Could Modern PCs Be More Sensitive to EMI/Power Noise? My Old i5 + GTX 1060 Feels Smoother Than a 9800X3D + 4070 Super

[Roxtar]

Hey everyone,

It’s been a while since my last post. I’ve spent quite a lot of time trying to solve an input-lag issue on my main PC, and I eventually found that a double-conversion UPS made a very noticeable improvement.

However, over the last few months I’ve made some observations that I honestly can’t explain, so I wanted to share them here in case someone with more knowledge about EMI, grounding, PSU behavior, power quality, or electrical noise can shed some light on what I’m experiencing.

The two PCs

My main PC:

Ryzen 7 9800X3D
RTX 4070 Super
32 GB DDR5
Modern motherboard/PSU
Connected to my double-conversion UPS

And then there is my old PC, which I originally bought for around $400 back in 2015:

Intel i5
GTX 1060
8 GB DDR3
Cheap/older PSU

I had to use the old PC temporarily because I was RMAing some fans from my main PC.

Normally, I wouldn't even bother testing it for competitive gaming because I assumed the experience would be significantly worse.

But something really strange happened.

The old PC felt better

I connected the old PC directly to the wall and didn't connect it to my UPS.

I launched Valorant expecting significantly worse input latency.

Instead, the mouse and movement felt extremely smooth.

Honestly, it felt better than my main PC.

At first I thought it was just placebo, so I started experimenting.

Experiment #1 — Old PC + UPS

I connected the old PC to the same double-conversion UPS that I normally use with my main PC.

Result:

No noticeable change.

The old PC still felt smooth.

Experiment #2 — Multiple power strips

I then tried something that probably sounds ridiculous.

I plugged a power strip into the wall, connected several devices to it, and then connected a second power strip to the first one. I basically tried to put as much load/devices on the same power path as I reasonably could.

I connected the old PC and both monitors to this setup.

Result:

Still completely smooth.

No obvious increase in input lag or change in mouse feel.

At this point I was extremely confused.

Then I switched back to my main PC

I connected my main PC to the UPS again.

It felt roughly like I remembered it.

The UPS definitely seems to help compared with running the PC directly from the wall, but after spending time with the old PC, I realized that the improvement wasn't 100%.

The modern PC still felt noticeably less responsive to me.

Then I decided to repeat the experiment without the UPS.

Main PC directly from the wall

Initially, it actually felt relatively normal.

But after playing for some time, the experience gradually became worse.

Eventually, I reached roughly the same level of input lag that originally made me start investigating this problem.

This is where things get really interesting.

My first thought was that perhaps some kind of electrical noise is getting through the PSU, or that something is changing thermally/electrically over time.

I'm not claiming that's actually what's happening — I have no measurements to prove it.

I'm just trying to understand why the behavior seems to correlate with the power source and time.

Then I tried the two-power-strip setup with the main PC

This time, instead of the UPS, I used the same kind of power-strip setup that seemed completely harmless with the old PC.

And, surprisingly, the main PC felt even worse.

At this point I'm completely confused.

The part I can't explain

How can this:

i5 + GTX 1060 + 8 GB DDR3 + old/cheap PSU

feel significantly smoother to me than this:

Ryzen 7 9800X3D + RTX 4070 Super + 32 GB DDR5

when both are running the same game and I'm using the same monitors/peripherals?

And why does the old PC seem almost completely unaffected by these power experiments while the modern PC seems much more sensitive?

This has made me wonder whether something related to:

EMI/RFI
conducted electrical noise
grounding
PSU filtering
motherboard VRMs
USB noise
GPU power delivery
switching power supplies
ground loops
monitor/PC interaction
or some other form of electrical interference

could somehow be affecting the system.

I'm especially curious about the possibility that modern PCs might simply have different or more complex switching/power circuitry, potentially making them behave differently in the presence of electrical noise.

Again, I'm not saying this is definitely EMI or "dirty electricity." That's just one hypothesis I'm trying to investigate.

There could obviously be something completely different going on.

What makes this even stranger

The UPS helped my main PC substantially, which is why I originally thought I had basically found the solution.

But after comparing it directly against the old PC, I'm starting to think that the UPS may only be masking part of the problem.

The old PC seems to be much less sensitive to whatever is happening.

So I'm wondering:

Could the difference actually be related to EMI/power quality, or am I looking in completely the wrong direction?

Could a modern 9800X3D/4070 Super system somehow be more susceptible to electrical interference than an older, much simpler system?

