My journey with EMI/RFI, a list of tested fixes
Posted: 03 Feb 2026, 22:02
This thread documents my short journey troubleshooting EMI/RFI-related issues in my PC setup. Over the last couple of weeks I sank most of my free time into it, (borderline obsessive at times…) but I’ve now reached the point of clear diminishing returns.
My main takeaway is that interference-related problems may be more common with modern hardware than many people realize, even in this forum, and they’re easy to miss unless you’ve understood the symptoms. Most will never realize. But the difference can be dramatic for both audio and display smoothness quality.
What surprised me most is how broad the contributing elements can be. Even basic changes like router Wi-Fi configuration (channel selection, transmit power etc.) produced tangible differences in my case. And nearly everyone has a router nearby, regardless of whether they live in an area with clean air and power.
I’m writing this as a PC enthusiast first. For what it’s worth, I have a PhD in EE (electromagnetics/photonics), and that mostly just helped confirm I wasn’t chasing ghosts.
I first got into latency optimization about 10 months ago, after upgrading my HW. I went down the latency-optimization rabbit hole: BIOS settings, memory timings, background processes, driver/OS tweaks, most things I knew about but hadn’t realized could matter this much in practice. A sad testament that modern consumer hardware/software is not build for latency in mind out of the box…
Months after I’d settled on what I thought was my “final” set of optimizations, I noticed, almost by accident while gaming, that routing some USB devices through simple adapters (adding an extra connection in the chain) improved visual smoothness. That didn’t make sense. I’d just added another interface and, in theory, another point of failure. My first suspicion was that I was unintentionally changing the electrical characteristics of the link, things like impedance and how much high-frequency noise can pass. This led me to diagnose a broader set of noise issues that had been heavily plugging my setup. To keep this short I’m going to switch to bullet points from here, listing all adjustments that worked. Right now, this feels like a fundamentally new experience, something I honestly didn’t think could be possible. It reminds me of gaming 10–15 years ago. But that may also be nostalgia…
My main takeaway is that interference-related problems may be more common with modern hardware than many people realize, even in this forum, and they’re easy to miss unless you’ve understood the symptoms. Most will never realize. But the difference can be dramatic for both audio and display smoothness quality.
What surprised me most is how broad the contributing elements can be. Even basic changes like router Wi-Fi configuration (channel selection, transmit power etc.) produced tangible differences in my case. And nearly everyone has a router nearby, regardless of whether they live in an area with clean air and power.
I’m writing this as a PC enthusiast first. For what it’s worth, I have a PhD in EE (electromagnetics/photonics), and that mostly just helped confirm I wasn’t chasing ghosts.
I first got into latency optimization about 10 months ago, after upgrading my HW. I went down the latency-optimization rabbit hole: BIOS settings, memory timings, background processes, driver/OS tweaks, most things I knew about but hadn’t realized could matter this much in practice. A sad testament that modern consumer hardware/software is not build for latency in mind out of the box…
Months after I’d settled on what I thought was my “final” set of optimizations, I noticed, almost by accident while gaming, that routing some USB devices through simple adapters (adding an extra connection in the chain) improved visual smoothness. That didn’t make sense. I’d just added another interface and, in theory, another point of failure. My first suspicion was that I was unintentionally changing the electrical characteristics of the link, things like impedance and how much high-frequency noise can pass. This led me to diagnose a broader set of noise issues that had been heavily plugging my setup. To keep this short I’m going to switch to bullet points from here, listing all adjustments that worked. Right now, this feels like a fundamentally new experience, something I honestly didn’t think could be possible. It reminds me of gaming 10–15 years ago. But that may also be nostalgia…
- Ferrite cores everywhere, on both ends of most cables (USB, power, display). I also added ferrites to other household devices (yes, even a toothbrush charger in the bathroom). Every single physical change was documented and tested in controlled gaming environments, and in my setup the improvements were consistent. The fact that ferrites helped on devices not even on the same outlet may suggests some degree of RFI current coupling through the home wiring. A high-end UPS may help with that, but ferrites still seem very beneficial overall.
- Farraday fabrics, everywhere, including the PC case panels (I never liked glass panels anyway), the backs of the monitors, bookshelf speakers and amplifier, my Stream Deck, and even parts of power strips (done safely, these materials are highly conductive). The only areas that were not covered are the front and top panels of my case where mesh screens were installed for airflow, they had a similar effectiveness. I also added shielding to the wall behind my desk. Everything was installed carefully so it didn’t ruin the look of the setup. Also, I realized you have to be careful about bonding/connecting different pieces of shielding. If you tie everything together incorrectly, you can unintentionally create a return path that routes conducted noise back toward the device, basically defeating the point. It took a fair amount of trial and error to get the grounding/bonding strategy right, but once I did, the results were noticeably better. You want them to act as shields, like a faraday cage.
- Ferrite rods & toroids. On top of the usual clamp-on ferrites, I tried those cheap antenna ferrite rods/toroids (a few bucks each on AliExpress). I dropped them in “hot spots” where cable bundles and shielding fabric pass close together, and stacked a few on the speakers. My intuition is they act as dissipators for RF currents passing through the shields, the same way cores do for wires. Perhaps these would not be needed if I did proper grounding/bonding everywhere, but that’s a whole project on its own. Arranging a few rods and regular ferrite clamps INSIDE the case also brought improvements
- Sleeving cables and building pre-shielded ones. You’ll see people online claim shielded Ethernet is “a waste of money,” but that could not be further from the truth, buy one, even if it’s only for peace of mind. I did an exhaustive test with the shielded and unshielded one and the change was very noticeable when I knew where to look. A lot of people simply have no idea how to notice interference issues. Audiophiles know.
- Wi-Fi tweaks (last step). I suspected most of the noise was coming from outside (I’m in urban LA), but in-home RF can absolutely add to the mess. I’m obviously not going to unplug a fridge for latency gains in a competitive FPS, but I did end up dialing my router’s transmit power down to 10%, shielding the router and simplifying the 2.4 GHz settings: G-only, channel 1, 20 MHz bandwidth, WPS off. Yeah, that’s “suboptimal” on paper, but in my apartment, Wi-Fi is basically just for smart lights/vacuums.