Unstable Pc performance and lack of visual smoothness.

[Ranayna]

I do know from experience a flash drive plugged in next to a bluetooth or wireless mouse receiver can interfere and cause the mouse to get glitchy... (EMI go brrr)
But the proximity of an un-powered flashdrive should not have any effects, unless the things emits ionising radiation.

That's a well known USB 3 issue. Apparently the base frequency of USB3 is 2.5 GHz, close enough to 2.4 GHz, that a badly shielded port can interrupt 2.4 GHz wireless connections, as they are used for Bluetooth and many other mouse dongles.

[temperance]

Where I am still stuck is the next practical step.
1) Mains-connected devices can make things worse in a repeatable way.
2) Removing them does not make the system stable.
3) The battery power source itself introduces noise.
Given this, what would you suggest as the most meaningful next experiment to identify the dominant coupling path into the PC?
What should be the next step forward?
1) Measurements to focus on.
2) Locations to measure.
3) Specific parts of the PC setup to isolate.
I am trying to move forward experimentally.

Moving forward with simple experiments is going to be difficult.

You must find a way to inject common mode noise over a wide frequency range into the mains cable of either the screen or the computer. Standard tests to perform such things do exist. But buying the equipment is very expensive.

You could try measuring the common mode noise with the TV and sat receiver connected/not connected in order to find a difference. But again, this requires expensive equipment. (Spectrum analyser, LISN network,...)

I do have both but you are not around the corner.

[Inputdelay]

Where I am still stuck is the next practical step.
1) Mains-connected devices can make things worse in a repeatable way.
2) Removing them does not make the system stable.
3) The battery power source itself introduces noise.
Given this, what would you suggest as the most meaningful next experiment to identify the dominant coupling path into the PC?
What should be the next step forward?
1) Measurements to focus on.
2) Locations to measure.
3) Specific parts of the PC setup to isolate.
I am trying to move forward experimentally.

Moving forward with simple experiments is going to be difficult.

You must find a way to inject common mode noise over a wide frequency range into the mains cable of either the screen or the computer. Standard tests to perform such things do exist. But buying the equipment is very expensive.

You could try measuring the common mode noise with the TV and sat receiver connected/not connected in order to find a difference. But again, this requires expensive equipment. (Spectrum analyser, LISN network,...)

I do have both but you are not around the corner.

That is exactly the wall I have run into. I understand that proper investigation of common-mode noise requires equipment like LISNs, spectrum analyzers. I fully agree that without this, conclusions are limited.

The problem is practical access. I am in Ukraine, and since the war started, access to university and lab equipment has become extremely difficult. I studied physics, electronics, and micro/nanoelectronics, but the relevant equipment was removed, sold, or transferred. I contacted former professors and several companies that do have such equipment, but either they didn’t have it or none agreed to perform on-site measurements or experiments.

I am simply stuck without the tools required to confirm anything.

If you know of people, labs, service companies, or even simplified diagnostic approaches that could be realistically accessible in my situation, that would be extremely helpful. Otherwise I am at a dead end where the next step is clear, but verification is not possible.

[temperance]

on-site measurements or experiments

On site doesn't help. A LISN isolates the PC from the mains noise anyhow. You have to bring the PC and the TV + sat receiver to the lab. That will do. But I don't know anyone in Ukraine with such equipment.


But you can investigate the GND connections inside the PC. (And the "EMI fingers" for all USB connections)

[radiolistener]

Have you checked the temperatures of the components? Is the case ventilation adequate?

The artifact-related symptoms resemble an intermittent GPU solder joint failure, which can occur, for example, due to overheating.

[Inputdelay]

Have you checked the temperatures of the components? Is the case ventilation adequate?

The artifact-related symptoms resemble an intermittent GPU solder joint failure, which can occur, for example, due to overheating.

