Arch Linux

orbit1849

Member

Registered: 2022-10-13

Posts: 13

Samsung G93SC goes to sleep and doesn't wake up

I have a Samsung G93SC which supports 5120x1440 at 240 Hz. This works fine on a Windows computer.

On my Arch system, even using the same physical graphics card in the Windows computer (a 4070ti), it only works at lower refresh rates, 60 and 120 Hz.

However, even at 60 Hz, it has a problem with going to sleep and not waking up again. This also only happens on the Arch system, not on Windows.

I've tried to disable everything I can in the OSD for the monitor related to sleep, power savings, adaptive sync, etc. I've tried on the host computer to disable DPMS with xset -dpms.

I've tried a different graphics card (3090 and 4070ti). I've tried three different cables (DP and HDMI, HDMI only goes up to 120 Hz though at that resolution, I'm aware of this).

I've tried different nvidia drivers (I cannot use nouveau, I need CUDA, etc.), probably 8 different drivers.

I have another smaller monitor also connected, but I've tried without that one connected, same issue.

I've tried to upgrade the firmware on the problematic Samsung monitor.

I've tried to make my own X config and fiddle around with it.

Section "ServerLayout"
    Identifier     "X.org Configured"
    Screen      0  "Screen0" 0 0
    InputDevice    "Mouse0" "CorePointer"
    InputDevice    "Keyboard0" "CoreKeyboard"
EndSection

Section "Files"
    ModulePath   "/usr/lib/xorg/modules"
    FontPath     "/usr/share/fonts/misc"
    FontPath     "/usr/share/fonts/TTF"
    FontPath     "/usr/share/fonts/OTF"
    FontPath     "/usr/share/fonts/Type1"
    FontPath     "/usr/share/fonts/100dpi"
    FontPath     "/usr/share/fonts/75dpi"
EndSection

Section "Module"
    Load  "glx"
    Load  "vnc"
EndSection

Section "InputDevice"
    Identifier  "Keyboard0"
    Driver      "kbd"
EndSection

Section "InputDevice"
    Identifier  "Mouse0"
    Driver      "mouse"
    Option      "Protocol" "auto"
    Option      "Device" "/dev/input/mice"
    Option      "ZAxisMapping" "4 5 6 7"
EndSection

Section "Monitor"
    Identifier "DP-0"
    Option "PreferredMode" "5120x1440"
    Option "Position" "0 480"
    Option "Rotate" "normal"
    Option "Rate" "60"
EndSection

Section "Monitor"
    Identifier "DP-2"
    Option "PreferredMode" "1920x1200"
    Option "Position" "5120 0"
    Option "Rotate" "left"
EndSection

Section "Device"
    Identifier  "Card0"
    Driver      "nvidia"
    BusID       "PCI:46:0:0"
    Option "AllowEmptyInitialConfiguration"
EndSection

Section "Screen"
    Identifier "Screen0"
    Monitor "DP-0"
    Monitor "DP-2"
    Device "Card0"
    Option "DPI" "96 x 96"
EndSection

Any ideas for things I could try? I can live with 60 Hz at 5120x1440, although it is quite sad considering it is capable of 240 Hz at 5120x1440 on Windows.

However, I cannot live with 3840x1080, the image is too unclear for me, and 3840x1080 is the only resolution it works reliably on so far. On 5120x1440 it goes to sleep, and never wakes up again. If I fiddle enough with xrandr, power the monitor on and off, remove and reinsert the cable, I can sometimes get it back, but this isn't a sustainable solution.

jason98893

Member

Registered: 2023-02-22

Posts: 7

Re: Samsung G93SC goes to sleep and doesn't wake up

Exact same problem. Works fine on a fresh boot, but the minute I've been gone for a while it doesn't wake and just loops over the "No Signal" message.

Did you find a solution?

Last edited by jason98893 (2024-08-12 19:09:01)

DarlCat

Member

Registered: 2025-07-23

Posts: 1

Re: Samsung G93SC goes to sleep and doesn't wake up

I just got a G93SC and stumbled across this thread looking for others having the same issue I was encountering. The findings and reported behaviors here match my issue exactly, so I want to offer the solution I was able to stumble across hoping it may help others since no resolution was ever reached in this thread.

The only others discussing this same issue with a proposed solution that I managed to find were using Windows, discussing the issue on Reddit where it was suggested that the PCI Express link state power management within Windows power saving settings could be disabled to avoid the undesirable behavior.

With this information I turned to documentation found on the wiki which educated me on adjusting ASPM. I opted to go with the performance policy supported by my system for my long-term fix because it worked well for me, but disabling the feature entirely also yielded positive results.

After activating the performance policy on my system the behavior stopped entirely and I've been a lot happier with my new monitor. To clarify for anyone revisiting this issue I'm using an Nvidia RTX 3090 on driver 575.64.03. The theory that I've adopted is that the level of GPU sleep achieved when the system sleeps the display causes issues with the Samsung monitor, and the ASPM performance policy avoids this problematic state, I'd assume while also losing the power saving benefits.

I suspect the same applies also to https://bbs.archlinux.org/viewtopic.php?id=296807