Strobed Motion Blur Reduction Is Same As PWM (but good kind of PWM)

[Chief Blur Buster]

I have a question about this (please feel free to delete or move if appropriate).

The new iPhone 14 is measured by some reviewers as having a 60Hz PWM rate with a 60Hz refresh rate. When those reviews were published, PWM sensitive users were furious that Apple went with such a low PWM refresh rate.

If I understand this correctly, this should actually be much better for PWM sensitive users as opposed to the non-synced rates of previous iPhones.

Interesting! The decision of framerate=Hz PWM is a tough decision, which is a pick-poison between PWM flicker and motion blur. Some people are more sensitive to motion blur, and some people are more sensitive to PWM.

Alternatively, it might be a confusion with the OLED brief black flicker (less than 5% refreshtime), which has no discernable effect in motionblur.

Are you able to tell me if there's some motion blur benchmarks with the iPhone 14? I have not been paying attention to iPhone 14's motion blur performance at various brightness settings, although I have been itching to buy an iPhone 14 Plus later this year.

[du57in]

I have a question about this (please feel free to delete or move if appropriate).

The new iPhone 14 is measured by some reviewers as having a 60Hz PWM rate with a 60Hz refresh rate. When those reviews were published, PWM sensitive users were furious that Apple went with such a low PWM refresh rate.

If I understand this correctly, this should actually be much better for PWM sensitive users as opposed to the non-synced rates of previous iPhones.

Interesting! The decision of framerate=Hz PWM is a tough decision, which is a pick-poison between PWM flicker and motion blur. Some people are more sensitive to motion blur, and some people are more sensitive to PWM.

Alternatively, it might be a confusion with the OLED brief black flicker (less than 5% refreshtime), which has no discernable effect in motionblur.

Are you able to tell me if there's some motion blur benchmarks with the iPhone 14? I have not been paying attention to iPhone 14's motion blur performance at various brightness settings, although I have been itching to buy an iPhone 14 Plus later this year.

The two best, which appear lacking, are probably these two:

https://www.notebookcheck.net/Apple-iPh ... 760.0.html

https://www.dxomark.com/apple-iphone-14-display-test/

[Chief Blur Buster]

The two best, which appear lacking, are probably these two:

https://www.notebookcheck.net/Apple-iPh ... 760.0.html

https://www.dxomark.com/apple-iphone-14-display-test/

Thank you!

I guess I will find out when I get an iPhone 14 Plus later this year to upgrade my old phone.

[blazerd123]

Yes, you will get 16.67 ms of blur (at 60 fp)s by strobing at extremely high rate since multiple images will eventually blend into the motion blur of sample and hold display. At 60fps@6000hz this would happen for movement speeds below 6000 pixels/s since the movement between strobes would be less than one pixel.

Look at these simulations for movement speed of 32 pixels/frame:

1 flash per frame = 60fps@60Hz strobed
motion_blur1.png

2 flashes per frame = 60fps@120Hz strobed
motion_blur2.png

4 flashes per frame = 60fps@240Hz strobed
motion_blur3.png

So what would the MPRT of an 120hz OLED device with a 120hz pwm frequency played at 120fps, 8ms or 4ms?

[Chief Blur Buster]

So what would the MPRT of an 120hz OLED device with a 120hz pwm frequency played at 120fps, 8ms or 4ms?

MPRT = pulsetime. Any Hz.

Demo: www.testufo.com/blackframes

PWM has infinite different pulsetimes (e.g. 50%:50%), but most OLED BFI uses 50%:50%.

That assumes MPRT(0%->100%) instead of MPRT(10->90%) and GtG is practically zero like OLED (so GtG doesn't distort MPRT)

[blazerd123]

I initially wanted to get a zotac zone with a 120hz pwm (testiest by notebookcheck). They didn’t go into detail but I’m pretty sure it’s off for the same period it’s on.

an lg c1 at 120hz monitor refresh with bfi (50%) enabled gives 4ms mprt.

Unless I’m misunderstanding the zotac zone natively is accomplishing what the C1 is doing so I assumed it was 4ms mprt

Unless again I’m misunderstanding

[kyube]

So what would the MPRT of an 120hz OLED device with a 120hz pwm frequency played at 120fps, 8ms or 4ms?