Yes, temperatures were checked. GPU is typically around 45-50 C and reaches at most 64-65 C under load. CPU shows similar behavior and stays well within normal limits. Case ventilation is adequate and airflow is not restricted. The issue appears even at idle and low load, not correlated with temperature or thermal stress, so overheating or intermittent solder joint failure due to heat does not seem likely.

[temperance]

In your first post you said that the bots appear fast but the player is lagging.

It means that the CPU is doing it's job. The game in the CPU is running full speed.

Next there are instruction buffers between the CPU and the GPU, the render queue. Depending the on the render queue size, you are, you are looking at a delayed version of the gaming going on in the CPU if the GPU can't keep up which seems to be happening in your case. What are the video card driver settings for the render queue size?

The same goes for the input devices. They put data in circular buffers and the CPU handles them when it has time to do so.

[Analog Kid]

Just throwing my 9¢ in here (adjusted for inflation):

So it seems there's something wrong with your PC.

Why don't you try getting a new fucking computer?

[Inputdelay]

Just throwing my 9¢ in here (adjusted for inflation):

So it seems there's something wrong with your PC.

Why don't you try getting a new fucking computer?

I already replaced all the hardware I used to have. Different PC, different CPU, motherboard, RAM, PSU, GPU, monitor, and cables. I also tested different architectures. The behavior does not change, so it does not appear to be tied to a specific piece of hardware.

[Inputdelay]

In your first post you said that the bots appear fast but the player is lagging.

It means that the CPU is doing it's job. The game in the CPU is running full speed.

Next there are instruction buffers between the CPU and the GPU, the render queue. Depending the on the render queue size, you are, you are looking at a delayed version of the gaming going on in the CPU if the GPU can't keep up which seems to be happening in your case. What are the video card driver settings for the render queue size?

The same goes for the input devices. They put data in circular buffers and the CPU handles them when it has time to do so.

I tested literally all NVIDIA driver versions available for my GPU. Each driver reinstall noticeably changes how the game feels. Not every driver makes it better or worse, but every driver changes the perception. This effect is repeatable.

I tested all NVIDIA Control Panel options and used NVIDIA Profile Inspector, including all available render queue values. None of these settings consistently resolves the issue. The driver version itself matters much more than individual settings.

In games GPU usage is typically around 25-50% at most. CPU usage is also low, usually below 35% and sometimes approaching 50%. The system is not CPU- or GPU-bound.

The same behavior is observable outside games (desktop mouse movement), so this does not look like a simple render queue backlog or GPU not keeping up with the CPU

[Manul]

How much longer will you continue these conversations without trying to pinpoint what exactly happens to your PC when it is affected? Is it not not at all interesting?

You say it affects any PC in specific locations. So it might also be my PC and my house, right? Anyone's. So if it is true, it is very interesting, we all can learn and benefit from that.

But you only say "desync", "mouse feels heavy". I have no idea what the fuck is desync or a heavy mouse.

If I feel that my car runs bad when humidity is high, I don't talk about my feelings or start studying meteorology, first of all I'm checking the engine. Cause perhaps number two spark plug has bad electrical isolation and that is the reason why humidity affects it. If it feels wrong, something must be wrong. And we can learn a lot by pinpointing what exactly is wrong.

So is it interference with USB devices, increased jitter, Display port interference, something with LCD panel, RTC glitching or what?

[thm_w]

If it feels wrong, something must be wrong.

I disagree, it can feel wrong and be working perfectly for various reasons:
- anxiety disorder
- palinopsia or other visual disorders https://my.clevelandclinic.org/health/symptoms/palinopsia

You can read hundreds of pages of this on the blurbusters forum section linked above, if you feel like wasting your time. I wouldn't. None of them will provide hard data.

[Analog Kid]

OK, let's look at this objectively:

OP has what they perceive as an ongoing problem with their PC.
They've exhausted every option to determine the cause by replacing stuff, both hardware and software.

