Isn't QD-OLED perfect for backlight strobing?

High Hz on OLED produce excellent strobeless motion blur reduction with fast GtG pixel response. It is easier to tell apart 60Hz vs 120Hz vs 240Hz on OLED than LCD, and more visible to mainstream. Includes WOLED and QD-OLED displays.
Chamber
Posts: 9
Joined: 08 Jul 2022, 23:17

Isn't QD-OLED perfect for backlight strobing?

Post by Chamber » 13 Jul 2022, 02:44

I've been reading a lot about backlight strobing and learned that OLED's strobing capabilities are mainly held back by it's low brightness and this is because of Talbot Plateu's law. Talbot Plateu's law from what I understand means that you loose brightness proportional to gains in motion clarity (reduction in MPRT) while backlight strobing.

The typical RGB AND WRGB OLED displays tend to struggle with brightness, so you either end up with disappointingly high MPRT while strobing or an image that's simply too dim to be usable. But QD-OLED's impressive brightness should be able to overcome this limitation while being less prone to burn in, as its emitters require less voltage to drive the same level of brightness.

Currently the LG C2 pushes 178 nits sustained brightness on SDR with 100% white window while the AW3423DW on HDR 1000 can push 290 nits sustained brightness on a 100% white window (source: RTINGS). The reason I chose SDR for the C2 and HDR 1000 for the AW3423DW is because those give the highest sustained brightness on 100% white window (best full screen brightness in worst case scenario).

Let's compare that brightness with the ViewSonic XG2431's, which is Blur Busters 2.0 approved. The XG2431 hits 343 nits sustained brightness on a 100% white window. With HDR, the XG2431 hits 499 nits sustained brightness on a 100% white window but iirc, you can't use backlight strobing with HDR on that monitor for some reason. So, the AW3423DW only looses by about 50 nits and I believe that the gap can be closed.

We know that the AW3423DW beats the LG C2 handily in 100% screen brightness and Dell is willing to provide a 3 year warranty for burn in. LG doesn't even offer warranty for burn in on their C series of OLEDs. This suggests that QD-OLED's brightness is being held back and can probably be pushed even further. If you slap a heatsink on the 34 inch QD-OLED panel, I think brightness can be pushed to 400 or even 500+ nits sustained on 100% white window, and maybe with backlight strobing turned on, more brightness can be pushed as I think backlight strobing should reduce the wear of the OLED pixels as they'll be off more often giving them more time to rest (correct me if I'm wrong on this).

But what would the advantages of strobing a QD-OLED display over LCD's? One is the faster GtG time, I've read that in order to make the ViewSonic XG2431's crosstalk almost disappear, you'd have to run it at 120Hz. This is because at lower refresh rates, the GtG pixel response time can be hidden between refresh rate cycles. But QD-OLED has almost instantaneous pixel response times, so there's no need to reduce your refresh rate to hide the GtG pixel response time between refresh cycles. So with the right tuning, it should be able to make crosstalk disappear at its maximum refresh rat.

Speaking of tuning, I believe that a QD-OLED display would require less of it to receive the Blur Busters 2.0 certification. This is because the AW3423DW doesn't have any overdrive settings and as mentioned before, this is thanks to GtG pixel response times not being an issue. According to Rtings the AW3423DW does have a bit of overshoot when going from pure black from another shade, but I think that can be resolved with tuning.

My dream monitor would be one that utilizes Samsung display's 34 inch QD-OLED panel from a brand like ViewSonic or BenQ which supports the BlurBusters strobing utility and receives the Blurbusters 2.0 certification.

I'd consider myself a noob when it comes to my understanding of backlight strobing so my apologies if I made any mistakes or errors in this post. I wanted to start this discussion as I haven't seen anyone else discuss the strobing potential of QD-OLED displays. I was really disappointed to find out that the AW3423DW doesn't have backlight strobing, so I hope that other monitor companies can consider adding this feature once they get a hold of this panel :D.

