Unstable Pc performance and lack of visual smoothness.

[Psi]

I've not read all pages, has anyone suggested recording gameplay with NVIDIA using the built-in GeForce Experience software/ShadowPlay and seeing if the issue is in the recording or not?

Another thing that occurs to me is unstable or glitchy timing for the 50/60hz mains freq and therefore any part of the computer or screen trying to handle this noise by syncing to it or avoiding conversions during a cycle is failing to do so properly.
A laptop running on DC/battery with NO connection to the mains in ANY way should be able to rule out that theory if it's still happens.

Another thing to check might be for any strong radioactive sources nearby. Might be a bit extreme, but who knows what effect that might have on a PC.

[Manul]

The thing is, such effects due to external interference are almost impossible, especially if consistent and not just a glitch in one frame out of a million. Any memory corruption events would manifest most notably as system crashes and are pretty random. Glitching digital video transmission (like hdmi link) would also manifest quite differently, by select variety of video defects or a total loss of video. I suggest to not go into "moon phases affect my frame rates" type of theories, it will be just a waste of time and a distraction from real issues.

it might be just an EMC, not the "moon phases affect my frame rates" type of theories.

I don't mean it to sound insulting, I mean it as a guidance. Source: I design, test and troubleshoot electronic circuits for over 20 years. You probably mean EMI, not EMC. Anyway, that is exactly my point. EMI causing such effects consistently, repeatably and across many devices is "moon phases" theory.

Gaming is demanding, I understand that. Every millisecond counts. But that's far from the only mission critical application for a PC. There are industrial, medical, aeronautical, defense applications where PCs are running mission critical stuff, sometimes in really harsh environments. And PCs are not some black magic put together, these systems are designed by humans and well understood, if it would not be understood, we would never make it work in the first place.

Sure there are edge cases of defective products, like a specific model of motherboard with unstable PLL or something. So you touch the case with your hand and it starts working. Or stops working. Or whatever. But these are again edge cases of serious design errors. But that does not sound at all like your case.

And we humans often have really hard time analyzing something and drawing conclusions. In fact, we are often really really bad at that. That's how historically we had so many superstitions (and still do). One time someone came up with an idea to sacrifice a young virgin girl and it happened that crops grew very well. So after that they started sacrificing a young girl every year.

Perhaps off topic, but this link is a really nice resource for an evening reading:

https://en.wikipedia.org/wiki/List_of_cognitive_biases

Anyway, in conclusion, as I said, too consistent to be EMI issue. In specific, isolated edge cases we could talk about that, but not here. And it's a digital system after all. It's not like we have sound recording studio and "my noise floor is -115dB with LED lights ON, -120dB if lights are OFF and -121dB if the LED lights are OFF and the microphone is facing north". Such effects are common in high-end analog systems, but not digital.

[Jeroen3]

I would recommend running some traces with the Windows Performance Recorder / Analyzer.

If there are recoverable errors somewhere in your system these should be logged somewhere.
So I would suggest looking into the event logs and turn on as much as possible.

Things that possible are not logged are issues in the CPU power state switching, but an EMI susceptibility on multiple platforms would be rare.

Anyway, Windows is terrible at handling IO errors. For example if an nvme drive craps out it will just BSOD, while with Linux you get entries in the kernel log and only that specific device will just stall. Contrary to windows where the entire io layer will stall.
So it may very well be as simple as a stalling SSD or misbehaving USB device.

If you expect it to be noise from the grid you can measure power quality, these meters are expensive though, maybe rent one.
However, I know from someone in Kyiv that power quality is abysmal right now. Not the most important thing when your infrastructure is getting shot to bits.

From the other angle, could you have an EMI susceptibility?

[Inputdelay]

The thing is, such effects due to external interference are almost impossible, especially if consistent and not just a glitch in one frame out of a million. Any memory corruption events would manifest most notably as system crashes and are pretty random. Glitching digital video transmission (like hdmi link) would also manifest quite differently, by select variety of video defects or a total loss of video. I suggest to not go into "moon phases affect my frame rates" type of theories, it will be just a waste of time and a distraction from real issues.

This is exactly what I am thinking. For something in the environment to cause this specific problem is inconceivable. It just is. I do not accept that we are looking at exactly the same hardware and software - just in two different places.