Or are there other mechanisms that could explain why two PCs connected to the same environment can have such different perceived input latency?

If anyone has experience with EMI measurements, PSU filtering, grounding, power-quality testing, USB noise, or competitive-gaming input latency, I'd really appreciate your thoughts.

I'm particularly interested in ideas for actually measuring what's happening, rather than just continuing to swap cables and power sources and relying on subjective mouse feel.

At this point I'm genuinely tempted to sell the 9800X3D system and go back to the ancient i5/1060, which is absolutely ridiculous considering the hardware difference

So before I do something stupid, I'd love to hear what you guys think is actually going on.

[cybepine]

You left one very important thing out.

Is your old PC case full metal case and is the new open glass panel?

If you can try to put new PC components to full metal case and then do the same tests.
And of course no overclocking.

I can't really answer your other questions, because there are so many variables.

[qbBA]

Hey everyone,

It’s been a while since my last post. I’ve spent quite a lot of time trying to solve an input-lag issue on my main PC, and I eventually found that a double-conversion UPS made a very noticeable improvement.

However, over the last few months I’ve made some observations that I honestly can’t explain, so I wanted to share them here in case someone with more knowledge about EMI, grounding, PSU behavior, power quality, or electrical noise can shed some light on what I’m experiencing.

The two PCs

My main PC:

Ryzen 7 9800X3D
RTX 4070 Super
32 GB DDR5
Modern motherboard/PSU
Connected to my double-conversion UPS

And then there is my old PC, which I originally bought for around $400 back in 2015:

Intel i5
GTX 1060
8 GB DDR3
Cheap/older PSU

I had to use the old PC temporarily because I was RMAing some fans from my main PC.

Normally, I wouldn't even bother testing it for competitive gaming because I assumed the experience would be significantly worse.

But something really strange happened.

The old PC felt better

I connected the old PC directly to the wall and didn't connect it to my UPS.

I launched Valorant expecting significantly worse input latency.

Instead, the mouse and movement felt extremely smooth.

Honestly, it felt better than my main PC.

At first I thought it was just placebo, so I started experimenting.

Experiment #1 — Old PC + UPS

I connected the old PC to the same double-conversion UPS that I normally use with my main PC.

Result:

No noticeable change.

The old PC still felt smooth.

Experiment #2 — Multiple power strips

I then tried something that probably sounds ridiculous.

I plugged a power strip into the wall, connected several devices to it, and then connected a second power strip to the first one. I basically tried to put as much load/devices on the same power path as I reasonably could.

I connected the old PC and both monitors to this setup.

Result:

Still completely smooth.

No obvious increase in input lag or change in mouse feel.

At this point I was extremely confused.

Then I switched back to my main PC

I connected my main PC to the UPS again.

It felt roughly like I remembered it.

The UPS definitely seems to help compared with running the PC directly from the wall, but after spending time with the old PC, I realized that the improvement wasn't 100%.

The modern PC still felt noticeably less responsive to me.

Then I decided to repeat the experiment without the UPS.

Main PC directly from the wall

Initially, it actually felt relatively normal.

But after playing for some time, the experience gradually became worse.

Eventually, I reached roughly the same level of input lag that originally made me start investigating this problem.

This is where things get really interesting.

My first thought was that perhaps some kind of electrical noise is getting through the PSU, or that something is changing thermally/electrically over time.

I'm not claiming that's actually what's happening — I have no measurements to prove it.

I'm just trying to understand why the behavior seems to correlate with the power source and time.

Then I tried the two-power-strip setup with the main PC

This time, instead of the UPS, I used the same kind of power-strip setup that seemed completely harmless with the old PC.

And, surprisingly, the main PC felt even worse.

At this point I'm completely confused.

The part I can't explain

How can this:

i5 + GTX 1060 + 8 GB DDR3 + old/cheap PSU

feel significantly smoother to me than this:

Ryzen 7 9800X3D + RTX 4070 Super + 32 GB DDR5

when both are running the same game and I'm using the same monitors/peripherals?

And why does the old PC seem almost completely unaffected by these power experiments while the modern PC seems much more sensitive?

This has made me wonder whether something related to:

EMI/RFI
conducted electrical noise
grounding
PSU filtering
motherboard VRMs
USB noise
GPU power delivery
switching power supplies
ground loops
monitor/PC interaction
or some other form of electrical interference

could somehow be affecting the system.