Therefore the problem—if it actually exists and is not just a matter of perception on their part—must be due to one of the following things:

• interfering magnetic fields
• interfering EM fields (RF, etc.)
• noise or other anomaly on the power source
• some other unknown source of interference (????)

So realistically, which of these could it be?
Have I omitted something from this list?
Anything in this list that can be automatically discounted?

One thing, possible RF interference, has not been definitively ruled out according to what the OP posted, since the RF scan that was done was limited in frequency (forget what the figure was), so anything above that was not measured.

[thm_w]

OK, let's look at this objectively:

OP has what they perceive as an ongoing problem with their PC.
They've exhausted every option to determine the cause by replacing stuff, both hardware and software.

Therefore the problem—if it actually exists and is not just a matter of perception on their part—must be due to one of the following things:

• interfering magnetic fields
• interfering EM fields (RF, etc.)
• noise or other anomaly on the power source
• some other unknown source of interference (????)

So realistically, which of these could it be?
Have I omitted something from this list?
Anything in this list that can be automatically discounted?

One thing, possible RF interference, has not been definitively ruled out according to what the OP posted, since the RF scan that was done was limited in frequency (forget what the figure was), so anything above that was not measured.

There is zero point in scanning RF and coming up with crackpot theories until you've quantified and measured the problem and proven it exists.

Interference/EMI/RF interacting with the hardware inside the PC would be clearly seen on: frame times, latency tools, screen recordings, error log, or nvidia reflex (if compatible).
The only place it wouldn't be measured without an external tool is in the monitor cable, monitor itself, and mouse (input latency). And that should already be ruled out as multiple monitors have "the exact same issue".

[Inputdelay]

How much longer will you continue these conversations without trying to pinpoint what exactly happens to your PC when it is affected? Is it not not at all interesting?

You say it affects any PC in specific locations. So it might also be my PC and my house, right? Anyone's. So if it is true, it is very interesting, we all can learn and benefit from that.

But you only say "desync", "mouse feels heavy". I have no idea what the fuck is desync or a heavy mouse.

If I feel that my car runs bad when humidity is high, I don't talk about my feelings or start studying meteorology, first of all I'm checking the engine. Cause perhaps number two spark plug has bad electrical isolation and that is the reason why humidity affects it. If it feels wrong, something must be wrong. And we can learn a lot by pinpointing what exactly is wrong.

So is it interference with USB devices, increased jitter, Display port interference, something with LCD panel, RTC glitching or what?

Of course it is interesting to me what exactly is happening to the PC. That is the whole reason I am still investigating this. Over a long period of time I tried many different tests and tools, but I did not find a stable correlation that clearly points to a single system failure.

Yes, based on my observations, it likely affects any PC, laptops included, and I also notice abnormal behavior on phones while charging in the same locations. If this is true, I agree it would be interesting and relevant to many people.

By “desync” or “heavy mouse,” I mean that mouse behavior is not consistent over time: the same physical movement produces different on-screen responses. So far I have not found any standard system metric that correlates with this change in behavior.

As for possible causes: USB interference is possible, DisplayPort coupling is also possible, and the monitor panel itself is unlikely, as multiple different monitors show the same behavior. Timers are harder to judge; I did not observe obvious timer anomalies using tools like MeasureSleep.

I am not refusing measurements. If specific, realistic measurements are suggested, I am willing to run them. What I am not willing to do is buy very expensive lab equipment just to prove that a problem exists. My goal here is to understand and mitigate the issue, not to convince anyone for the sake of argument.

I also ran an animation error test using Intel PresentMon. Here's a test I made when I had the problem occurring.

[Inputdelay]

If it feels wrong, something must be wrong.

I disagree, it can feel wrong and be working perfectly for various reasons:
- anxiety disorder
- palinopsia or other visual disorders https://my.clevelandclinic.org/health/symptoms/palinopsia

You can read hundreds of pages of this on the blurbusters forum section linked above, if you feel like wasting your time. I wouldn't. None of them will provide hard data.

