35 ms·
If I get that right the test is a single blinking led. I think it's much harder than detecting a flickering motion. I can definitely tell 30 vs 60 vs 120fps wh
by bombela 3y ago
If I get that right the test is a single blinking led. I think it's much harder than detecting a flickering motion.
I can definitely tell 30 vs 60 vs 120fps when moving a mouse cursor for example.
For flickering LED light bulbs (120/100hz) I usually notice during a rapid head movement that something is off. Or that the light is blindly bright to look at while at the same time what it illuminates doesn't feel as bright as it should be. I learned to assume that it means the light is flickering, because I can easily use my phone to record a slow motion video for confirmation.
- zamalek 3y agoYup, flicker fusion is for the bare minimum. My second-hand knowledge is that we can reliably identify a single flash at roughly 1/900s. We are even subtly conscious of single photons. What I can say for certain is that the difference 75FPS and 60FPS is the difference between enjoying VR and leaving your lunch on the floor.
- IAmGraydon 3y agoIt’s not testing flicker fusion threshold unless it’s activating the exact same spot on the retina each time it flickers. When you move, the spots are distributed along a path across your retina.
- snailmailman 3y agoI hate those light bulbs. They blink at I think 120hz? Idk maybe that saves energy as half the time the light isn’t on. And maybe most people can’t tell. As an object moves in sunlight it looks smooth to your eyes, but under a blinking light you only see the motion during moments that get lit up. This is much more obvious for faster motion. I’m not 100% certain but I think this sort of lighting triggers bad headaches for me. I get bad headaches some times, and I’ve always been pretty sure the lighting was why but I’ve only recently started confirming with my phone that all those places where I get headaches use blinking overhead lights for the entire room. More and more places around me have adopted it and it’s awful. I have tons of clips in my phone now of random locations in slow motion, many of which have these lights. I can’t walk through a store that’s using them for very long.
- smodo 3y agoVery possible. I’ve suffered a brain injury that has made me more aware of this kind of discontinuous input. It is definitely more taxing for the brain. LED lights with poor drivers are a rudeness to me now.
- bombela 3y ago120Hz US, 100Hz EU. It flickers at twice the main's frequency. It is because the LED drive/power supply electronics is too cheap to have enough capacitance to maintain the LED powered while the alternating voltage is crossing zero. I used to get headache from flickering CFL lights in store when I was a kid. Those days only the worst LEDs make me uncomfortable. Some LED bulb will flicker with a really sharp cutoff, they are the worst. Many will flicker, but not down to zero light. This does make a difference.
- aimor 3y agoYes, I think that the light being stationary is an important distinction, though it's a good thing to control for because motion complicates things. I came to the comments here because of a flickering effect (different from what they're testing for) I often observe: the tail lights of cars. Sometimes when I move my eyes (not my head) very quickly I will see a dashed trail from the red tail lights of certain cars (I'm assuming they're pulse modulating for some reason). I've been thinking about this in terms of the often-cited phenomena where our brains 'black out' our vision briefly when moving our eyes. I think this (brain 'blacking out' our vision) is not what happens, and some others have similar evidence: https://www.science.org/doi/10.1126/sciadv.abf2218 https://www.science.org/doi/10.1126/sciadv.abf2218
- bombela 3y agoI don't think I see a dashed trail, but I will pay more attention. I mostly find that many modern car LED lights are flickering very obnoxiously. And it is worse during motion.