Unstable Pc performance and lack of visual smoothness.

[negativ3]

Have you had your main fusebox and interconnects insulation tested to various outlets?

Have you isolated the power, disconnected the copper at both ends, shorted one end together, then measured resistance between pairs at the unshorted end?

Probably better if you consolidated all your tests/checks and actual data into an easily readable document.

Its either a physical phenomenon or RF related for visitors to experience the same, and if as you say you have the same issues running off fully isolated batteries... that leaves RF. The fault clears when you travel.

[Inputdelay]

Have you had your main fusebox and interconnects insulation tested to various outlets?

Have you isolated the power, disconnected the copper at both ends, shorted one end together, then measured resistance between pairs at the unshorted end?

Probably better if you consolidated all your tests/checks and actual data into an easily readable document.

Its either a physical phenomenon or RF related for visitors to experience the same, and if as you say you have the same issues running off fully isolated batteries... that leaves RF. The fault clears when you travel.

No, I have not yet performed insulation resistance testing of the main fusebox and branch wiring, nor have I disconnected conductors end-to-end and tested pair resistance in the way you described.

That is a valid gap in my investigation, and I agree it’s one of the remaining areas that can still explain unstable neutral behavior, DC components, coupling that bypasses local isolation, and batteries.

The only thing that I tried is I powered my PC and monitor not using the wiring of my house, de-energizing it and powering it directly after the meter. But the problem itself still remained. Which is the main concern about the branch of wiring in the house.

I do want to clarify one point:

Battery operation and isolation did not eliminate the effect, which suggests coupling not strictly via conducted AC power - potentially via wiring acting as antennas, reference drift, or environmental RF coupling.

I agree that consolidating all tests and measurements into a single structured document is the right next step.

If you have specific recommendations on insulation test voltage levels, acceptable resistance thresholds, or how to safely perform these tests in a residential setting, that would be genuinely helpful.

[Inputdelay]

Have you had your main fusebox and interconnects insulation tested to various outlets?

Have you isolated the power, disconnected the copper at both ends, shorted one end together, then measured resistance between pairs at the unshorted end?

Probably better if you consolidated all your tests/checks and actual data into an easily readable document.

Its either a physical phenomenon or RF related for visitors to experience the same, and if as you say you have the same issues running off fully isolated batteries... that leaves RF. The fault clears when you travel.

The list might not be full, since there are many other things that I have tried, but there is not enough time to name all of them, since it's a lot.

1. Hardware & peripherals

Multiple PCs and laptops tested (different platforms, vendors, and generations)
Multiple monitors tested (LCD, high refresh rate, different panels)
Multiple GPUs, PSUs, RAM kits
Multiple mice, keyboards, mousepads
HDMI / DP / DVI cables replaced
Issue persists across all combinations

2. Software / OS

Multiple Windows versions and builds
Linux (Ubuntu)
Clean installs
BIOS/UEFI updates
GPU VBIOS updates
Chipset / GPU / USB drivers swapped
All OC/XMP disabled
MSI mode tested
VRR / G-Sync on/off
Frame caps on/off
Extensive stability tests (Prime95, OCCT, MemTest, Cinebench, and many, many others)
Internal telemetry (FPS, frametimes, clocks, temps, voltages) unchanged
The problem was occurring on the desktop when there are additionaly moving objects they behave abnormal.
I checked, as Dervu said, the timer measurements inside CPU-Z, and I can confirm that they were not acting; they were glitching in a way that it would do like microstops every x seconds that you could visually notice.

3. Network / ISP

Multiple ISPs (FTTH, coax, LTE)
Online and offline behavior is identical.
Router connected/disconnected
Ethernet connected/disconnected

4. Power sources

Online double-conversion UPS (online mode)
Online UPS disconnected from mains
Battery power stations (pure sine)
Isolation transformer
Autotransformer (Variac)
EMI/RFI filters: Industrial EMI filters (including EMSBI FZP-125-1, 30 kHz - 18 GHz range)
Ferrites on power and signal cables
Different sockets and circuits
Result: Symptom persists even fully isolated from the grid.