User avatar
RealNC
Site Admin
Posts: 3741
Joined: 24 Dec 2013, 18:32
Contact:

Re: Isn't QD-OLED perfect for backlight strobing?

Post by RealNC » 13 Jul 2022, 03:25

I don't think anyone is willing to read that impenetrable wall of text.
SteamGitHubStack Overflow
The views and opinions expressed in my posts are my own and do not necessarily reflect the official policy or position of Blur Busters.

User avatar
jorimt
Posts: 2481
Joined: 04 Nov 2016, 10:44
Location: USA

Re: Isn't QD-OLED perfect for backlight strobing?

Post by jorimt » 13 Jul 2022, 08:32

Chamber wrote:
13 Jul 2022, 02:44
RealNC wrote:
13 Jul 2022, 03:25
I don't think anyone is willing to read that impenetrable wall of text.
I've now added paragraphs ;)
(jorimt: /jor-uhm-tee/)
Author: Blur Busters "G-SYNC 101" Series

Displays: ASUS PG27AQN, LG 48CX VR: Beyond, Quest 3, Reverb G2, Index OS: Windows 11 Pro Case: Fractal Design Torrent PSU: Seasonic PRIME TX-1000 MB: ASUS Z790 Hero CPU: Intel i9-13900k w/Noctua NH-U12A GPU: GIGABYTE RTX 4090 GAMING OC RAM: 32GB G.SKILL Trident Z5 DDR5 6400MHz CL32 SSDs: 2TB WD_BLACK SN850 (OS), 4TB WD_BLACK SN850X (Games) Keyboards: Wooting 60HE, Logitech G915 TKL Mice: Razer Viper Mini SE, Razer Viper 8kHz Sound: Creative Sound Blaster Katana V2 (speakers/amp/DAC), AFUL Performer 8 (IEMs)

Futuretech
Posts: 35
Joined: 11 Oct 2020, 23:52

Re: Isn't QD-OLED perfect for backlight strobing?

Post by Futuretech » 13 Jul 2022, 15:52

There is also PQD or Perskovite Quantum Dot. Combining QD with the Perskovite material.

https://www.perovskite-info.com/ There is also microled-info, OLED-info, and probably going to expand into LET and OLET as well.

Although PQD is perfect for OLED for even more. In the microLED sector it can shrink microLED further. Nanosys downscaled by 83% 35micrometer to 8micrometer back around 2017 or so a microled they were working on.

I wouldn't be surprised if LET and OLET are receiving the PQD treatment as well. Light Emitting Transistors and Organic Light Emitting Transistors.

Still these are augmentations there is still more R&D and more maximization offerings. Still I understand your enthusiasm O.P. but these can sometimes be band-aid issues that might require other technologies available. Sure you have QD OLED monitors with quite low blurring, enough blurring it can be seen but that doesn't mean we can't further bring down the blur. These current OLED panels are more like a taste of what's to come who knows what other material can bring forth.

User avatar
Chief Blur Buster
Site Admin
Posts: 11647
Joined: 05 Dec 2013, 15:44
Location: Toronto / Hamilton, Ontario, Canada
Contact:

Re: Isn't QD-OLED perfect for backlight strobing?

Post by Chief Blur Buster » 13 Jul 2022, 20:34

RealNC wrote:
13 Jul 2022, 03:25
I don't think anyone is willing to read that impenetrable wall of text.
I am...

So this is the actual translation:
Most readers in this forum won't read this in this specific forum area. Chief Blur Buster may move these types of threads to one of the Laboratory forums. Walls of texts about technical monitor/testing are highly encouraged in the Laboratory forums, such as Area 51 or Pursuit Camera, etc
Honestly, I want to encourage more forum posts like these on Blur Busters, because I like to discuss technology matters.

