Just had a chance to check PWM on the MBP M4 14". Here are results: 1. PWM is 15 kHz. 2. PWM appears on 9 bars of brightness. It looks like DC dimming is used in the range of 0-8 bars. 3. PWM modulation is about 50% in the full range of 9+ bars of brightness. 4. PWM magnitude very depends of the colour. Each colour has its own PWM pattern. Worst PWM is for the white colour. So moving photodiode over the screen which displays different colours changes PWM visualization on the oscilloscope.

MacBook Pro M4 PWM

NewDwarf

Just had a chance to check PWM on the MBP M4 14".

Here are results:

PWM is 15 kHz.
PWM appears on 9 bars of brightness. It looks like DC dimming is used in the range of 0-8 bars.
PWM modulation is about 50% in the full range of 9+ bars of brightness.
PWM magnitude very depends of the colour. Each colour has its own PWM pattern. Worst PWM is for the white colour. So moving photodiode over the screen which displays different colours changes PWM visualization on the oscilloscope.

AshX

Curious if you were able to detect the gray color flicker or a form of FRC on the M4 Pro. It’s fascinating that the color changes the PWM pattern, which makes me think dithering is interacting with PWM and could be a good or bad thing.

I’m also curious if power saving settings affect it when turned on and off. Did you test with True Tone, Night Shift, and Auto Brightness, and if so, any notable effects?

Donux

NewDwarf This does not sound bad at all from pure technical point of view. In store views also seam to suggest to me that Air models have sharp sensation in the yes, some Pro models do not. Of course it could be placebo or batch variences with Pros.

Also, as I have tested on my external BENQ PD2705Q and now on Asus Zenscreen, and per some tests performed on this forum (Intel vs apple silicon flicker redzones), it is absolutely clear, Air models send "dirty signal" to internal and external panels.

NewDwarf

AshX Gray colour has minimal PWM. This happens because of the microLED nature where each pixel is driven independently. So, I expect these results.

Power specific settings, True Tone, Auto Brightness don't affect PWM.

AshX

NewDwarf

Very interesting. This seems to tentatively confirm the gray color flicker is LCD specific and MiniLED and OLED don’t seem to utilize it (please correct me if I’m wrong). I wasn’t aware MiniLED had the same level of per-pixel control as OLED - I assumed control in the former was relegated to larger “zones” as opposed to the individual pixels.

So my follow-up question would be what is your educated guess as to the implementation of dithering (either GPU or FRC)? Is it more difficult to detect on these MiniLED 8-bit+FRC screens? Does the PWM mask dithering, or is it that interaction that changes the PWM pattern?

By DC dimming up to 8 bars, wouldn’t that traditionally mean there is no PWM/flicker? I’m also assuming that the 50% modulation depth is what makes many of us feel dizzy when looking at these screens?

Lauda89

NewDwarf No, macbook pros are miniLED not microLED. So it is a simple IPS screen backlit by multiple zones of LEDs.
It is OLED and microLED (almost only prototypes at the moment) that have independent management for each pixel.

NewDwarf

AshX It is difficult to say how large is the specific area controlled individually from the backlight perspective. Reading "quantum dot" I would think it is pixel level backlight control.

But, anyway, the area is too small as there is a sharp change of the PWM pattern on moving the photodiode over the colour bar table.

AshX By DC dimming up to 8 bars, wouldn’t that traditionally mean there is no PWM/flicker?

At least, there is no PWM flickering and other low frequency fluctuations.

AshX I’m also assuming that the 50% modulation depth is what makes many of us feel dizzy when looking at these screens?

Not sure, as 15 kHz is pretty high frequency. Moreover I cannot use some M-based Apple's laptops with zero PWM and disabled dithering. Obviously, there is something which causes eye strain. To figure out what is this, I guess, 180x magnification and 1000 FPS recoding is required. I have no such equipment.

AshX

Donux per some tests performed on this forum (Intel vs apple silicon flicker redzones), it is absolutely clear, Air models send "dirty signal" to internal and external panels

I noticed this on the M4 Air. Thought I was imagining things.

NewDwarf

Donux

Donux it is absolutely clear, Air models send "dirty signal" to internal and external panels.

Air M2/M3 have weird 60 Hz fluctuation. I think it is related somehow to the GPU frame rate but I am not sure.