Inputdelay: how do you explain the obvious judder in the second half of the video?

If only I knew what the cause of it was, I would gladly tell you, but the only change is two different places; there is nothing to add.

[Inputdelay]

I've not read all pages, has anyone suggested recording gameplay with NVIDIA using the built-in GeForce Experience software/ShadowPlay and seeing if the issue is in the recording or not?

Another thing that occurs to me is unstable or glitchy timing for the 50/60hz mains freq and therefore any part of the computer or screen trying to handle this noise by syncing to it or avoiding conversions during a cycle is failing to do so properly.
A laptop running on DC/battery with NO connection to the mains in ANY way should be able to rule out that theory if it's still happens.

Another thing to check might be for any strong radioactive sources nearby. Might be a bit extreme, but who knows what effect that might have on a PC.

I would suggest reading all the pages in the first place. And yes, I have recorded gameplay, and there is not a single logical reason for doing that because the problem on screen and on record is completely different. If you just check the recorded video, you probably won't even notice anything suspicious.

I did check the sine wave via oscilloscope, and it is just fine.

A laptop is having literally the same issue on-grid and off-grid, using the portable power station or not using it.

[Inputdelay]

I would recommend running some traces with the Windows Performance Recorder / Analyzer.

If there are recoverable errors somewhere in your system these should be logged somewhere.
So I would suggest looking into the event logs and turn on as much as possible.

Things that possible are not logged are issues in the CPU power state switching, but an EMI susceptibility on multiple platforms would be rare.

Anyway, Windows is terrible at handling IO errors. For example if an nvme drive craps out it will just BSOD, while with Linux you get entries in the kernel log and only that specific device will just stall. Contrary to windows where the entire io layer will stall.
So it may very well be as simple as a stalling SSD or misbehaving USB device.

If you expect it to be noise from the grid you can measure power quality, these meters are expensive though, maybe rent one.
However, I know from someone in Kyiv that power quality is abysmal right now. Not the most important thing when your infrastructure is getting shot to bits.

From the other angle, could you have an EMI susceptibility?

I did run the Performance Recorder/Analyzer millions of times, and as I said earlier, it has nothing to do with the OS in the first place.

CPU power state is constant and super stable.

I have already switched entire systems, and the problem still remains. It isn't the SSD or a misbehaving USB device either.

I do really want to measure power quality, but I didn't find a single way of doing that, even to rent, since I have a war going on here, and most of the labs aren't working.

I had been thinking about EMI problems, and it might be something outside of my private house, but I don't know where to look in the first place.

[Inputdelay]

The thing is, such effects due to external interference are almost impossible, especially if consistent and not just a glitch in one frame out of a million. Any memory corruption events would manifest most notably as system crashes and are pretty random. Glitching digital video transmission (like hdmi link) would also manifest quite differently, by select variety of video defects or a total loss of video. I suggest to not go into "moon phases affect my frame rates" type of theories, it will be just a waste of time and a distraction from real issues.

it might be just an EMC, not the "moon phases affect my frame rates" type of theories.

I don't mean it to sound insulting, I mean it as a guidance. Source: I design, test and troubleshoot electronic circuits for over 20 years. You probably mean EMI, not EMC. Anyway, that is exactly my point. EMI causing such effects consistently, repeatably and across many devices is "moon phases" theory.

Gaming is demanding, I understand that. Every millisecond counts. But that's far from the only mission critical application for a PC. There are industrial, medical, aeronautical, defense applications where PCs are running mission critical stuff, sometimes in really harsh environments. And PCs are not some black magic put together, these systems are designed by humans and well understood, if it would not be understood, we would never make it work in the first place.

Sure there are edge cases of defective products, like a specific model of motherboard with unstable PLL or something. So you touch the case with your hand and it starts working. Or stops working. Or whatever. But these are again edge cases of serious design errors. But that does not sound at all like your case.

And we humans often have really hard time analyzing something and drawing conclusions. In fact, we are often really really bad at that. That's how historically we had so many superstitions (and still do). One time someone came up with an idea to sacrifice a young virgin girl and it happened that crops grew very well. So after that they started sacrificing a young girl every year.

