Nvidia's DGX Spark mini PC, a machine built for AI workloads and research, has been tested with blockbuster games like Cyberpunk 2077. The result is clear: gaming performance from this $4,000 machine is disappointingly poor, which underlines the major architectural gap between compute-focused hardware and graphics hardware built for entertainment. The device shipped after a long delay, and buyers should understand that its main purpose is running AI models, not conquering games that demand high frame rates.
The Technical Challenge: Running Cyberpunk 2077 on a Non-Specialized ARM Chip
Gamer and Reddit community member Retrotom took on a complicated technical challenge. He managed to get Cyberpunk 2077 running on the DGX Spark, a device that uses an ARM chip and runs the Linux operating system. To do it, he had to use Box64, an x86-to-ARM translation layer. It acts as a bridge, allowing Linux programs compiled for the x86_64 architecture to run on the DGX Spark's ARM platform. The process demands deep expertise, including compiling Box64 yourself and working through complicated scripts to install Steam and launch the game.
The technical feat worked, but the computing performance was a letdown. According to Retrotom's report, running the game at 1080p with medium graphics settings, the frame rate hovered around only 50fps. That is especially notable because the DGX Spark does not support DLSS, the technology created to improve performance on Nvidia GPUs in demanding games. Stability was not perfect (there were a few freezes), but overall the system held up reasonably well.
Retrotom said he enjoyed the process and called playing x86 PC games on an ARM platform a "fun" experience. Still, he offered a strongly worded warning: "Don't waste $4,000 on a DGX Spark if you plan to game." Clearly, investing in dedicated AI hardware does not mean owning a powerful gaming machine.
Analyzing the Limits: Why AI Power Doesn't Translate Into FPS
Other tests back this up. The YouTube channel ETA Prime used the RPCS3 emulator to run Skate 3 (an older PS3 game) on the MSI EdgeXpert (a Spark-based device) and managed only about 60 FPS at 1080p. This highlights the DGX Spark's limits as an entertainment machine. The problem is that the architecture of the AI chips in the DGX Spark is optimized for the parallel calculations needed by large language models (LLMs) and data training, not for real-time 3D graphics rendering.
https://www.youtube.com/watch?v=aKcmx1cWTHc
With a $4,000 budget, gamers have far better options from dedicated gaming PC brands like Alienware or Origin. Even building a DIY PC with high-end components (such as an RTX 5090 and a Ryzen 7 9800X3D) would deliver much higher performance per dollar. This test is a reminder that while AI chips are the future of computing, traditional GPUs remain the irreplaceable foundation for entertainment and gaming.