This is not a psychological or perceptual issue. I am medically healthy, and this behavior is observed not only by me. Multiple other people have visited my place, including friends, family members, and people working in electronics, and they independently noticed the same abnormal behavior. Friends have also brought their own computers, and the issue was observable there as well.

Because the same effect is noticed by different people on different systems in the same location, this cannot reasonably be explained by anxiety, visual disorders, or subjective perception. Redirecting to psychological explanations does not help move this case forward.

I am not refusing measurements. On the contrary, I am explicitly asking for them.

What I am not willing to do is spend thousands of dollars on specialized lab equipment just to "prove" something on the internet. My goal is not to convince others that a problem exists, but to identify what it is and how to eliminate it.

If you have concrete, practical measurement ideas that do not require extremely expensive equipment, I am open to trying them.

[Inputdelay]

OK, let's look at this objectively:

OP has what they perceive as an ongoing problem with their PC.
They've exhausted every option to determine the cause by replacing stuff, both hardware and software.

Therefore the problem—if it actually exists and is not just a matter of perception on their part—must be due to one of the following things:

• interfering magnetic fields
• interfering EM fields (RF, etc.)
• noise or other anomaly on the power source
• some other unknown source of interference (????)

So realistically, which of these could it be?
Have I omitted something from this list?
Anything in this list that can be automatically discounted?

One thing, possible RF interference, has not been definitively ruled out according to what the OP posted, since the RF scan that was done was limited in frequency (forget what the figure was), so anything above that was not measured.

I am willing to re-invite the national RF institute (Ukrainian Radio Frequency Center) to perform measurements, but to do this properly I need to understand exactly what should be measured and where.

Previously, specialists came with RF measurement equipment (the device photo was posted earlier in this thread). They could observe the same abnormal behavior on the display, but we were unable to identify any clear RF source inside the house. Without precise guidance, the measurements become unfocused and inconclusive.

So the practical question is:

1) Which parameters should be measured specifically?

2) In what frequency ranges

3) And at which physical points (main cable, near PC, near monitor, near DisplayPort/USB, etc.)

I have frequent and long power outages here. During these outages, the PC runs from battery power stations only. Even in this case, the system behavior still changes over time while running on the same battery source. This happens at both battery stations.

If RF is still considered a realistic candidate, I need concrete guidance on what exactly should be measured so that calling the RF institute again makes sense and is not just another blind scan.

[jc101]

So is the issue reflected in things connected to the PC via cables, the mouse, monitor etc.?

Does a laptop, with everything integral, show similar issues? No mouse connected, run on internal battery etc.

[Inputdelay]

So is the issue reflected in things connected to the PC via cables, the mouse, monitor etc.?

Does a laptop, with everything integral, show similar issues? No mouse connected, run on internal battery etc.

This was already addressed earlier in the thread. Yes, the issue remains on a laptop both on-grid and off-grid. Yes, it also remains when everything is disconnected and the laptop is running purely on battery.
We already identified that the touchpad itself can be a contributing factor. However, even if no keys are pressed and the touchpad is not used at all - simply opening an application and observing motion - the behavior is still present on battery.

[Manul]

If it feels wrong, something must be wrong.

I disagree, it can feel wrong and be working perfectly for various reasons:
- anxiety disorder
- palinopsia or other visual disorders https://my.clevelandclinic.org/health/symptoms/palinopsia

You can read hundreds of pages of this on the blurbusters forum section linked above, if you feel like wasting your time. I wouldn't. None of them will provide hard data.

Yes, when we watch a display and hold mouse in a hand, we are iherently in the loop, we are part of the system. So it still holds, that something must be wrong.

OK, let's look at this objectively:

OP has what they perceive as an ongoing problem with their PC.
They've exhausted every option to determine the cause by replacing stuff, both hardware and software.

