PWM flickering vs. refresh type display flickering

[xykreilon]

Hi, I have a question regarding flickering.
Why is PWM flickering noticeable (either visually or through latent issues like nausea, headache, etc.) when less than or equal to around 500Hz, despite flickering from the refresh rate of "refresh type" displays (CRT, strobbing LCD/OLED, etc.) being hard to perceive and/or be strained from past about a 75Hz refresh rate?

 

[majiannoodle]

Random guess... Maybe LED PWM dark and bright difference is very large and fast and much more noticeable? CCFL and maybe CRT has a lot of residual light when it is turned off so the difference between bright and dark is less than LED PWM. CRT and CCFL PWM does give people headache.

 

[XoR_]

I have hard time believing you can notice 500Hz flickering.
Even W-LED 175Hz flickering which can be seen by effect it has on motion is not really seen as flickering

 

[sharknice]

Hi, I have a question regarding flickering.
Why is PWM flickering noticeable (either visually or through latent issues like nausea, headache, etc.) when less than or equal to around 500Hz, despite flickering from the refresh rate of "refresh type" displays (CRT, strobbing LCD/OLED, etc.) being hard to perceive and/or be strained from past about a 75Hz refresh rate?

Strobing and scanning can have visual benefits such as improved motion clarity whereas pwm only has drawbacks, but I don't think it's normal to be affected by pwm flickering and not the other flickering. Often when people blame PWM on headaches etc, PWM isn't the true cause.

 

[xykreilon]

Random guess... Maybe LED PWM dark and bright difference is very large and fast and much more noticeable? CCFL and maybe CRT has a lot of residual light when it is turned off so the difference between bright and dark is less than LED PWM.

That's something to think about- LEDs definitely have substantially less residual light.

CRT and CCFL PWM does give people headache.

CCFLs and CRTs don't use PWM. PWM stands for pulse width modulation, and is used for a large assortment of things in electronics. But, for displays, it's used for dimming the brightness when it can no longer be dimmed by just lowering the voltage without causing distortion. Many desktop LCDs don't need PWM at all, and mobile/laptop LCDs only need it at very low brightness levels- usually half brightness at most.
However, OLED screens currently need PWM for even the slightest bit of dimming because of the inhibiting technical difficulties in lowering the voltage going to each individual pixel.

Strobing and scanning can have visual benefits such as improved motion clarity whereas pwm only has drawbacks, but I don't think it's normal to be affected by pwm flickering and not the other flickering. Often when people blame PWM on headaches etc, PWM isn't the true cause.

Well, I have some anecdotal evidence in favor of AMOLED PWM being worse than CRT flickering in regards to fatigue, soreness, etc. I've been using AMOLED screened phones since 2014- first with a Nokia Lumia Icon and then with a Galaxy S8. Now, frankly, I don't recall much fatigue or soreness from using the Lumia at all. This could be because its PWM flickering had a particularly high refresh rate, or because the eye-soreness I get from the S8 isn't due to PWM at all like you suggest. When I first noticed the eye-soreness I got from the S8, I just thought it was a consequence of general smartphone use and moved on. However, later down the road when I switched to a Nexus 5X with an LCD display, I noticed my eyes getting a lot less sore for some reason. Not really thinking about why that might be and just thinking the difference is likely in my head. However, when the Nexus 5X's CPU cores split apart (sad violin noises) and bricked the device to the moon and back, I was forced to switch back to the S8. I very quickly noticed my eyes getting sore quicker despite the display being just as dim and extended use of a blue light filter.
Again, this was all before I knew anything about PWM. Considering that, measured by notebookchecker, the PWM flickering of my S8 is at a relatively very low rate of 200-250Hz, it seemed to be a reasonable explanation for the fatigue.
However, I'd also been using a CRT monitor at a minimum refresh rate of 76Hz considerably, and did not have any eye-soreness issues over my previous LCD. In fact, I'm able to use it comfortably for longer than the LCD- likely for reasons outlined here: https://mike.pub/19960719-lcd
Anyway, I'd be really interested in a blind test that delves into this. With more strobbing displays on the horizon, it certainly wouldn't only be relevant to prehistoric CRT users like me.

 

[majiannoodle]

I think PWM is started before LED backlight and start with CCFL backlight. I have a Panasonic TV from 2008? and a Dell 20" from 2004. If I use camera to take pics of them with increasing shutter speed, Panasonic TV has partially darken image (not completely black) from PWM while Dell does not. Here is link to tftcentral:
https://www.tftcentral.co.uk/articles/pulse_width_modulation.htm

 

[jeremyshaw]

I think PWM is started before LED backlight and start with CCFL backlight. I have a Panasonic TV from 2008? and a Dell 20" from 2004. If I use camera to take pics of them with increasing shutter speed, Panasonic TV has partially darken image (not completely black) from PWM while Dell does not. Here is link to tftcentral:
https://www.tftcentral.co.uk/articles/pulse_width_modulation.htm

PWM definitely did start earlier than LED, though most light sources also had quite a bit of persistence, whereas LEDs can go from 0-100-0-100 quite fast. It definitely made the strobing effect more powerful, though obvious the impact on individuals is quite varied.

 

[XoR_]

CCFLs and CRTs don't use PWM. PWM stands for pulse width modulation, and is used for a large assortment of things in electronics. But, for displays, it's used for dimming the brightness when it can no longer be dimmed by just lowering the voltage without causing distortion. Many desktop LCDs don't need PWM at all, and mobile/laptop LCDs only need it at very low brightness levels- usually half brightness at most.

I do not thing there were any CCFL LCD's which did not use PWM. I certainly do not remember seeing any which didn't.
It is from electronic of these displays why many later W-LED LCD's used 175Hz PWM as this value was optimal for CCFL lamps.

Later when people started reporting issues with flickering W-LED screens manufacturers started either switching to controlling with voltage and/or very fast PWM. Though I am actually skeptical about this voltage control. You can switch electronic components such a W-LEDs on and off very very fast, even in MHz range far beyond capabilities of any measurement devices and it will be better for W-LED color response and cheaper than using real voltage control. If I was creating voltage control unit I would make it like that.

 

[xykreilon]

I do not thing there were any CCFL LCD's which did not use PWM. I certainly do not remember seeing any which didn't.
It is from electronic of these displays why many later W-LED LCD's used 175Hz PWM as this value was optimal for CCFL lamps.

Ah, for some reason I thought CCFL flickering was just inherent to the bulb for other reasons and not due to always-on PWM. Guess I missed that.

Later when people started reporting issues with flickering W-LED screens manufacturers started either switching to controlling with voltage and/or very fast PWM. Though I am actually skeptical about this voltage control. You can switch electronic components such a W-LEDs on and off very very fast, even in MHz range far beyond capabilities of any measurement devices and it will be better for W-LED color response and cheaper than using real voltage control. If I was creating voltage control unit I would make it like that.

Well, PWM seems to be commonly in frequencies low enough to cause various issues. If it was consistently above 1000Hz, I wouldn't heed it too much. But, that's not the case- even sub 300Hz isn't entirely rare. So, I do.