Perhaps off topic, but this link is a really nice resource for an evening reading:

https://en.wikipedia.org/wiki/List_of_cognitive_biases

Anyway, in conclusion, as I said, too consistent to be EMI issue. In specific, isolated edge cases we could talk about that, but not here. And it's a digital system after all. It's not like we have sound recording studio and "my noise floor is -115dB with LED lights ON, -120dB if lights are OFF and -121dB if the LED lights are OFF and the microphone is facing north". Such effects are common in high-end analog systems, but not digital.

I understand your point, and I want to clarify something important.

I am not trying to claim that this is a classic EMI problem in the sense of digital logic malfunctioning or systems crashing. I fully agree that modern digital systems are well understood, robust, and widely used in far more critical applications than gaming.

What I am struggling with is not explaining the root cause yet, but identifying what measurable parameter is actually changing over time.

I also agree that human perception is unreliable, which is exactly why I am trying to move away from subjective descriptions and towards instrumentation and logging. If this were purely perceptual, I would expect it to disappear under objective measurement - but so far I simply don’t know what to measure correctly. And yet it has nothing to do with my perception; then everyone, including professors and engineers and my family, is blind, which is not the case. It has nothing to do with cognitive biases.

At this point, I am not committed to EMI, power quality, grounding, or any specific hypothesis.

My working assumption is only this:

Something in the system’s behavior changes over time, and I have not yet identified the parameter that captures it.

That is why I am here - not to defend a theory, but to learn which conditions would be the correct place to look and how to falsify wrong assumptions instead of reinforcing them.

[Gyro]

I thought this thread sounded familiar... Maybe it's the humidity!

https://www.eevblog.com/forum/beginners/pc-performance-correlated-with-relative-humidity/

[Inputdelay]

It has nothing to do with humidity.

[Uunoctium]

An issue that reminds me of yours occured here suddenly some years ago. I'm using an Microsoft BT mouse without an additional BT receiver, mainly to keep the USB ports accessible. Computer is an Envy NB with UHD Display. Over two years with a cadence of approx ten times or more a year mouse freezes at diplay and with a lag of 3 seconds the pointer jumps in centimeter steps. Unusable for hours.
My concern was, that anyone in the building uses an illegal unapproved BT device such a wireless headphone with too much RF-power or bandwith cosumption.

The only effective countermeasure against was, to jamm with a generator, set to sweep from 2.400 to 2.480MHz, AM mod and +13dBm feeding an log-per antenna directed to the floor below.
In near all cases the trouble ended after 10 minutes.

Knowing, that would be not a long term solution, i replaced the combo Wifi/BT card with the newer Intel AX201.
Issue never occured again since.

[DimitriP]

Small recap so we don't lose track:)
Lightbulbs: went from "normal" brightness", to dim, a tranformer was replaced and brought the brightness back, and now they are dimmer again.
PC or  laptop on battery or plugged in exhibits the symptom.
For all we know there is  a mad scientist  two houses down the road  polluting the neighborhood with "something" that's causing that.
I did use to have a tiny notebooks PC connected to my HF radio to decode and send morse. Decoding had no issues. Sending was another matter, with high-is SWR, and a 100W tranceiver 2 feet away from it. It would act sluggish, keys would appar "stuck" soemtimes it would lock up for the duration of the transmission, until the tranceiver was manually switched to receive, since the notepbook was too slow or in that "pause mode" unable to switch the tranceiver off.
Everything was in front of me so it "made sense". At least on of the 4 monitors on the other side of the room would do some watercolor effects, like they use to do dream sequences" in movies.
But that was a tiny notepook laptop next to a 100W HF tranceiver being flooded by RF.

I can't imagine soemthing external causing this amount of interference for this long a time (over 2 years now as the OP mentioned).

At the same time, don't be limited by my imagination.

[Inputdelay]

An issue that reminds me of yours occured here suddenly some years ago. I'm using an Microsoft BT mouse without an additional BT receiver, mainly to keep the USB ports accessible. Computer is an Envy NB with UHD Display. Over two years with a cadence of approx ten times or more a year mouse freezes at diplay and with a lag of 3 seconds the pointer jumps in centimeter steps. Unusable for hours.
My concern was, that anyone in the building uses an illegal unapproved BT device such a wireless headphone with too much RF-power or bandwith cosumption.

