Chief Blur Buster wrote: ↑02 Oct 2020, 22:44
howiec wrote: ↑02 Oct 2020, 22:14
I'm sure someone else has said this before but as an electrical engineer, I can tell you that EMI should have zero effect on your "input lag"
UNLESS you're talking about wireless networking or peripherals (e.g. mouse & keyboard).
Part 1 of 2
Actually, it's rare but it has happened. I am crossposting something else I wrote on another forum at ArsTechnica:
____________
Interference problems can create really WEIRD effects for a lot of modern gadgets. When interference is strong enough -- and/or the shielding is weak enough.
Like the
LG 5K display that glitched if a WiFi router was nearby.
I've also gotten anecdotes from spectrum-analyzer using engineers/techs where they found weird videogaming sutters/glitches caused by electricity noise.
The interference was nearly (but not quite) strong enough to crash a computer. Instead, the computer was severely slowing/lagging from apparent error correcting behaviours.
We don't know what specific slowdowns were caused by -- flood of IRQ interrupt noise or error-correcting behaviours on SATA cables or noise on ECC DRAM buses or some other mysterious EMI-based slowdown effects, or a PCI Express bus trying to retransmit data during noise errors -- probably error correcting by the millions, like death by a million nanoseconds or microseconds that added up to continually visible slowdowns (game stutter, Window-dragging stutter, scrolling stutter, etc). Even when nothing was running in background (CPU/GPU near 0%), and scanners found no viruses.
Like thousands of microfreezes per minute.
The computer really performed erratically slow in a way that wasn't blameable on Microsoft Windows in particular as patterns persisted regardless of what was installed or reinstalled. And it was only when the interference was occuring. When the culprit appliance was removed, the computer went back to full speed. The engineer's testing equipment confirmed massive interference noise was somehow being injected into the computer that slowed down / stuttered everytime the noise occured. The massive amounts of error correction layers that portions of a modern computer has, apparently prevents the computer from crashing.
It's possible some components were suboptimal/defective or it was a cheap motherboard, or a poorly EMI-rejecting computer power supply. But at some point, some forms of interference can become too strong for a hardened computer to resist fully. And they apparently don't always crash right away thanks to all those error-correcting buses (PCI, SATA, NVMe, ECC RAM, USB, LAN, DisplayPort, etc, all of which use error correction now). It can be bad enough like an Internet connection losing 50% of its packets, but on a PCI Express bus, or USB cable, or NVMe SSD or whatever -- things really slow down there too if interference is strong enough but still lets enough data through that the computer doesn't crash.
All the numerous error correcting buses found in modern PCs, that didn't exist 25+ years ago. Error correction behaviours often involve latency, including bandwidth reductions or retransmission latency or extra wait states or ultrabrief computer freezes, etc. We don't know which error correcting layer occured, but we know computer reliably sped up/slowed down everytime the known electrical interference was added/removed.
Usually error corrections in a PC on these layers are nanoseconds or microseconds but they can apparently daisychain by the thousands into visible milti-millisecond microfreezes covering multiple display refresh cycles! Which just appears as stutter (choppy motion on a screen). It's hard to diagnose exact causes of specific stutters (from driver bugs or thermal throttling or background processing).
But apparently, this computer slowdown was EMI-caused -- visible stutter from electricity noise coming from the power outlet!
There's a ton of weakly-shielded cheap electronics or even shielding-defective high end electronics. And time can degrade many device like these TVs degrading into unintentional small-scale EMPs.
Very weird!
_____
--end of crosspost; begin of reply--
WARNING / CAVEAT: This is usually a wild goose to a red herring more than 99% of the time. However, it's not out of the realm of possibility for strong-enough EMI to start "thousand-microfreezes" lagging an insufficiently-shielded PC. The type of "EMI strength + shielding weakness" combo that is strong enough to almost, but not quite, crash a PC. I've seen at least a few examples in the last few years. TL;DR: EMI can cause computer lag as described, BUT it's very rare. It definitely ain't a five-sigma "No", just rare. Unless you have the time and absolutey exhausted other avenues, do not waste time troubleshooting longshots; troubleshoot the low-lying apples first!