I initially wanted to get a zotac zone with a 120hz pwm (testiest by notebookcheck). They didn’t go into detail but I’m pretty sure it’s off for the same period it’s on.
an lg c1 at 120hz monitor refresh with bfi (50%) enabled gives 4ms mprt.
Unless I’m misunderstanding the zotac zone natively is accomplishing what the C1 is doing so I assumed it was 4ms mprt
Unless again I’m misunderstanding

I believe there's a misunderstanding at hand here.
The Zotac ZONE doesn't have proper PWM dimming (be it multi-strobe or single-strobe).
This is how actual global scan, single-strobe PWM ("backlight strobing" if LCDs or "BFI" if OLED) looks like
To be specific, this is how the waveform of a fixed duty cycle implementation looks.

It's suffering from the display scan-out related brightness dip (which all OLEDs suffer from) related to the internal capacitance of the OLED voltage driver. I've also seen it referred to as "TFT Reset" by a few chinese reviewers.
Perhaps these visualizations I've provided might give you a better idea on what's happening behind the back.

I haven't seen a single handheld employ backlight strobing (LCD) or HW BFI (OLED) yet...
I haven't seen any <2ms MPRT TVs as well

The LG C1 TV can supposedly get ~2,778ms MPRT (~360Hz-like eye-tracked clarity) at 120Hz, as per rough visual estimates by a member who owns the TV.

[blazerd123]

The display backlight flickers at 120 Hz (worst case, e.g., utilizing PWM) Flickering detected at a brightness setting of 100 % and below. There should be no flickering or PWM above this brightness setting.

The frequency of 120 Hz is very low, so the flickering may cause eyestrain and headaches after extended use.

In comparison: 53 % of all tested devices do not use PWM to dim the display. If PWM was detected, an average of 8118 (minimum: 5 - maximum: 343500) Hz was measured.

This is directly taken from the notebook check article on the zotac zone. I’m a newb so bare with me. It sounds like it does pwm dimming that remains at 120hz frequency from 100% brightness to the minimum of the display. Are you saying that it’s not well implemented because it’s not on a fixed duty cycle? I’m confused on that

https://www.reddit.com/r/SteamDeck/s/uVi6VlOEB2

This forum post suggest that the pwm dimming at the indicated frequency (120hz) implemented in the zone would offer better motion clarity for the same reasons it does on the steam deck

[kyube]

The display backlight flickers at 120 Hz (worst case, e.g., utilizing PWM) Flickering detected at a brightness setting of 100 % and below. There should be no flickering or PWM above this brightness setting.

The frequency of 120 Hz is very low, so the flickering may cause eyestrain and headaches after extended use.

In comparison: 53 % of all tested devices do not use PWM to dim the display. If PWM was detected, an average of 8118 (minimum: 5 - maximum: 343500) Hz was measured.

This is directly taken from the notebook check article on the zotac zone. I’m a newb so bare with me. It sounds like it does pwm dimming that remains at 120hz frequency from 100% brightness to the minimum of the display. Are you saying that it’s not well implemented because it’s not on a fixed duty cycle? I’m confused on that

https://www.reddit.com/r/SteamDeck/s/uVi6VlOEB2

This forum post suggest that the pwm dimming at the indicated frequency (120hz) implemented in the zone would offer better motion clarity for the same reasons it does on the steam deck

Yes, but take a look at this part in the Zotac Zone review:

Secondary Frequency: 240 Hz

There are 2 light flickers occurring:
• Brightness-related dimming method (the 240Hz portion), a full 0-100% (on-off) flicker
• Scan-out related one (the 120Hz portion),a ~12% brightness dip every refresh cycle (60hz = 16,67ms; 120hz = 8,33ms...)

Together, they're a form of multi-strobe PWM, which is counter-productive in reducing eye-tracked motion quality.
That post is a bit misleading IMO... Re-reading this thread from the beginning or this part might make it easier for you.

Your goal is to have the actual on-off pulse (used to hide the persistence blur) be the same frequency as your refresh rate (also referred to as vertical frequency) to achieve a single-strobe PWM scenario.

[blazerd123]

Thank you for explaining that. I didn’t see the relevance of the secondary frequency