Personally, I encourage Walls of Text in Area51 or other Laboratory forums, and we badly need to recruit more researchers & Walls of Text over there, because some of us (myself, Futuretech, CRTgimp, STL8K, thatoneguy, etc) actually read these walls of text. Those are the forum members that [almost] never posts anywhere except in one of those "Laboratory" forums.

Since over 50% of Blur Busters corporate income is via collaborations with manufacturers/vendors/clients/etc involving very Aera51 topics; the revenues help keep these forums running... To me the General forums is to keep fans happy, while the Blur Busters Special Sauce is Laboratory stuff ( www.blurbusters.com/area51 ). Big Walls Of Text included.

If you look closely at Area 51 Post-Read Counts you notice many read walls of texts at Blur Busters, but they are mainly a totally different audience in Laboratory. I check my web logs too, and many are coming from real users instead of bots, and curiously a high read count from various corporate headquarters (despite low public reply counts).
wow.png
wow.png (75.02 KiB) Viewed 10823 times
This is considered, to me, a medium size wall of text, not nearly as big as the wall of texts found in Laboratory.

The totem pole generally is walls of texts gets posted in a forum further down the screen at https://forums.blurbusters.com such as one of the Laboratory forums.

Small posts about what monitor to buy? General.
Big technical questions about display nuts and bolts? Area51.

But we never discourage walls of text about display nuts and bolts in the Laboratory section of Blur Busters

Blur Busters is a longtime crowdsourced incubator of some display tech ideas, some of which eventually rolled into the works I do (test inventions, collaborations with manufactures, or into a related research paper. Some outlandish ideas, while others are less so. Over 100 research papers have been inspired by Blur Busters readings/ideas, and 25-30+ even cite Blur Busters / TestUFO / Mark Rejhon ... Companies such as Samsung and NVIDIA already cite me, for example.

I think this is more of an Area51 question.

Walls of text like this usually gets moved to either Area51. Done.

I'll answer the question in fuller subsequent post -- sometimes it takes a few days or few weeks, but I usually reply to almost every single Area51 thread (eventually).
Head of Blur Busters - BlurBusters.com | TestUFO.com | Follow @BlurBusters on Twitter

Image
Forum Rules wrote:  1. Rule #1: Be Nice. This is published forum rule #1. Even To Newbies & People You Disagree With!
  2. Please report rule violations If you see a post that violates forum rules, then report the post.
  3. ALWAYS respect indie testers here. See how indies are bootstrapping Blur Busters research!

User avatar
Chief Blur Buster
Site Admin
Posts: 11647
Joined: 05 Dec 2013, 15:44
Location: Toronto / Hamilton, Ontario, Canada
Contact:

Re: Isn't QD-OLED perfect for backlight strobing?

Post by Chief Blur Buster » 13 Jul 2022, 20:55

Chamber wrote:
13 Jul 2022, 02:44
Isn't QD-OLED perfect for backlight strobing?
Short answer: Potentially, with an asterik, depending on how the panel is layered.

Generic Considerations

The neat thing about QD-OLED and QLED (in general) is that it's two stage. The "backlit" terminology is not correct, but we'll accept it for the purpose of answering this question generally. It's 2-layers of lighting control (light emission followed by spectrum conversion). Basically it's a monochrome OLED, lighting up a quantum dot phosphor layer, essentially.

A blue light (or ultraviolet) excites the red/green/(blue) pixels. Theoretically the blue light layer can be turned on/off, as long as the red/green pixel Quantum Dot Phosphors don't continue to glow for long (KSF afterglow effect like www.blurbusters.com/red-phosphor ...) but the good news is QD is very fast-responding.

The key challenge is which layer is the TFT layer, and whether an LCD is also involved. It can be either or both (locally-dimmed OLED-backlit QD LCDs). TFT layers complicate strobing, so if the blue light layer is the TFT layer, it's not easy to strobe. But if the blue light layer is a global illumination layer for the other TFT layer, then it's easy to strobe.