I'm especially curious about the possibility that modern PCs might simply have different or more complex switching/power circuitry, potentially making them behave differently in the presence of electrical noise.

Again, I'm not saying this is definitely EMI or "dirty electricity." That's just one hypothesis I'm trying to investigate.

There could obviously be something completely different going on.

What makes this even stranger

The UPS helped my main PC substantially, which is why I originally thought I had basically found the solution.

But after comparing it directly against the old PC, I'm starting to think that the UPS may only be masking part of the problem.

The old PC seems to be much less sensitive to whatever is happening.

So I'm wondering:

Could the difference actually be related to EMI/power quality, or am I looking in completely the wrong direction?

Could a modern 9800X3D/4070 Super system somehow be more susceptible to electrical interference than an older, much simpler system?

Or are there other mechanisms that could explain why two PCs connected to the same environment can have such different perceived input latency?

If anyone has experience with EMI measurements, PSU filtering, grounding, power-quality testing, USB noise, or competitive-gaming input latency, I'd really appreciate your thoughts.

I'm particularly interested in ideas for actually measuring what's happening, rather than just continuing to swap cables and power sources and relying on subjective mouse feel.

At this point I'm genuinely tempted to sell the 9800X3D system and go back to the ancient i5/1060, which is absolutely ridiculous considering the hardware difference

So before I do something stupid, I'd love to hear what you guys think is actually going on.

Yeah, there really are a lot of variables, but I had a similar situation—the key difference was the motherboard, I think, because the RAM and motherboards were different. Plus, the older generation of DDR3 boards didn’t have all that extra stuff like C-State or R-Rebar; it’s really hard to list all the settings on the new boards. The older motherboards don’t have any of that, and even if they do, it’s just a very minimal set on the top-of-the-line models from the older generation. Even the PC case can have an impact, since it’s a solid-piece design. The guy above mentioned that the new ones have glass panels and are practically made of foil—a lot of things play a role. But you’re absolutely right that the older generation performed much better in games, and the K/D ratio was higher. even if you disable everything unnecessary on a new motherboard via a custom BIOS, it’ll be better—but not stable—and you’ll waste a lot of time because you won’t understand anything without the manual. Plus, new motherboards have a hidden BIOS that no program can detect; if it does, it’ll just show a set of numbers. Compare what’s available on the old and new models and put two and two together.

[MK92]

No, unless by "modern PC" you mean a 1991 PC vs 2026 PC, not just a couple of generations of difference in GPU and CPU.

My reserve PC is almost exactly the same as yours but with even older GPU, also from 2015 (gtx 960 2gb, i5 6500, and some random budget MSI board + low budget PSU), and it worked exactly the same as my main PC (with a quality PSU) until dedicated current was installed, and both PCs have a glass panel.

[jterri]

Hey everyone,

It’s been a while since my last post. I’ve spent quite a lot of time trying to solve an input-lag issue on my main PC, and I eventually found that a double-conversion UPS made a very noticeable improvement.

However, over the last few months I’ve made some observations that I honestly can’t explain, so I wanted to share them here in case someone with more knowledge about EMI, grounding, PSU behavior, power quality, or electrical noise can shed some light on what I’m experiencing.

The two PCs

My main PC:

Ryzen 7 9800X3D
RTX 4070 Super
32 GB DDR5
Modern motherboard/PSU
Connected to my double-conversion UPS

And then there is my old PC, which I originally bought for around $400 back in 2015:

Intel i5
GTX 1060
8 GB DDR3
Cheap/older PSU

I had to use the old PC temporarily because I was RMAing some fans from my main PC.

Normally, I wouldn't even bother testing it for competitive gaming because I assumed the experience would be significantly worse.

But something really strange happened.

The old PC felt better

I connected the old PC directly to the wall and didn't connect it to my UPS.

I launched Valorant expecting significantly worse input latency.

Instead, the mouse and movement felt extremely smooth.

Honestly, it felt better than my main PC.

At first I thought it was just placebo, so I started experimenting.

Experiment #1 — Old PC + UPS

I connected the old PC to the same double-conversion UPS that I normally use with my main PC.

Result:

No noticeable change.

The old PC still felt smooth.

Experiment #2 — Multiple power strips

I then tried something that probably sounds ridiculous.

I plugged a power strip into the wall, connected several devices to it, and then connected a second power strip to the first one. I basically tried to put as much load/devices on the same power path as I reasonably could.

I connected the old PC and both monitors to this setup.

Result:

Still completely smooth.

