4 ms·

Windows will boot without a monitor, or at least, it used to, not sure about Windows 11. But Steam Link mirrors your display, and so doesn’t work without one.

gaudystead 2y ago
I built a NUC running Windows 11 into a tiny portable server for a project I was building and can confirm it boots and functions just fine without being plugged into a monitor. I just plug it into a power source and it does what I need it to do, but I can plug a monitor and keyboard (and sometimes a mouse because keyboard-only navigation seems to be getting less and less supported/intuitive...) if I need to perform troubleshooting.
- dicknuckle 2y ago
  The issue is usually with the graphics card itself in my experience. This is easily "fixed" on a DVI port by plugging a resistor of the correct value into two of the tiny pin sockets. The diagram is very easy to find online and you don't have to open the computer. That's become a thing of the past as far as I know.
  - withinrafael 2y ago
    NVIDIA GPUs still require "dummy" HDMI plugs. They won't be fully usable without something plugged in; it's not clear why.
    - rincebrain 2y ago
      This always seemed to be a very deliberate design choice by them to avoid you being able to use their consumer cards headless versus paying them a large amount on the Quadro or DG cards, since the big problem we saw at $OLDJOB was always that you couldn't use CUDA on them headless. At said $OLDJOB, we ended up soldering dummy VGA plugs that had resistors across the right pins when we wanted to experiment with building a low-power cluster of NVIDIA Ion boards and seeing how it competed with big cards. Ah, memories.
      - alias_neo 2y ago
        Would bet that this is exactly why. I run Tesla GPUs in my server rack which don't even have display ports, but they run any OS just fine with the vGPU drivers, which Nvidia make an absolute pain to obtain.
        rincebrain 2y ago
        The _very_ first gen of Tesla cards did have those headers on them, IIRC, and then successive ones had the headers on the board but not connected for another generation or so, IIRC. You also used to be able to edit the PCI IDs for the drivers to get the Tesla ones to attach to consumer GPUs, but that stopped working at some point.
  - TiredOfLife 2y ago
    https://github.com/timminator/Enhanced-GPU-PV https://github.com/timminator/Enhanced-GPU-PV
wongarsu 2y ago
Can confirm. I boot Windows 11 without screen, then connect via RDP without issue
TiredOfLife 2y ago
https://github.com/timminator/Enhanced-GPU-PV https://github.com/timminator/Enhanced-GPU-PV or https://github.com/ClassicOldSong/Apollo https://github.com/ClassicOldSong/Apollo (sunshine fork with auto virtual display) note that RDP works without plugs/virtual displays.

gaudystead 2y ago

I built a NUC running Windows 11 into a tiny portable server for a project I was building and can confirm it boots and functions just fine without being plugged into a monitor. I just plug it into a power source and it does what I need it to do, but I can plug a monitor and keyboard (and sometimes a mouse because keyboard-only navigation seems to be getting less and less supported/intuitive...) if I need to perform troubleshooting.

dicknuckle 2y ago
The issue is usually with the graphics card itself in my experience. This is easily "fixed" on a DVI port by plugging a resistor of the correct value into two of the tiny pin sockets. The diagram is very easy to find online and you don't have to open the computer. That's become a thing of the past as far as I know.
- withinrafael 2y ago
  NVIDIA GPUs still require "dummy" HDMI plugs. They won't be fully usable without something plugged in; it's not clear why.
  - rincebrain 2y ago
    This always seemed to be a very deliberate design choice by them to avoid you being able to use their consumer cards headless versus paying them a large amount on the Quadro or DG cards, since the big problem we saw at $OLDJOB was always that you couldn't use CUDA on them headless. At said $OLDJOB, we ended up soldering dummy VGA plugs that had resistors across the right pins when we wanted to experiment with building a low-power cluster of NVIDIA Ion boards and seeing how it competed with big cards. Ah, memories.
    - alias_neo 2y ago
      Would bet that this is exactly why. I run Tesla GPUs in my server rack which don't even have display ports, but they run any OS just fine with the vGPU drivers, which Nvidia make an absolute pain to obtain.
      - rincebrain 2y ago
        The _very_ first gen of Tesla cards did have those headers on them, IIRC, and then successive ones had the headers on the board but not connected for another generation or so, IIRC. You also used to be able to edit the PCI IDs for the drivers to get the Tesla ones to attach to consumer GPUs, but that stopped working at some point.
- TiredOfLife 2y ago
  https://github.com/timminator/Enhanced-GPU-PV https://github.com/timminator/Enhanced-GPU-PV

wongarsu 2y ago

Can confirm. I boot Windows 11 without screen, then connect via RDP without issue

TiredOfLife 2y ago

https://github.com/timminator/Enhanced-GPU-PV https://github.com/timminator/Enhanced-GPU-PV or https://github.com/ClassicOldSong/Apollo https://github.com/ClassicOldSong/Apollo (sunshine fork with auto virtual display) note that RDP works without plugs/virtual displays.