5. Electrical observations (measured)
Neutral-to-ground voltage fluctuates from ~0.1 V up to 5–10 V under load.
Ground-neutral resistance varies from ~4 Ω up to >450 Ω.
Neutral behaves worse than local ground
DC components detected on phase and neutral during outages
AC voltage near zero during outages, DC current still present
Changes observed after distribution transformer replacement
Day/night behavior disappeared after adding local ground reference (TT, single outlet).

6. RF/EMI

All household devices were disconnected, and there was no change.
Router and networking gear removed, and there was no change
The issue persists on battery.
RF monitoring performed (no obvious narrowband source detected)
Behavior similar to other independent users in different countries

7. Reproducibility
The same hardware behaves perfectly in other locations.
Returns immediately when brought back
Everyone who visited my place observed the same problem, which eliminates internal factors completely.
Problem present since building construction (~2007). Sometimes the problem was even worse; it all depended on the electricity company and their changes around the transformer and its neutral.
Improvements occur sporadically for minutes to hours, then disappear.

8. Correlated reports

Multiple users report identical symptoms:
delayed input without FPS drops
“heavy” movement feeling
worse at higher refresh rates
Mouse feeling heavy, sluggish, floating on ice, not precise, not snappy
Keyboard feeling nonresponsive, with visible delay even when typing in Windows
disappears when relocating with the same hardware
Reduction in overall image clarity.
Ghosting: the remaining trail of moving objects.
Image splitting during sudden mouse movements.

Open technical questions

Which external electrical parameters (neutral drift, common-mode noise, reference instability) can modulate timing/jitter without visible errors?

What mechanisms could affect display synchronization without altering internal mechanisms?

If someone believes this cannot happen:
Please explain why the same system behaves differently under different electrical reference conditions while reporting identical internal states.

That is the unresolved point.

[negativ3]

in different countries

So its a particular piece of gaming software which is common to all users having the experience.

That explains a lot.

I'm out.

[Inputdelay]

in different countries

So its a particular piece of gaming software which is common to all users having the experience.

That explains a lot.

I'm out.

Your conclusion does not follow.

The common factor is not gaming software.

The common factor is location-dependent degradation that:

affects offline applications (desktop, typing, local tests),

affects different games and non-game software, just a simple desktop expierence.

That directly contradicts a software-only explanation.

If this were a specific game engine or OS issue, relocation would not instantly resolve it without any configuration change.

You are free to disengage, but this is not a valid technical inference.

[negativ3]

Your certainty about many thing implies there is a known issue which you're not sharing.

[Inputdelay]

Your certainty about many thing implies there is a known issue which you're not sharing.

No. Certainty does not imply hidden knowledge.

It implies elimination.
I am certain about what this is not, because those variables were systematically changed and isolated:
different hardware
different OS and drivers
different software and games
offline operation
battery operation
different ISPs
different locations
What remains is not “a known issue”, but an unresolved one.
If I had a known root cause, I would not be here asking which external parameters should be measured and with what instruments.
If you think my certainty is unjustified, point to a specific falsified test that I ignored.
Otherwise, this line of questioning goes nowhere.

[NE666]

External electrical parameters (neutral-ground voltage, resistance, DC offsets, transformer changes) correlate with behavior.

How can that be, when you've told us that the same effect can also be seen when you're running on battery power bank, during periods of mains power loss?

Besides, those AC line parameters aren't important in and of themselves. It's only what's happening on the DC side of the SMPS that matters, so look for quality or interference issues there.

[al777]

I understand your point, and I agree with you on one fundamental thing:

If the system produces correct, stable frames and timing, then the result must be correct and measurable.

If the system works incorrectly, the result is incorrect and also measurable. Most importantly, in what exact way it is incorrect often hints to where the problem is. For instance, it might hint what is more likely, PC, display link or display itself misbehaving, thus narrowing further troubleshooting. Anyway, I'm clearly not needed here, so I will refrain from further discussion.