The only effective countermeasure against was, to jamm with a generator, set to sweep from 2.400 to 2.480MHz, AM mod and +13dBm feeding an log-per antenna directed to the floor below.
In near all cases the trouble ended after 10 minutes.

Knowing, that would be not a long term solution, i replaced the combo Wifi/BT card with the newer Intel AX201.
Issue never occured again since.

Small recap so we don't lose track:)
Lightbulbs: went from "normal" brightness", to dim, a tranformer was replaced and brought the brightness back, and now they are dimmer again.
PC or  laptop on battery or plugged in exhibits the symptom.
For all we know there is  a mad scientist  two houses down the road  polluting the neighborhood with "something" that's causing that.
I did use to have a tiny notebooks PC connected to my HF radio to decode and send morse. Decoding had no issues. Sending was another matter, with high-is SWR, and a 100W tranceiver 2 feet away from it. It would act sluggish, keys would appar "stuck" soemtimes it would lock up for the duration of the transmission, until the tranceiver was manually switched to receive, since the notepbook was too slow or in that "pause mode" unable to switch the tranceiver off.
Everything was in front of me so it "made sense". At least on of the 4 monitors on the other side of the room would do some watercolor effects, like they use to do dream sequences" in movies.
But that was a tiny notepook laptop next to a 100W HF tranceiver being flooded by RF.

I can't imagine soemthing external causing this amount of interference for this long a time (over 2 years now as the OP mentioned).

At the same time, don't be limited by my imagination.

Thanks for both replies; they are useful, so let me clarify a few important points and add missing context.

First, regarding the Bluetooth/RF interference example:

I fully understand that localized RF sources (especially in 2.4 GHz) can cause very specific and reproducible failures, like mouse freezes. In your case, the behavior was clearly tied to BT/WiFi and was resolved by replacing the radio hardware, which makes perfect sense.

In my case, however:

- The issue stays the same for all kinds of wired and wireless USB mice and keyboards.

- present with laptops running purely on battery,

- present with different PCs, monitors, and even TVs, including smartphones (when charging them).

– and persists when everything inside the house is disconnected at the breakers.

Because of that, I don’t currently see a single RF band or interface that could explain all observed effects in a consistent way.

Second, about long-term external interference:

I agree that a constant strong RF source affecting a whole neighborhood for years would be highly unlikely. That’s also why I’m not assuming a “mad scientist” scenario or some unknown transmitter nearby.

What makes my case confusing is that the issue has been present since the house was built. The distribution transformer for us was the same from ~2006 until its replacement in 2025. The only events that ever caused noticeable changes were actions on the power grid side (transformer replacement, switching events, emergency modes), and these changes were temporary. Except for the brightness of all of the devices.

I sometimes get spontaneous improvements lasting from 10 minutes up to a couple of hours:

The image becomes visibly cleaner (as if “wiped”), motion clarity improves, and mouse response becomes sharper. These improvements never last and disappear without any action on my side.

About grounding and day/night behavior:

Earlier there was a noticeable difference between day and night behavior.

After I implemented a local TT grounding for a single outlet (PC + monitor) in April 2024, the day/night difference disappeared completely.

The issue itself did not disappear, but its time-of-day correlation changed.

To be clear: I’m not claiming a specific cause at this point.

I’m trying to understand what class of parameters could change slowly or sporadically over time and affect timing stability across many digital devices, even when running on battery.

If you think this is not EMI, I’m absolutely open to that.

What I’m looking for is guidance on:

- what would be the next sensible thing to measure,

- what kind of disturbance could realistically affect multiple systems without causing outright crashes,

- and how to falsify wrong hypotheses instead of reinforcing assumptions.

[Buriedcode]

Given the bizarre an inconsistent effects, I'm surprised no-one has mentioned carbon monoxide. We've all seen that legendary reddit thread about it - I don't mean to be insulting here, could be worth checking out.

[Manul]

The question is, how willing are you to put effort and set up lab tests?

Two ideas:

1. Confirm and possibly log your mouse update rate. What mouse rate do you have according to pages like this?

https://cpstest.org/polling-rate-test/

Setup something which rotates continuously in front of optical mouse sensor to simulate a constant movement (alternative: program a microcontroller to act as HID mouse simulating continuous movement). Write some piece of software, which logs and analyzes: position update rate (is it stable) and position delta per update (are there no abnormal position jumps).