Donux

I personally will buy buying pro model at some point, but probably will manage risk in this fashion:

Buy external measuring device (accuretely enough)
Buy display unit (probably 1 generation lagging) and in store test for 1-2 hours

If someone could throw a link to Amazon, ebay, alibaba or elsewhere, for decent but budget measuring device they have personally used and tested I would appreciate.

And 15 kHz I would accept as normal. It is really high. And realistically, if you are sitting in the cafe or other public space, it probably has a lot of dodgy lighting here and there which partly overshaddows little nitty gritty perfections of a screen.

Donux

NewDwarf I keep thinking maybe its a depletion of some chemical in the brain. For example there are EMF sensitive people who say if they get triggered, they get magnessium and it solves it immediately. On another hand last time I checked with ChatGPT, there was no obvious link between eye strain and daily high frequency EMF dose. Maybe, due to prolonged work eyes get so sensitive, that it starts to pickup every single noise imperfection.

Donux

NewDwarf Something is off, the question is how much agency there is at our end to find a solution. Apple would never themselves admit mistakes or issue a recall, this would be damaging to the brand.

AshX

NewDwarf Air M2/M3 have weird 60 Hz fluctuation. I think it is related somehow to the GPU frame rate but I am not sure.

It’s probably FRC. I don’t think the Airs are capable of sending a pure signal like the Mac Mini can to an external display. It’s going to apply the processing from the internal display unless you can disable the internal display’s FRC, which would require disabling it within the TCON, and no one has been able to figure out what software commands would disable it like Stillcolor has with the GPU.

My best guess is Apple wants to maintain consistency between the internal display and the external display so it will force this FRC into the signal. If it’s not that, then it’s ColorSync. But it’s probably likely ColorSync is playing a role in everything and I highly doubt anyone could easily retrofit a solution like was possible in the past using RGB color profiles.

One could prove or disprove both our theories by using different forms of each chipset: M2, M2 Pro, M2 Ultra, M2 Max for example (I don’t know it that exists for that generation; I know the new Mac Studio has an M3 Ultra and M4 Max option, not an M4 Ultra Option).

It’s anecdotal but I’ve noticed the M3 Ultra Mac Studio pushes a cleaner display output than the M4 Max. I would be surprised if lower power mobile devices like the Air don’t utilize power saving tricks that introduce more noise into the signal than desktop devices that have active cooling and a powerful chipset with more GPU cores. It only seems logical to me given how much of an importance Apple places on battery life these days.

Botvinic

Hello! Recently I bought a MacBook Pro m4 Pro and was horrified. After turning off the dithering, I still felt strain in my eyes and did not understand what was happening. I decided to check through a camera with high shutter speed and found out that there is a pwm in the mini led. What a surprise. And it can't be turned off. I had to switch to m3 air, because m4 couldn't turn off dithering. Thus, the best poppy at the moment without dithering and without shim is air m3 13 or 15. We are waiting for new technologies))

Donux

Botvinic Well if you can pick it up with consumer camera, then it is bad for sure. PWM is not a big deal if it is 15000 Hz, but it should not show anything on your camera I think. How is external monitor, any better ?

NewDwarf

Botvinic MacBook Pro M4 is much better for me than MacBook Air's.

As @Donux said, 15 kHz is safe from the PWM perspective.

Donux

Botvinic Hi, are you still suffering from Macbook pro m4 eye strain when using it without tinkering with settings apple did not create themselves ?

Botvinic

External monitors are flickering, I can't find it yet, all my eyes hurt.

PWM m2 pro causes more pain for me than dithering on AIR m3. It is interesting that only one type of dithering is disabled there gstillvolor). The flickering of pixels still remains, I checked through a microscope.

Donux

Botvinic Remove all those crazy workarounds and keep it standard. Make sure external monitor gets lower grade color mode, as usually they leave on standard high FRC mode. Inspect color profile that macbook generated for you, to see what is the gamut. If it is lower then sRGB, keep it, as it probably reflects real 6 bit capabilities. Otherwise change to Generic RGB or sRGB. Generic RGB usually is better, and it is pretty muck close to sRGB, just Apple's own version of it.

Update:
Well at least you can keep that dithering off that is already implemented, but uniformity I would skip.