The only noticeable difference is the fluctuation of the core speed of the CPU and the constant change of bus speed; that's the only metric that stands out.

      ^^^-- this makes me think of a (crazy) hypothesis... An external ~8...12GHz transmission could cause localized heating (only few degrees or even less) of silicon dies, which in turn could trigger thermal throttling, exhibiting itself in core/bus frequencies changing. The thermal throttling control could be completely buried in the firmware/microcode in the CPU/GPU/etc. - no software drivers may be necessary/able to register it, nor anything would be exposed other than the frequency changes. And for a gaming system it would be plausible that performance is already maxed out/dancing on the edge of thermal throttling, so a minor push could trigger it. As for the source of the multi-GHz emission... your flag tells us that there could be plenty of that, both sides are using X-band and the power output on those systems is not trivial, as long as there is a direct line of sight, especially over a large body of water/flat land - it will be tangible on the receiving end. Radar trauma is known to first exhibit itself to a human victim as a "wave of warmth" (before other nasty things happen), silicon on the edge of its thermal envelope is much more sensitive to such heating. And as 12GHz has 25mm wavelength - your prior attempt to shield the system may have been ineffective for that frequency, a minor (few mm scale) crack in the cage could still allow it to leak in, and ingress cables can act as (poor) antennas/waveguides.

Going further on the crazy path - the periodic nature of the issue (periodic "microfreezes") could correlate to rotation speed of the transmitting antenna. In a system with active cooling thermal inertia is fairly low, those few extra degrees may be shed fast enough to fit between rotation cycles. BTW, the act of frequency switching as a result of thermal throttling in itself could be taxing the system (e.g. due to re-buffering or other frequency switching-related extra activity). If only switched once in a long time (like every few minutes) it may not be noticeable, but repeated frequently - it may be bogging down the system. On a wild whim I'd try a 1N5711 Schottky anti-parallel to a green (emerald) LED, with uncut 1N5711's leads serving as a "dipole antenna". Your WiFi router will likely not light it up, but those other sources can, if they are close enough.

I did note that you've said that it started in 2007, which is well before the current situation unfolded in your region - I do not have an explanation for that within my crazy hypothesis. Good luck in your search of the culprit.

[Inputdelay]

Your certainty about many thing implies there is a known issue which you're not sharing.

https://youtu.be/LDQQy5oPnRY
The result is at FRIEND'S APARTMENT!  THIS IS NOT MY HOUSE!

This video demonstrates a reproducible difference in cursor motion behavior under different electrical conditions.

PART 1- Powered from battery:
The laptop is disconnected from the mains and running only on its internal battery.
Visible effect:
- Cursor motion becomes noticeably more stable.
- Reduced micro-stutter compared to mains power

PART 2 - Powered from mains (before electrical work):
Laptop connected to the main grid.
Visible effect:
- Micro-stutter / “floating” effect is clearly visible
- Cursor does not move linearly despite constant hand motion
- Cursor motion is uneven.

PART 3 - After electrical work:
A licensed electrician performed the following actions in the building:
- Cleaned and restored neutral bus connections
- Removed twisted/spliced neutral connections
- Improved contact quality
- Replaced circuit breakers for the entire stairwell

NO changes were made to:
- Laptop hardware
- Software / OS / drivers
- Mouse
- Camera
- Environment layout

Result (same mains power, after neutral remediation):
- Cursor motion becomes smooth and linear.
- The previously visible floating/stuttering effect disappears.
This video is not intended to “prove a theory.” It demonstrates that correcting neutral integrity alone can change observable input behavior on a digital system without any internal changes.
The question is not whether this happens. The question is which electrical parameters explain it and how to measure them properly.

Important note:
This is not about mouse polling rate, drivers, games, or perception.
The same effect was reproducible:
- on desktop
- offline
- before and after neutral remediation
If this were a purely internal timing or software issue,
changing building neutral integrity would not alter the result.
He also took a couple of photos before and after.