2. Make an optical sensor from a photodiode with amplifier. You can plug it into sound card, cause frequencies of interest are within audio range. Make software which renders a quickly and steadily changing scene like a sliding smooth gradient (use Unity or something). Do spectrum analysis on the captured signal. Factual display update rate should be visible as a clear spectrum line (let's say 360Hz). You can write software to compute FFT in real time and log the factual display update rate. You may be able to expand on this idea and analyze ghosting too.

Otherwise I don't know what else to suggest.

[tszaboo]

I agree that a constant strong RF source affecting a whole neighborhood for years would be highly unlikely. That’s also why I’m not assuming a “mad scientist” scenario or some unknown transmitter nearby.

Why? There could be EW equipment around from either side, if your flag is set up correctly.

[Inputdelay]

The question is, how willing are you to put effort and set up lab tests?

Two ideas:

1. Confirm and possibly log your mouse update rate. What mouse rate do you have according to pages like this?

https://cpstest.org/polling-rate-test/

Setup something which rotates continuously in front of optical mouse sensor to simulate a constant movement (alternative: program a microcontroller to act as HID mouse simulating continuous movement). Write some piece of software, which logs and analyzes: position update rate (is it stable) and position delta per update (are there no abnormal position jumps).

2. Make an optical sensor from a photodiode with amplifier. You can plug it into sound card, cause frequencies of interest are within audio range. Make software which renders a quickly and steadily changing scene like a sliding smooth gradient (use Unity or something). Do spectrum analysis on the captured signal. Factual display update rate should be visible as a clear spectrum line (let's say 360Hz). You can write software to compute FFT in real time and log the factual display update rate. You may be able to expand on this idea and analyze ghosting too.

Otherwise I don't know what else to suggest.

I understand what you are suggesting, and I agree with the general principle: if something is real, it should be measurable.

However, the reason I’m not pursuing mouse polling tests or DIY photodiode display measurements is not because I’m avoiding measurements, but because these tests only validate subsystems that have already been ruled out multiple times.

Let me be very clear about the scope:

 The issue is present with multiple mice (wired), keyboards, PCs, laptops, monitors, and TVs.
It is present when running on battery, off-grid power stations, and generators.
It persists when the entire house is disconnected at the breakers.
It does not correlate with USB polling rates, drivers, OS versions, BIOS, or hardware changes.

It does not manifest as packet loss, freezes, dropped frames, or obvious malfunction - but as timing degradation and loss of motion clarity.

So testing mouse polling stability or display refresh stability in isolation will almost certainly produce “correct” results, because the devices themselves are functioning within spec.

What I’m trying to understand is not: “Does my mouse report 4000 Hz?”or “Does my panel refresh at 360 Hz?”

But rather:

“What external or infrastructural parameters can slowly or intermittently shift timing stability, noise margins, or reference conditions across many digital systems simultaneously, without causing outright failures?”

This is why I’m looking outside the PC - power quality, grounding topology, neutral stability, common-mode noise, reference potential drift, etc.

If you believe those classes of issues are impossible or irrelevant, that’s fine - but then the productive discussion would be:
why they can be excluded, and what other non-PC-internal domain should be investigated instead.
I’m not looking to prove that a mouse or a display is broken.
I’m trying to identify which physical domain I should be measuring next.

[Inputdelay]

I agree that a constant strong RF source affecting a whole neighborhood for years would be highly unlikely. That’s also why I’m not assuming a “mad scientist” scenario or some unknown transmitter nearby.

Why? There could be EW equipment around from either side, if your flag is set up correctly.

Then a radio-monitoring inspection at my place should have caught that, right?

[tszaboo]

I agree that a constant strong RF source affecting a whole neighborhood for years would be highly unlikely. That’s also why I’m not assuming a “mad scientist” scenario or some unknown transmitter nearby.

Why? There could be EW equipment around from either side, if your flag is set up correctly.

Then a radio-monitoring inspection at my place should have caught that, right?

If it was on at the time. And the inspectors weren't told to just ignore those signals.
I mean, there are a few things going on*, not telling ordinary citizens about nearby EW stuff is going to be standard.
* I would write some other things, but we were asked not to discuss it on this forum.

