Hello deepin users,

As a Global Ambassador deepin Indonesia, I want to share an exciting experience from my deepin presentation to my classmates in my study program today. Although the main focus was deepin basics, I specifically highlighted the superior handling of NVIDIA Optimus (dual graphics: integrated and dedicated) compared to my experience on Ubuntu.

Hardware Context

My Lenovo laptop has a common configuration where the external HDMI port is hardwired directly to the dedicated NVIDIA GPU. This is a common configuration that often causes efficiency problems in Linux.

GPU Handling Differences

1. Experience on Ubuntu: On Ubuntu (and many other distros with classic driver settings), to ensure my HDMI port works for an external monitor, I must set the NVIDIA driver to "NVIDIA (Performance Mode)" or "GPU Only".
   Why is this mandatory? Because the HDMI port is wired directly to the NVIDIA card, the X server must be forced to run using the NVIDIA GPU as the primary renderer to access and send a signal to that specific port. This sacrifices efficiency: the dedicated GPU is always on, leading to:
   • Faster battery drain.
   • Increased laptop temperature.
   • Higher overall power consumption.
2. The deepin Solution (On-Demand Works Perfectly!): In deepin, I was very impressed. I can use the "On-Demand" mode (the most power-efficient mode), and the HDMI port still works normally and perfectly! deepin appears to implement NVIDIA Prime Render Offload or a similar mechanism exceptionally well. This means the main system runs on the integrated GPU (power saving), but it intelligently activates and deactivates the NVIDIA GPU ONLY to handle the HDMI output when a monitor is plugged in, without needing to switch the global mode.

This is a very significant efficiency boost and proves that deepin has highly optimized power management and NVIDIA Optimus driver handling without sacrificing external port functionality.

Unfortunately, time was limited during the presentation, but I managed to get photos from my classmate to document the settings I used.

I am very happy with deepin's capability here. Does anyone have similar experiences or additional insights on NVIDIA Optimus management in deepin? Let's discuss it here!