[xvr]

The questioner mentioned an airport located 5 kilometers from his home. For an airport radar, 5 kilometers isn't that far. Of course, there's no direct beam to the questioner's apartment (the result of a direct beam would be obvious and catastrophic), but even a side or reflected beam can be significant.

[Manul]

Wow, finally, something which shows something of substance.

This is a widely known and understood issue with touchpads and it is caused by common mode noise. Touchpad has capacitive sensing matrix which is scanned at high frequency detecting miniscule capacitance changes when a finger is moved. Quite sensitive and finicky devices. Isolated switching adapters produce lots of common mode noise and when you put finger on the touchpad, it becomes a capacitive path for the common mode, overloading and disturbing the capacitve sense matrix. Usually happens with crappy touchpads or if power adapter is not original. Original manufacturer adapters are designed with specific switching frequencies in mind which don't interfere with coresponding touchpads on their laptops and in general has higher quality and filtering. If it happens with original adapter - it is just crap.

There is also noise which comes from AC line, not just from power adapter, although it is rare to be so strong. Proper ground in the building can reduce that, it is true.

But such common mode noise should not affect USB mouse, if it is not at cosmic high levels. Touchpad and external mouse might both seem like a mouse, but under the hood they are very different devices. Regarding the external noise, touchpad is very sensitive, USB mouse is very robust.

Make a video with USB mouse. I want to see it.

[Inputdelay]

External electrical parameters (neutral-ground voltage, resistance, DC offsets, transformer changes) correlate with behavior.

How can that be, when you've told us that the same effect can also be seen when you're running on battery power bank, during periods of mains power loss?

Besides, those AC line parameters aren't important in and of themselves. It's only what's happening on the DC side of the SMPS that matters, so look for quality or interference issues there.

Let's say you're right. What devices do I need to measure this, what methods are needed for this, what indicators should I look at, what should I look at, what is normal, and what is not?

[Inputdelay]

I understand your point, and I agree with you on one fundamental thing:

If the system produces correct, stable frames and timing, then the result must be correct and measurable.

If the system works incorrectly, the result is incorrect and also measurable. Most importantly, in what exact way it is incorrect often hints to where the problem is. For instance, it might hint what is more likely, PC, display link or display itself misbehaving, thus narrowing further troubleshooting. Anyway, I'm clearly not needed here, so I will refrain from further discussion.

The only noticeable difference is the fluctuation of the core speed of the CPU and the constant change of bus speed; that's the only metric that stands out.

      ^^^-- this makes me think of a (crazy) hypothesis... An external ~8...12GHz transmission could cause localized heating (only few degrees or even less) of silicon dies, which in turn could trigger thermal throttling, exhibiting itself in core/bus frequencies changing. The thermal throttling control could be completely buried in the firmware/microcode in the CPU/GPU/etc. - no software drivers may be necessary/able to register it, nor anything would be exposed other than the frequency changes. And for a gaming system it would be plausible that performance is already maxed out/dancing on the edge of thermal throttling, so a minor push could trigger it. As for the source of the multi-GHz emission... your flag tells us that there could be plenty of that, both sides are using X-band and the power output on those systems is not trivial, as long as there is a direct line of sight, especially over a large body of water/flat land - it will be tangible on the receiving end. Radar trauma is known to first exhibit itself to a human victim as a "wave of warmth" (before other nasty things happen), silicon on the edge of its thermal envelope is much more sensitive to such heating. And as 12GHz has 25mm wavelength - your prior attempt to shield the system may have been ineffective for that frequency, a minor (few mm scale) crack in the cage could still allow it to leak in, and ingress cables can act as (poor) antennas/waveguides.

Going further on the crazy path - the periodic nature of the issue (periodic "microfreezes") could correlate to rotation speed of the transmitting antenna. In a system with active cooling thermal inertia is fairly low, those few extra degrees may be shed fast enough to fit between rotation cycles. BTW, the act of frequency switching as a result of thermal throttling in itself could be taxing the system (e.g. due to re-buffering or other frequency switching-related extra activity). If only switched once in a long time (like every few minutes) it may not be noticeable, but repeated frequently - it may be bogging down the system. On a wild whim I'd try a 1N5711 Schottky anti-parallel to a green (emerald) LED, with uncut 1N5711's leads serving as a "dipole antenna". Your WiFi router will likely not light it up, but those other sources can, if they are close enough.