Therefore the problem—if it actually exists and is not just a matter of perception on their part—must be due to one of the following things:

• interfering magnetic fields
• interfering EM fields (RF, etc.)
• noise or other anomaly on the power source
• some other unknown source of interference (????)

So realistically, which of these could it be?
Have I omitted something from this list?
Anything in this list that can be automatically discounted?

One thing, possible RF interference, has not been definitively ruled out according to what the OP posted, since the RF scan that was done was limited in frequency (forget what the figure was), so anything above that was not measured.

That are all reasonable suggestions. But may I ask a question, if you have a PCB which works while cold, but fails when hot, what do you do? Pretty much everyone will start troubleshooting along the lines of cooling some places with cold spray, heating places with hot air and so on. Often it will take less than 10 minutes to pinpoint.

So perhaps it is affected by temperature, external EMI or whatever. But it means some area, some component is vulnerable to that. It is very important to find which. That is the only professional, scientific approach.

I once designed a one-off USB device. And the customer started complaining that it is repeatedly failing. Turns out, he was running a device in very dry environment, connected to a laptop and synthetic carpet in a room. Static electricity made it misbehave. So it was my design blunder, and I fixed that, I just never noticed it during development, cause my lab environment is all antistatic.

I

Interference/EMI/RF interacting with the hardware inside the PC would be clearly seen on: frame times, latency tools, screen recordings, error log, or nvidia reflex (if compatible).
The only place it wouldn't be measured without an external tool is in the monitor cable, monitor itself, and mouse (input latency). And that should already be ruled out as multiple monitors have "the exact same issue".

I agree, but I would not say it is 100% ruled out by replacing devices. Every USB cable is similar to another one, most LCD panels are very similar in design, so it might be possible that multiple devices are affected. Although not likely. Anyway, if it is true, pinpointing the exact component is the key.

[NE666]

If RF is still considered a realistic candidate, I need concrete guidance on what exactly should be measured so that calling the RF institute again makes sense and is not just another blind scan.

Why?

That's the same as not going to see a doctor because you can't properly instruct them on how to perform a medical diagnosis.

If you suspect that there's an external interference source causing you problems, then just tell them that. Explain the symptoms. Explain that it affects not just one piece of equipment but all similar equipment moved into that location. Tell them that it is not removed by running from battery, so it's likely not a power supply issue.

Then step back and let them do the work they are educated and trained to do. They don't need you to be able to hold their hand.

[tszaboo]

I think do this, and close this thread, as everything has been discussed already.

[Inputdelay]

If RF is still considered a realistic candidate, I need concrete guidance on what exactly should be measured so that calling the RF institute again makes sense and is not just another blind scan.

Why?

That's the same as not going to see a doctor because you can't properly instruct them on how to perform a medical diagnosis.

If you suspect that there's an external interference source causing you problems, then just tell them that. Explain the symptoms. Explain that it affects not just one piece of equipment but all similar equipment moved into that location. Tell them that it is not removed by running from battery, so it's likely not a power supply issue.

Then step back and let them do the work they are educated and trained to do. They don't need you to be able to hold their hand.

A doctor already has a well-defined diagnostic framework, standardized procedures, and a clear scope of responsibility. An RF institute does not. They measure exactly what you ask them to measure, within a defined frequency range, with specific probes, for a limited time. If the task is vague, the result is a vague "everything is within limits" report - which already happened.

I have already called them once. They came, measured, and saw the issue visually but could not correlate it to anything measurable. That’s why I’m asking for concrete guidance now, not because I want to micromanage them, but because repeating the same blind scan is a waste of time and money.

[SteveThackery]

They came, measured, and saw the issue visually but could not correlate it to anything measurable.

Hang on, what do you mean, "they saw the issue"? What issue? You don't know what the issue is, yet. How do you know that what they saw is, in fact, what is causing your problem?

[negativ3]

The OP said a couple hundred posts ago that the phenomenon was repeatable on visitor systems and that symptoms go away when he travels to another location.