Various Ways of Layering a Screen That Includes An OLED and Includes Quantum-Dots

When QD is involved with OLED, there are many combinations where OLED and LCD can help each other, or can be pure OLED:
  • Global OLED light, TFT-matrix phosphor (OLED-backlit phospor-pixel LCD).
    OLED is just a global light source for an LCD that uses QD-phosphor-subpixels instead of normal color filters. Better color gamut.
    .
  • TFT-matrix OLED light, non-TFT phosphor (Blue-OLED excited non-TFT phosphor pixels).
    OLED blue subpixels lights up phoshors. Fastest GtG but hardest to strobe. No LCD is involved.
    .
  • Locally dimmed OLED light, TFT-matrix phosphor (TFT OLED-backlit FALD for phosphor-pixel LCD).
    OLED is zone-controlled light source for an LCD that utilizes QD-phosphor-subpixels.
    .
  • Etc.
    There are other engineering permutations possible of a hybridized OLED / OLED+LCD
Including unreleased new layering formulations -- the hybridization possible of OLED+LCD working together, makes discussion a little complicated, and requires a deep-dive on how the display is manufactured, and manufacturers sometimes are tight-lipped at the beginning, especially when they're only showing off prototypes.

The problem is that some of the layering diagrams doesn't seem to show it as a possibility, while others do, but there are many ways to layer a two-stage OLED (a "backlight" layer for short-wavelength light, and a "phosphor" layer to create the color gamut).

Samsung has not released detailed engineering specs of their panel and the locations of all of their TFT layers (though there might be a public white paper PDF somewhere that I haven't looked yet at SID.org). Currently, it is my initial my understanding Samsung is a TFT-matrix blue OLED, and if so, that will be potentially difficult to strobe without rolling scan electronics in the OLED drive, since it's not something you can necessarily globally strobe easily, depending on how the transistors operate (TFT OLED).

Simple single-transistor and Darlington-transistor pixel drives are generally not globally strobeable, because you need to row-column addressing several million subpixels to turn off the pixels for strobing. While other OLED panels have an ability to refresh in the dark and a single-wire global illumination voltage applied (some OLEDs do this to allow PWM dimming). There are many ways to design the TFT circuit of a TFT OLED (1 transistor, 2 transistor, or 3+ transistor per subpixel).

The PWM dimming logic can be easily commandeered to function as motion blur reduction strobing (by using 1 PWM pulse per Hz). TFT OLEDs with PWM capability requires a different transistor wiring in the active matrix (TFT layer) to allow pixels to refresh to a color value independently of illumination voltage. It is not known if Samsung's QD-OLED panel is capable of this.


Consideration: Layering Construction Affects Ease Of Strobing

In order from easiest to hardest to strobe:
  • OLED global illumination backlight for an LCD layer (as long as LCD is fast-GtG "1ms" tech);
  • PWM-controlled OLED illumination voltage for a pure TFT OLED that is capable of PWM;
  • Non-PWM-capable OLED with OLED TFT being blue subpixels (because you have to refresh all pixels one at a time, to turn them off, far more complicated than simply turning off a single PWM wire or single backlight wire).
PWM dimming is evil but makes strobing capability far easier.

If the panel is constructed right AND there's a strobeable feature in the blue/ultraviolet layer, this can theoretically produce deliciously good OLED strobing, although (calibrateable) color distortions can occur with the different phosphor decay speeds of the different quantum dots (which might or might not be negligible).

However, with short MPRTs like 0.25ms, even a 0.02ms phosphor decay is up to 10% color distortion because of the pulsewidth:decay ratio that varies between R/G/B.

Consideration: QD Phosphor Decay Differences + Color Shifts During Strobing

QD is a very fast phosphor so this is usually insignificant. However, at least minor color shifts will occur, because this is where microseconds become human visible because of a twist:

Regardless of global or per-pixel strobing, Strobing a phosphor-excitation layer requires custom color calibration for each pulsewidth. If you use 2ms MPRTs the pulsewidth:phosphordecay ratio is protected more and color distortions are much more invisible. For voltage-boosted strobing (same nits), color distortions at 0.2ms MPRT is 10x worse than color distortions from 2ms MPRT when using strobeable phosphor-excitation layers -- Since at 0.2ms MPRT you need to strobe 10x brighter for same Hz, to get as bright as the same 2ms MPRT, but phosphor decay dramatically worsens despite same average screen brightness.