I did note that you've said that it started in 2007, which is well before the current situation unfolded in your region - I do not have an explanation for that within my crazy hypothesis. Good luck in your search of the culprit.

I appreciate the effort and creativity in the hypothesis, but I want to be very clear here; otherwise, this discussion drifts in the wrong direction.

I do NOT believe this is caused by:
- external radar heating
- GHz-range directed energy
- rotating antennas
- military X-band exposure

Reasons:

1) Timeline mismatch
The issue has existed continuously since ~2007.
Long before any current military activity.
That alone eliminates this class of explanation.

2) Location dependence

The behavior changes with:
- neutral integrity
- grounding quality
- transformer replacement
It does NOT correlate with:
- outdoor orientation
- shielding attempts

3) Thermal throttling does not fit observations.
- Core temperature does not rise
- Power limits are not hit
- Performance counters remain normal.
- Frequency changes occur without thermal correlation.

Also, if this were thermal throttling:
Battery operation would consistently improve behavior.
It does not.

4) Frequency scale mismatch
If 8–12 GHz radiation were strong enough to cause localized silicon heating:
- RF inspection would detect it.
- Other electronics would misbehave.
- This would not selectively manifest as input smoothness degradation.
That level of exposure would not be subtle.

I agree on one thing only:
Frequency instability is observable. But I attribute it to reference instability and synchronization degradation, not thermal or radiative heating.
So while the hypothesis is interesting, it does not match the long-term, location-bound, infrastructure-dependent nature of the issue I am dealing with.

[Inputdelay]

The questioner mentioned an airport located 5 kilometers from his home. For an airport radar, 5 kilometers isn't that far. Of course, there's no direct beam to the questioner's apartment (the result of a direct beam would be obvious and catastrophic), but even a side or reflected beam can be significant.

Is there any way to detect and understand that it's radar? Even though the airport hasn't been used for military purposes for at least a year.

[Inputdelay]

Wow, finally, something which shows something of substance.

This is a widely known and understood issue with touchpads and it is caused by common mode noise. Touchpad has capacitive sensing matrix which is scanned at high frequency detecting miniscule capacitance changes when a finger is moved. Quite sensitive and finicky devices. Isolated switching adapters produce lots of common mode noise and when you put finger on the touchpad, it becomes a capacitive path for the common mode, overloading and disturbing the capacitve sense matrix. Usually happens with crappy touchpads or if power adapter is not original. Original manufacturer adapters are designed with specific switching frequencies in mind which don't interfere with coresponding touchpads on their laptops and in general has higher quality and filtering. If it happens with original adapter - it is just crap.

There is also noise which comes from AC line, not just from power adapter, although it is rare to be so strong. Proper ground in the building can reduce that, it is true.

But such common mode noise should not affect USB mouse, if it is not at cosmic high levels. Touchpad and external mouse might both seem like a mouse, but under the hood they are very different devices. Regarding the external noise, touchpad is very sensitive, USB mouse is very robust.

Make a video with USB mouse. I want to see it.

I have asked a friend to make a video; it will take some time for him to get one and record it, but that was just a showcase, since at that place there is no longer an issue.

He didn't record 'before' with the wired mouse. So there would be only 'after the electrical work video' that showcases the cursor movement via USB mouse. So I don't think it has any point, only to see the end result without the 'before' result.

[al777]

If 8–12 GHz radiation were strong enough to cause localized silicon heating:
- RF inspection would detect it.

...from your earlier message: "I am attaching a file of the device used to take measurements at my place via the Central Branch of the UCRF."
...and a cropped fragment from that image is attached.

Nope. They wouldn't. All the best.

[Manul]

I have asked a friend to make a video; it will take some time for him to get one and record it, but that was just a showcase, since at that place there is no longer an issue.

