lock.
What changed
On native Linux, RTX Spark laptops' GPU is capped around 780 MHz without the Windows-only Nvidia Platform Controllers and Framework component, against about 2.2 GHz on Windows.
Type: Other · Applies to: Nvidia RTX Spark, Nvidia Platform Controllers and Framework
Why it matters
Teams planning to run local inference on Linux with this hardware may see large throughput losses that depend on a Windows-only component.
What you can do
Before buying RTX Spark hardware for Linux inference, benchmark it on the exact OS setup you plan to use.
Original source
We have no link to the original source for this story yet.
Where we saw it
Latent Space
AI wrote this summary from the sources listed, and nobody reviewed it individually before publication. Check the original source before you rely on it. How it works