No obvious increase in input lag or change in mouse feel.

At this point I was extremely confused.

Then I switched back to my main PC

I connected my main PC to the UPS again.

It felt roughly like I remembered it.

The UPS definitely seems to help compared with running the PC directly from the wall, but after spending time with the old PC, I realized that the improvement wasn't 100%.

The modern PC still felt noticeably less responsive to me.

Then I decided to repeat the experiment without the UPS.

Main PC directly from the wall

Initially, it actually felt relatively normal.

But after playing for some time, the experience gradually became worse.

Eventually, I reached roughly the same level of input lag that originally made me start investigating this problem.

This is where things get really interesting.

My first thought was that perhaps some kind of electrical noise is getting through the PSU, or that something is changing thermally/electrically over time.

I'm not claiming that's actually what's happening — I have no measurements to prove it.

I'm just trying to understand why the behavior seems to correlate with the power source and time.

Then I tried the two-power-strip setup with the main PC

This time, instead of the UPS, I used the same kind of power-strip setup that seemed completely harmless with the old PC.

And, surprisingly, the main PC felt even worse.

At this point I'm completely confused.

The part I can't explain

How can this:

i5 + GTX 1060 + 8 GB DDR3 + old/cheap PSU

feel significantly smoother to me than this:

Ryzen 7 9800X3D + RTX 4070 Super + 32 GB DDR5

when both are running the same game and I'm using the same monitors/peripherals?

And why does the old PC seem almost completely unaffected by these power experiments while the modern PC seems much more sensitive?

This has made me wonder whether something related to:

EMI/RFI
conducted electrical noise
grounding
PSU filtering
motherboard VRMs
USB noise
GPU power delivery
switching power supplies
ground loops
monitor/PC interaction
or some other form of electrical interference

could somehow be affecting the system.

I'm especially curious about the possibility that modern PCs might simply have different or more complex switching/power circuitry, potentially making them behave differently in the presence of electrical noise.

Again, I'm not saying this is definitely EMI or "dirty electricity." That's just one hypothesis I'm trying to investigate.

There could obviously be something completely different going on.

What makes this even stranger

The UPS helped my main PC substantially, which is why I originally thought I had basically found the solution.

But after comparing it directly against the old PC, I'm starting to think that the UPS may only be masking part of the problem.

The old PC seems to be much less sensitive to whatever is happening.

So I'm wondering:

Could the difference actually be related to EMI/power quality, or am I looking in completely the wrong direction?

Could a modern 9800X3D/4070 Super system somehow be more susceptible to electrical interference than an older, much simpler system?

Or are there other mechanisms that could explain why two PCs connected to the same environment can have such different perceived input latency?

If anyone has experience with EMI measurements, PSU filtering, grounding, power-quality testing, USB noise, or competitive-gaming input latency, I'd really appreciate your thoughts.

I'm particularly interested in ideas for actually measuring what's happening, rather than just continuing to swap cables and power sources and relying on subjective mouse feel.

At this point I'm genuinely tempted to sell the 9800X3D system and go back to the ancient i5/1060, which is absolutely ridiculous considering the hardware difference

So before I do something stupid, I'd love to hear what you guys think is actually going on.

No I don’t think modern equipment is more sensitive to EMI than older equipment. It might be something with the electricity in your house and how your equipment filters out the power. I played on multiple older and worse PCs and laptops than what you have before I got my current high end gaming laptop and the issue has been completely the same regardless. I think the electricity in my house is so bad that as soon as I use anything on the electricity the equipment gets bricked completely and permanently. But idk if i were you and my older pc felt better than my newer pc I would just use the older pc and completely ditch the new one.

[gjk]

I want to say that I have the exact same behaviour here, my lower end system was operating way smoother than my current system.
(7800X3D/5070Ti)

my undervolted 9900K at 4.5Ghz with a 2080ti was way smoother than my current setup, where the lower end system was much sharper.
It is pretty obvious in games where you need to pan\move the camera fast, like RTS games, pulling over 600fps with the X3D feels very delayed and heavy.

installing the chipset drivers makes this problem worse, or even enabling expo.

there are threads to be found here and there about it:
https://www.reddit.com/r/overclocking/c ... ive_tried/

and yes,the double conversion ups does improve it when under load!

[ablemor]

No, this phenomenon also occurred on an old i5 3470 machine, although I don't know how modern that is considered to be.
For me, the UPS helped once and never again, plus I honestly don't think there's any expert here who could answer this question.