He didn't record 'before' with the wired mouse. So there would be only 'after the electrical work video' that showcases the cursor movement via USB mouse. So I don't think it has any point, only to see the end result without the 'before' result.

There is no point in making video if it works well, I can look at my own mouse. Show a badly working one. Wired on USB. If you have issue at your place, make a clear side by side slow motion video like your friend did, even better, put grid paper under the mouse, lay the monitor horizontally and film from up, the more scientific, the better. This is engineering forum. We like data, we like numbers and measurements. You will get much more involvement if you provide that. Let everyone see how that mouse behaves.

[Manul]

Bottom line is, your friend's video showing misbehaving touchpad is a pretty common issue. Nothing weird there. But what you described is weird, almost unbelievable. And I'm pretty sure the nature of your issue is different. So provide some data clearly showcasing the issue and we will see then.

[NE666]

External electrical parameters (neutral-ground voltage, resistance, DC offsets, transformer changes) correlate with behavior.

How can that be, when you've told us that the same effect can also be seen when you're running on battery power bank, during periods of mains power loss?

Besides, those AC line parameters aren't important in and of themselves. It's only what's happening on the DC side of the SMPS that matters, so look for quality or interference issues there.

Let's say you're right. What devices do I need to measure this, what methods are needed for this, what indicators should I look at, what should I look at, what is normal, and what is not?

A relatively inexpensive DSO is the tool for the job. See video's from Dave et al for a discussion on the relevance of these values, and how they measured. Compare to, ideally, the specifications provided by the manufacturer of your PC's actual power supply model, or against the generic specs for its type e.g. ATX

[NE666]

The behavior changes with:
- neutral integrity
- grounding quality
- transformer replacement

I'll say it again. NO, it doesn't. If, as you've told us, you've run the affected system from battery, when the mains power grid is offline, then any observed effect has NOTHING to do with any of the above.

It does NOT correlate with:

- shielding attempts

Again, no. As I posted previously, your shielding attempts as you described them were flawed. Therefore they do not provide evidence to support your conclusion.

[R3]

External electrical parameters (neutral-ground voltage, resistance, DC offsets, transformer changes) correlate with behavior.

How can that be, when you've told us that the same effect can also be seen when you're running on battery power bank, during periods of mains power loss?

Besides, those AC line parameters aren't important in and of themselves. It's only what's happening on the DC side of the SMPS that matters, so look for quality or interference issues there.

Let's say you're right. What devices do I need to measure this, what methods are needed for this, what indicators should I look at, what should I look at, what is normal, and what is not?

A relatively inexpensive DSO is the tool for the job. See video's from Dave et al for a discussion on the relevance of these values, and how they measured. Compare to, ideally, the specifications provided by the manufacturer of your PC's actual power supply model, or against the generic specs for its type e.g. ATX

Hello, I have a similar problem and took similar measures, including replacing the PC, just like this person, and it didn't help. That is, when replacing different PCs, does it even make sense to try to measure anything in them? They can't all break down at the same time in one place and be instant

[Inputdelay]

First, about battery operation.

Running a system from a battery does NOT isolate it from the electrical environment in the way you are assuming.

The system is still capacitively and parasitically coupled to:
-building wiring
-earth reference
-surrounding conductive structures
-monitor and peripheral grounds
-cable shields acting as antennas

Especially when:
-The monitor remains mains powered.
- The PC chassis is floating.
-Neutral and ground potentials are unstable.
Battery operation removes differential-mode noise from the mains.
It does NOT remove common-mode noise, reference drift, or environmental coupling.

So the statement:
"If it happens on battery, it cannot be related to neutral or grounding."
is technically incorrect.

Second, about shielding.

I am not using shielding attempts as proof that EMI is or is not the cause.
I am stating that shielding and filtering attempts did not produce stable or lasting changes.

Second, about shielding.

I am not using shielding attempts as proof that EMI is or is not the cause.
I am stating that shielding and filtering attempts did not produce stable or lasting changes.

