I'm having this same issue using two different PCs both with Windows 10.
- Both are iGPU only, so no NVIDIA/AMD graphics card control panel such-and-such to configure.
- One is intel CPU and one is AMD CPU, so it's not a brand specific problem.
- I have a 4k VA monitor (Sceptre) that has never had issues running at 60hz (max 60hz @4k).
- The monitor with issues is a 4k IPS monitor (KTC) that currently is set to 30hz (max 144hz @4k).
- The monitor is compatible with G-Sync but has no way to turn it off from the monitor settings.
- The monitor is the only thing plugged into a power strip other than the other monitor. this power strip previously handled multiple PCs and network devices just fine.
- Power settings for both PCs are set to max performance, and no screen turnoff.
- All drivers are freshly updated on both systems.
- Both PCs are currently using HDMI cable to connect to the monitor.
I suspect it's a compatibility issue between certain monitors or HDMI itself and Windows. My power setup is fine. No reports of faulty power on these models of monitor. Using an expensive high quality surge protector. Many are having this with a multi-monitor setup, and many are having to set their monitor to match the one with minimum refresh rate or minimum G-Sync/FreeSync settings. So it seems that Windows has an issue with using two monitors with different refresh rates and different G-Sync/FreeSync capabilities. People are reporting the problem sometimes goes away when the monitors are set to the same refresh rate, usually the highest rate of the lowest quality monitor, and sometimes it goes away when G-Sync/FreeSync are force-disabled. People also seem to run into this problem when using two different monitors as opposed to exact same model. Some have also fixed the black screen by only using one monitor. This corroborates the possibility that Windows 10/11 are incorrectly handling two monitors with different capabilities unless the settings are forced so that they match one another in terms of performance.
Multiple people have recorded an HDMI event happening in logs around the same time as the black screen. So whatever inside the Windows kernel that communicates with the HDMI firmware is having some kind of glitch and potentially triggering a 'reset' or renegotiating the initial handshake. This is why the problem persists regardless of manufacturer drivers from GPUs, CPUs, or monitors. This could only happen if it was something more integral to how Windows communicates with hardware (ie the kernel). This is probably why Microsoft is reluctant to fix it. The kernel is the main thing they want to touch THE LEAST compared to just pushing a basic security update or fudging up the UI/UX again and again. Linux users also experience this but the problems seem to be less common. I didn't experience this glitch when using both monitors with Debian, but different distros have different modifications that can affect how the basic Linux kernel treats certain hardware.
People with DisplayPort or USB-C have occasionally experienced this, but it seems to be a problem that most people are having when using HDMI (even when replacing cables, and switching to a different port). So based on all of the evidence it looks to me like Windows 10/11 have sporadic issues when communicating a complex monitor setup between multiple HDMI ports.
Please correct me if I'm off-base with any of this analysis!
One thing that I haven't seen people mention is how they have their monitor settings inside the monitor itself configured. I just turned off OverDrive and HDR inside my monitor and I'll report back whether that helps at all.