Yes/No Tool for PWM and Dithering

Dizzy

Is there a tool that can report a simple Yes/No if there is PWM and Dithering (maybe other detection?). It’s really hard to tell by following a scrolling line test to know if PWM is present or what could be causing symptoms.

moonpie

No.

Desktop monitor - PWM is easy. Dithering requires a microscope.

Laptops are even harder. You will have to physically just try them out.

Dizzy

moonpie Can you explain how on the desktop?

moonpie

Dizzy
An intel or amd x86 desktop hooked up to a monitor:

Smartphone camera on manual shutter and you can see black or grey bars. (up to 1khz)
Radex Lupin or Opple Light Master (up to 5-10khz)
Scope with a photodiode (up to however much you want to spend)

+
microscope for dithering test

You are better off just buying the exact make, model, and year produced/tested by others. e.g. If rtings.com tests the HP BZHd8he gaming monitor in 2024 and says it is okay, you run out in 2024 and buy the HP BZHd8he gaming monitor. Business/office models have a greater chance of keeping the same panel throughout the production.

Note: Laptop and any external screen attached to a laptop is completely hit or miss. Any ARM based "desktop" can be as well which means Mac Mini MX devices.

Dizzy

moonpie Thank you

jordan

You're best off getting a opple light master iv which will show what frequency/waveform for the display your testing. It makes it easy to check for PWM. Does not matter if its a laptop, desktop or a phone. Dithering is harder to check for. I know some people have done gradient tests which if you see banding then usually less likely its dithering. The only true method is getting a cheap microscope which many seem to use the carson microflip x250 one on amazon. Using that with a 240fps slo mo camera which can be found on certain phones. There is no simple yes no tester for dithering sadly.

Dizzy

jordan Thank you

moonpie

jordan 😕 You sure those are good for laptops? Don't they quit at 10kHz and only show results in ms?

FNIRSI DSO152 is $36 at amazon and does 200kHz. A $0.10 LED or $5 photodiode is all you need to add to that.

jordan

moonpie Nah I've seen them read 20kHz even. I forgot what they max out at. Someone also posted a modified version of the opple app which is a lot faster too.

moonpie

jordan Provide a link showing that it can read up to 20kHz. That's still bad for dealing with laptops.

Running a Chinese app is bad enough, but running a modiied Chinese APK is even worse. This makes the $36 FNIRSI DSO152 a superior solution which can do 200kHz without that.

Edit:
Radex Lupin maxes at 6kHz. IIRC, I used their software to extend that to 8-10kHz:
https://ledstrain.org/d/776-radex-lupin-are-6000-hz-enough

jordan

moonpie https://www.reddit.com/r/PWM_Sensitive/s/1LFh9JIiLL

It's just good for quick portable tests. You can easily tell if one is dc or not. IDC using a modified version of the app personally. It's posted here if anyone is curious.

https://www.reddit.com/r/flashlight/comments/1bt9it5/i_modified_opple_light_master_apps_for_android_apk/

I think the other alternatives are better but not portable. Be careful with the lupin because I bought one and the reading was so off. I mailed it to my friend and he tested his against mine and it was definitely off. I wonder how many other ones are too 🤔

moonpie

jordan Do you have a link where someone with a real scope actually showed it to be accurate? My guess is not at all considering their own graphic stops at 10kHz and the LM4's main purpose is for spectrum not flickering.

"UPDATE: After using this laptop for 1hr, I'm having slight nausea and my back of the head is very tired, probably because of the dithering and constant eye focus on these flickering pixels 🙁"

Looks not good via scope below 100%:
https://laptopmedia.com/review/apple-macbook-air-15-m3-review-big-screen-power-in-a-thin-and-light-package/#p3

I bought the Lupin and with PC software it could do almost 10Khz. I returned it because using PC software made it not portable, a cheap soundcard could do the same thing, and a smartphone camera on manual was faster to confirm such low PWM.

jordan

moonpie I do not unfortunately. I know alot of people use it in the flashlight subreddits but some also say the light master 3 is more accurate 🤔

I know M silicon MacBooks dither bad so any discomfort would definitely be from that

moonpie

jordan Then I would not tell people it is a "yes/no" for laptops. The manufacturer is very clear that 10kHz is the max they trust and you shouldn't assume anything beyond that is accurate until tested with a scope.

laptopmedia.com is showing it has some kind of low amplitude PWM or VRR active. This isn't dithering:

vs

jordan

moonpie it's still usable on laptops below that. Especially OLED ones. I mean I don't know how accurate it is fully but it's good enough to get an idea. Several people use it. Obviously there are better options but that's the best option for the money in terms of portable meters. If you want portable you can spend a lot more and get a uprtek mk350n premium but I think it caps at 50kHz. The uprtek mk550T is designed for measuring screens and I think it may be capable of higher frequencies.

macsforme

I would personally not trust the Opple LightMaster frequency reading unless you can physically see it on the graph to verify it. I recall seeing some things that made me question it (possibly, wildly inconsistent readings in the upper ranges). I understand that the graph purports to reach 10KHz, but I question whether the device can detect such high frequencies with accuracy.

moonpie

jordan For laptops you need something higher than 10kHz and accurate.

Any cheap scope will do. $36 FNIRSI DSO152 will do 200kHz. There is absolutely no reason to buy the ridiculously expensive products you have posted when a picoscope is $100-$200 and overkill at 10MHz.

jordan

moonpie yeah those others are great just not portable. I guess with the uprtek you get other testing stuff such as measuring SVM, color spectrum, and whatever else they are capable of but yeah way pricey

moonpie

jordan DSO152 literally fits in your pocket. The problem isn't portability but the devices being used are incapable of of going beyond 10kHz and giving people a "yes" answer when it is actually a "no."

What "other stuff" has an effect on eye strain? PWM and dithering are the only objective things anyone can agree on bothers them. Everything else is just a handful of people who think those issues could be a problem. Manufacturers already don't care about eyestrain because we are a minority, so they will never care to fix an issue affecting a minority of a minority.

jordan

moonpie your right that is very tiny. I don't think anyone believes the safe / not safe output on the opple as it'll show a device not safe when it is if it's dc like dimming/OLED for example. I just use it to get an idea for devices under 10kHz, mostly seeing waveform. Do those mini oscilloscopes have any noise that effects the measurement? They're cheap that's why I ask.

Other stuff as in different types of light measurements. I didn't say other things that cause eye strain. Speaking of that I would say color spectrum is definitely one other issue to some. We know certain phosphors on led backlit panels cause eye strain to some people which is easily detectable with a spectrometer. I know it affects me because the orange backlight on the daylight tablet caused me issues and it uses a specific phosphor (I forgot the name)

moonpie

jordan

OLED flickers and should never be tagged as "safe." The OP wants a "yes/no" tool not a "get an idea" tool. That is why I said you shouldn't use them on laptops.

Cheap =/= bad. You can find soundcards in the trash for free that can be used to accurately detect PWM over 10KHz. No clue about phosphors. The only two objective yes/no tests are for PWM and dithering that seems to address the vast majority of issues here. The rest seem to be ignorance of settings/configurations.