That is an observation, not a conclusion.
So far, no such model has been proposed.

What I am claiming is narrow and specific:

The behavior correlates with changes in external reference conditions and infrastructure state.
It does not correlate with internal software or hardware changes.

This is based on:
-identical systems behaving differently in different locations
-identical internal telemetry in good and bad environments
-behavior changes after transformer and neutral maintenance
- Day/night effects disappearing after introducing a local ground reference
If you believe this cannot physically happen, then the correct next step is not dismissal,
but identifying which external parameters must be measured to falsify this hypothesis.
I am explicitly asking for that.
I am here to narrow causes, not to win an argument.

[Inputdelay]

Bottom line is, your friend's video showing misbehaving touchpad is a pretty common issue. Nothing weird there. But what you described is weird, almost unbelievable. And I'm pretty sure the nature of your issue is different. So provide some data clearly showcasing the issue and we will see then.

I understand your position, but I need to draw a clear boundary here; otherwise, this discussion keeps looping.
I am not refusing measurements.
I am refusing measurements that do not map to the hypothesis space I am investigating.

About video proof:
I physically cannot reproduce my friend's 1000 fps experiment at my location.
I do not have a high-speed camera capable of resolving behavior above my display refresh rate.
My phones are limited to 240 fps, while my main display is 360 Hz.
Filming a 360 Hz system at 240 fps will, by definition, alias and hide the effect.

Yes, I can film:
- a 120 Hz laptop at 240 fps
But that does not represent the system where the problem manifests most strongly.
So the result would be technically misleading, not “more scientific.”

What I am investigating is

- Why the same hardware, same OS, same drivers, and same software?
- reports identical internal metrics
- but produces different temporal behavior in different environments
-The only measurable internal difference I ever see is the difference in the way of cable management, especially the display port or HDMI position, closer to a wall or further, and the same goes for power cables, and it still depends on the display port or HDMI position.

Everything else stays nominal.

That strongly suggests:
The effect is upstream of observable application-level input/output.
and downstream of environmental reference conditions.

This is why I am focusing on:
- neutral / ground integrity
- reference potential drift
- distribution transformer changes
- infrastructure-level effects and not on re-proving that a mouse cursor can move smoothly on a clean system.
I am not asking anyone to “believe” me.
I am asking which external parameters should be measured.
to explain why the same digital system behaves differently without internal state changes. If the only acceptable path forward is
“Show me a visibly broken mouse, or it doesn’t exist.” then we are no longer doing engineering diagnostics, we are doing courtroom theatrics.

[Inputdelay]

I want to clarify a few points once, because they are repeatedly misunderstood.

First, about cables. The behavior of the system is directly affected by:
DisplayPort/HDMI cable routing, distance of these cables from walls, proximity to power cables, whether signal and power cables cross or run in parallel, relative spacing between all connected cables
Changing only the physical layout of cables, without changing any hardware, software, or power source, produces noticeable changes in responsiveness and visual clarity.
This alone indicates coupling effects. No internal PC metric will reflect this.

Second, about internal metrics and ECC.
There are no meaningful changes in FPS, frametimes, temperatures, voltages, or logged errors between good and bad states. That is expected.
They do not necessarily expose their activity to user-level telemetry.
Their job is to preserve correctness and stability, not to report discomfort.
A system that immediately crashed under interference would be easy to diagnose.
A system that stays operational by absorbing timing and synchronization stress is much harder to diagnose.
This does not mean the system is “working correctly” in a performance-sensitive context.

Third, about “proof” and measurements.
I am not here to prove that the problem exists.
That has already been established by:
identical systems behaving differently in different locations, visitors observing the same behavior, and the disappearance of the issue when hardware is moved elsewhere. The remaining task is to identify the external cause.
If someone believes this is impossible, then the correct response is not dismissal but proposing concrete measurements that target external coupling, reference stability, or environmental conditions.
Finally, a practical note.
I am willing to pay for direct, hands-on technical help.
I am not looking for arguments or philosophy.
I am looking for a person who understands where to look next.