RTX Spark laptop Linux inference performance is limited
What happened
On November 4, a review of an Nvidia RTX Spark laptop on the Hacker News front page said encrypted firmware leaves native Linux without power management support, so on AC power the GPU clock drops to about 780 MHz and inference throughput falls sharply. Running GPT-OSS-120B on an MSI Prestige, the review measured 65.4, 65.1 and 13.9 tokens per second on Windows, WSL and native Linux. WSL came close to Windows, while native Linux was clearly lower.
Written by AI from the coverage · updated 3 hours ago
Coverage
Follow the reports to see the story from different sides.
- Hacker News front pageNvidia RTX Spark encrypted firmware slows inference speed to 35% on Linux
Nvidia RTX Spark 笔记本在原生 Linux 下接电运行时，GPU 因电源管理支持缺失降至约 780 MHz，推理吞吐大幅下降。MSI Prestige 上的 GPT-OSS-120B 测试中，Windows、WSL 和原生 Linux 分别达到 65.4、65.1 和 13.9 tokens per second。
Heat over time
Heat now 9·Comparable peak 9（Nov 4 20:00）·Comparable change over 24 hours –
The trend only compares accounts observed without gaps, so its range may be smaller than the current heat. Hover or tap the chart for each hour; the left and right arrow keys step through it.