And if Red or Green decays slower, you will have color tinted ghosts during strobing. Probably not as bad as KSF, but the pulsewidths will need to be carefully calibrated, and then color-recalibrated to compensate for the colorshifts of the different phosphordecay speeds of R/G/B, because colorshifts occur if pulsewidth:phosphordecay ratios varies between the color channels.

A 0.01ms phosphor decay divergence in a 1ms strobe -- that's 1% of 1ms. That is bigger than the color tint difference of RGB(128,128,128) and RGB(128,128,130). TEN microseconds, but it's 1% of the photons! Because of tiny ratios involved, this is where microseconds become human visible (a famous example of The Amazing Human Visible Feats of The Millisecond)

Consideration: OLED Brightness Limitation & Bright Strobe Pulses

Talbot-Plateau Law strikes back again with its evil red light saber!

Another major problem with blue OLED is brightness compared to the brightest LCD backlights. Using blue OLED light excitation sources for QD layers (whether it be phosphor-on-TFT-OLED, or an LCD-driven phosphor layer), is that voltage boosting the OLED layer can degrade their lifetimes faster than an LCD backlight, if you want short MPRTs without losing brightness. So you need to use very durable blue OLEDs that are burnin and wear resistant, in order to brightly strobe them without being too dark.

Hope this answers the technical questions of strobing a multilayered screen technology involving OLED in one of its layers...

Long Term Ultrabight HDR OLEDs + Rolling Scan Strobe

Ultrabright HDR-capable OLEDs, with rolling-scan logic, can do a lagless strobe. Far more complex to strobe than global strobing, but fully lagless like a CRT tube. Requires blue/ultraviolet TFT OLED backplanes with individually controllable subpixels, lighting up simple QD phosphors that is just simple inkjet printed paint onto OLED subpixels (one stage of inkjet printing an OLED -- this is currently starting to make cheaper OLEDs possible, as Samsung uses inkjet printing to print the phosphors onto the OLED subpixels).

The 2000-nit OLED backplanes are now becoming possible with custom formulations in some OLED lab screens, so in theory...

The current-carrying backplane can't HDR-bright all pixels simultaneously (windowing) but if the rolling scan window is tight (e.g. 10% of OLED height), you can use the nit headroom of 2000 nit HDR to get 200 nits of high quality strobing -- while only needing to strobe 2000nit for only 10% of pixels at any one time. The HDR-illumination-power windowing design needs to accomodate the needs of strobing. Done this way, this does not overheat the OLED, as 200nit strobing of a 2000-HDR never (A) exceeds the momentary HDR window due to the rolling nature; and (B) average brightness is only, say 200nits (for a 10% rolling scan window). This of course assumes that you can do 2000nits in a 10% HDR window. Alternatively, you could go 5% HDR window, and settle for 100nits. But regardless, strobed will be HDR-less, despite using the HDR nit surge headroom to keep rolling strobes bright.

This is actually easiest/safest done with built-in automatic pixel-turnoff logic built into each TFT subpixel, so you simply program a pixel-off delay, and now the panel is self-strobing with only a single scanout. Some OLED BFI designs are starting to experiment with TFTs that automatically turns off transistors, without needing a separate de-illumination scanout pass.

Also see Custom OLED Rolling Scan Patterns for related discussion about TFT-driven rolling-scan strobing (for CRT-matching lagless strobing since panel scanout can theoretically be in sync with cable scanout).
Head of Blur Busters - BlurBusters.com | TestUFO.com | Follow @BlurBusters on Twitter

Image
Forum Rules wrote:  1. Rule #1: Be Nice. This is published forum rule #1. Even To Newbies & People You Disagree With!
  2. Please report rule violations If you see a post that violates forum rules, then report the post.
  3. ALWAYS respect indie testers here. See how indies are bootstrapping Blur Busters research!