[Inputdelay]

I agree that a constant strong RF source affecting a whole neighborhood for years would be highly unlikely. That’s also why I’m not assuming a “mad scientist” scenario or some unknown transmitter nearby.

Why? There could be EW equipment around from either side, if your flag is set up correctly.

Then a radio-monitoring inspection at my place should have caught that, right?

If it was on at the time. And the inspectors weren't told to just ignore those signals.
I mean, there are a few things going on*, not telling ordinary citizens about nearby EW stuff is going to be standard.
* I would write some other things, but we were asked not to discuss it on this forum.

Then how do I even check this by myself? I don't have any equipment for that.

[tszaboo]

I agree that a constant strong RF source affecting a whole neighborhood for years would be highly unlikely. That’s also why I’m not assuming a “mad scientist” scenario or some unknown transmitter nearby.

Why? There could be EW equipment around from either side, if your flag is set up correctly.

Then a radio-monitoring inspection at my place should have caught that, right?

If it was on at the time. And the inspectors weren't told to just ignore those signals.
I mean, there are a few things going on*, not telling ordinary citizens about nearby EW stuff is going to be standard.
* I would write some other things, but we were asked not to discuss it on this forum.

Then how do I even check this by myself? I don't have any equipment for that.

Probably the cheapest is an RTL-SDR, about 20 EUR.
Or a TinySA, it's a bit more expensive.
I mean at this point you are just guessing that maybe the second moon of jupiter is causing it.

IMHO something is going to be inside your PC, and it's unlikely that it's some outside source. But you are not willing to investigate that. And unlikely is not impossible. I actually don't know how much trouble EW can cause.

[Inputdelay]

I agree that a constant strong RF source affecting a whole neighborhood for years would be highly unlikely. That’s also why I’m not assuming a “mad scientist” scenario or some unknown transmitter nearby.

Why? There could be EW equipment around from either side, if your flag is set up correctly.

Then a radio-monitoring inspection at my place should have caught that, right?

If it was on at the time. And the inspectors weren't told to just ignore those signals.
I mean, there are a few things going on*, not telling ordinary citizens about nearby EW stuff is going to be standard.
* I would write some other things, but we were asked not to discuss it on this forum.

Then how do I even check this by myself? I don't have any equipment for that.

Probably the cheapest is an RTL-SDR, about 20 EUR.
Or a TinySA, it's a bit more expensive.
I mean at this point you are just guessing that maybe the second moon of jupiter is causing it.

IMHO something is going to be inside your PC, and it's unlikely that it's some outside source. But you are not willing to investigate that. And unlikely is not impossible. I actually don't know how much trouble EW can cause.

Funny enough, my friend has the exact same problem, and he has RTL-SDR. He might do some tests. But what exactly are we looking for, and how then?

[Uunoctium]

What is the area, you living?
Residential or industrial/mixed? Is an railway nearby and if so, is it electric? Medical Offices esp radiology (unproper shielded MRT or x-ray)? Industry operating RF heating? Workshops with magnetizing processes? Huge PV installation? On rooftop mobile phone BTS?

[Manul]

However, the reason I’m not pursuing mouse polling tests or DIY photodiode display measurements is not because I’m avoiding measurements, but because these tests only validate subsystems that have already been ruled out multiple times.

I'm not sure if you understood what I'm saying. I don't suggest to test mouse and display. It would not be isolated tests of these subsystems. There is much more going on under the hood. For instance, if you log mouse poll rate and deltas, what is included (but not limited to) is:

mouse itself (sensor, firmware), USB bus (physical differential pair which might be catching EMI), USB hub, it's firmware, windows USB layer drivers, HID drivers, some windows user mode code, and finally some application receiving mouse events. So if mouse is not smooth itself, if USB frame gets corrupted, if some interrupt routine takes too long, perhaps some unrelated device stutters and temporarily locks kernel execution, power states transition, page faults happen or something else jitters in the system - you will see it as inconsistent rate / delta.

Same with video output. Furthermore, these days it's really quite complicated, there is not only v-sync, but also technologies like g-sync (backwards sync from display to GPU), variable refresh rates, in between frame generation (AI driven interpolation) and such.

