Reading through the existing answers on this board, I had a look at my /var/log/Xorg.0.log and would see this when both external screens go black for a couple seconds:
[ 34570.436] (II) systemd-logind: got resume for 13:76
[ 34570.436] (II) event12 - Video Bus: is tagged by udev as: Keyboard
[ 34570.436] (II) event12 - Video Bus: device is a keyboard
[ 51961.624] (II) event256 - PixArt Dell MS116 USB Optical Mouse: SYN_DROPPED event - some input events have been lost.
[ 51961.624] (EE) event257 - Dell KB216 Wired Keyboard: client bug: event processing lagging behind by 241ms, your system is too slow
[147298.769] (EE) event256 - PixArt Dell MS116 USB Optical Mouse: client bug: event processing lagging behind by 11ms, your system is too slow
The "client bug" thingy on the last 2 lines and/or the SYN_DROPPED on the 4th line could show the source of the issue, e.g. could be the hub or the connected devices tripping the whole connection? I couldn't test without these devices yet.
(NVIDIA drivers: 520.56.06; XOrg: 1.21.1.4 (12101004))
EDIT: I have all 3 screens (laptop's and 2x external HDMI Dell monitors) set to 60Hz refresh rate and still have the issue.
]]>I found that setting the refresh rate of the laptop and external monitor to the same rate, or turning off the laptop screen completely mitigates this issue.
So these tips probably won't help you but maybe someone else.
Random guess: Try passing "pcie_aspm=off" to the kernel.
Doesn't seem to work in my case
]]>```
$ lspci | grep VGA
01:00.0 VGA compatible controller: NVIDIA Corporation GP107 [GeForce GTX 1050 Ti] (rev a1)
```
This only happens when the screen is having too much whiteness
]]>lspci:
0b:00.0 VGA compatible controller: NVIDIA Corporation GP107 [GeForce GTX 1050] (rev a1)
Anyway, I'll continue to keep my packages updated, consider if I have some config issue, hardware issue, etc. Just posting my impression.
]]>> lspci |grep VGA
01:00.0 VGA compatible controller: NVIDIA Corporation GP106 [GeForce GTX 1060 6GB] (rev a1)
------
I'm leaving this here since I found this thread via a search engine and I'm sure others will too.
I've been using a USB Wi-Fi adapter for years. But I bought a PCIe one recently. I ordered external antennas with it but didn't receive them yet. So I've been testing with the antennas directly attached to the adapter.
I started having those flickering issues during speed tests (other network-heavy tasks would cause similar "trouble" obviously). I moved the antennas a little bit away from the HDMI wire attached to the GPU, and voila, no more flickering.
There are probably many causes for HDMI screen flickering. But this might not be an immediately obvious one.
]]> $ lspci |grep VGA
01:00.0 VGA compatible controller: NVIDIA Corporation GP107 [GeForce GTX 1050 Ti] (rev a1)
The dual monitors are both BenQ BL 2205. I suspect the problem is with a recent nvidia driver update (using ver. 460.67). I don't know how to solve that, downgrading caused some other problems to me.
Also check out this thread on nvidia formus: https://www.nvidia.com/en-us/geforce/fo … opicPage=3
]]>I'm dual booting
3rd link in my signature…
Other than that, look into the power/performance setting in nvidia-settings and maybe try to enforce the full composition pipeline, https://wiki.archlinux.org/index.php/NV … en_tearing (ignore the context, it'll prevent the GPU from powering down)
You can also try the lts kernel and/or the 390xx driver, https://aur.archlinux.org/packages/nvidia-390xx-dkms/
sometimes momentarily show a message saying the input is not supported
means there's a bogus signal and with the described context that sounds a lot like the GPU isn't drawing or getting enough power.
]]>