The point is - if you have stable, rock solid input chain from physical mouse to end application and if you have rock solid chain from application to display you have rock solid everything, cause these two chains encapsulate pretty much everything what's at play when you run a game on PC. So my guess is that if the problem is real, it will show up in these measures as unstable jittery numbers.

[Inputdelay]

What is the area, you living?
Residential or industrial/mixed? Is an railway nearby and if so, is it electric? Medical Offices esp radiology (unproper shielded MRT or x-ray)? Industry operating RF heating? Workshops with magnetizing processes? Huge PV installation? On rooftop mobile phone BTS?

I live in a small rural/suburban village of around 2,000 people, not an industrial zone. Which is 500 m away from the big 1 million city. It's a private house for the most part, and

Nearby infrastructure includes:

General regional railway lines in the region (electrified lines of the railway), but no large rail traction power substations or heavy industry directly adjacent. Around 1 km and 600 m.

The city with its industrial base is several kilometers away (not immediately adjacent to my residence).

The local airport serving is at some distance (~5 km), not within my immediate vicinity.

There are no known:

medical radiology or MRT facilities nearby, RF heating industrial facilities, large telecommunications transmitters, or BTS on rooftops directly near the house.

My immediate residential area does not contain high-power industrial equipment or specialized RF sources. I live in a typical mixed residential/rural environment.

[Inputdelay]

However, the reason I’m not pursuing mouse polling tests or DIY photodiode display measurements is not because I’m avoiding measurements, but because these tests only validate subsystems that have already been ruled out multiple times.

I'm not sure if you understood what I'm saying. I don't suggest to test mouse and display. It would not be isolated tests of these subsystems. There is much more going on under the hood. For instance, if you log mouse poll rate and deltas, what is included (but not limited to) is:

mouse itself (sensor, firmware), USB bus (physical differential pair which might be catching EMI), USB hub, it's firmware, windows USB layer drivers, HID drivers, some windows user mode code, and finally some application receiving mouse events. So if mouse is not smooth itself, if USB frame gets corrupted, if some interrupt routine takes too long, perhaps some unrelated device stutters and temporarily locks kernel execution, power states transition, page faults happen or something else jitters in the system - you will see it as inconsistent rate / delta.

Same with video output. Furthermore, these days it's really quite complicated, there is not only v-sync, but also technologies like g-sync (backwards sync from display to GPU), variable refresh rates, in between frame generation (AI driven interpolation) and such.

The point is - if you have stable, rock solid input chain from physical mouse to end application and if you have rock solid chain from application to display you have rock solid everything, cause these two chains encapsulate pretty much everything what's at play when you run a game on PC. So my guess is that if the problem is real, it will show up in these measures as unstable jittery numbers.

Yes, I understand what you mean, and I agree with the logic behind it.

I understand that logging mouse polling rate/deltas or optical measurements of the display is not about testing the mouse or the monitor in isolation but about capturing the entire end-to-end chain.

The reason I am hesitant to go down this path is not because I reject measurements, but because I already spent a long time doing functional end-to-end testing of the whole system under different conditions, and the results point away from an internal jitter problem.

Here is the key observation that makes me skeptical about internal chain jitter as the root cause:

"The same PC, OS image, peripherals, BIOS, firmware, and configuration behave normally when moved to another physical location. Multiple different PCs, laptops, monitors, and even TVs exhibit the same degradation when used in this location. Battery-powered devices (laptops) are affected as well. Software changes, background processes, drivers, USB controllers, power states, etc. do not correlate with state changes. Improvements sometimes happen abruptly for minutes or hours without any software or hardware change."

This strongly suggests that what changes is not the internal timing of the system but the operating conditions under which the system runs.

My current working hypothesis is not “random jitter inside the PC,” but a shift in reference conditions: chassis, ground/neutral potential drift, common-mode noise, and slow changes in supply or environmental electrical conditions that alter how digital circuits operate without causing outright errors or crashes.

In other words, I am trying to find parameters that explain why the same digital system behaves differently without any internal state change.

If you believe that such effects would still necessarily manifest as measurable jitter in mouse or display timing, I am open to hearing which specific metric would uniquely distinguish between internal scheduling/driver/software issues and external reference or power-related effects.

I am not trying to avoid measurement-I am trying to avoid measurements that cannot falsify the main hypothesis.