User avatar
RealNC
Site Admin
Posts: 3741
Joined: 24 Dec 2013, 18:32
Contact:

Re: Isn't QD-OLED perfect for backlight strobing?

Post by RealNC » 14 Jul 2022, 02:54

Chief Blur Buster wrote:
13 Jul 2022, 20:34
RealNC wrote:
13 Jul 2022, 03:25
I don't think anyone is willing to read that impenetrable wall of text.
I am...
This is what it looked like before it was edited :P

Image
SteamGitHubStack Overflow
The views and opinions expressed in my posts are my own and do not necessarily reflect the official policy or position of Blur Busters.

User avatar
Chief Blur Buster
Site Admin
Posts: 11647
Joined: 05 Dec 2013, 15:44
Location: Toronto / Hamilton, Ontario, Canada
Contact:

Re: Isn't QD-OLED perfect for backlight strobing?

Post by Chief Blur Buster » 15 Jul 2022, 01:43

RealNC wrote:
14 Jul 2022, 02:54
This is what it looked like before it was edited :P
Really, I didn't see that.

Yeah, I missed the original context of a "wall of text" in formatting parlance.

Nontheless, this question is right up Area51's alley.
Head of Blur Busters - BlurBusters.com | TestUFO.com | Follow @BlurBusters on Twitter

Image
Forum Rules wrote:  1. Rule #1: Be Nice. This is published forum rule #1. Even To Newbies & People You Disagree With!
  2. Please report rule violations If you see a post that violates forum rules, then report the post.
  3. ALWAYS respect indie testers here. See how indies are bootstrapping Blur Busters research!

blurfreeCRTGimp
Posts: 42
Joined: 28 May 2020, 20:36

Re: Isn't QD-OLED perfect for backlight strobing?

Post by blurfreeCRTGimp » 15 Jul 2022, 16:08

We want to prioritize the fastest GTG but also have bright strobe pulses. Couldn't we theoretically coat a blue OLED material on the substrate with a clear yet reflective coating (so the organic material is still self illuminating, and easy to drive) but we could also then augment the brightness with an inorganic blue light source without compromising its self emissive qualities?

Or if coating the OLED with retro reflective coating is problematic augment the panel with artificial blue light brought in on a layer of one of those microlens arrays that exist on some OLED panels?

I was thinking (again, surprise) using a MEMS scanning system (with its low voltage for decent brightness, and less heat) with 3 blue laser diodes running at a very high scanrate of 960hz?

As an example, a MEMS system only adding say 5 watts of power providing only 70+ ansi lumens would be enough to push current OLED to basic LCD strobe nit levels if panel refresh rate was high enough, and you used a MEMS system in concert with a rolling scan.

That way, you get the boost in blue light, it lifts brightness in all QD RGB pixels to strobe at adequate nits, but you don't incur any afterglow, blooming, or persistence penalty from a conventional approach of conversion layers/polarizers phosphor decay times. It also wouldn't have to complicate the way you drive the OLED pixels themselves. Kind of a more mechanical solution rather than more complicated electronics to drive the pixels.

MEMS being inherently a scanning technology presumably shouldn't have the same limitations or circuitry woes of strobing whether a backlight or OLED.

Need it faster? Build it into the mirror, or get a better motor. Need infinite contrast at any one point? simply Shut off the scanning lasers.

Its like adding the bright dot of the electron gun sweeping the panel, but with blue lasers and mirror augmenting a reflective blue organic emitter.

Might also be able to run even less voltage through the actual OLED material and make the panel last longer at the same brightness.

You could use blue lasers of differing wavelength to boost gamut even higher too : )

Another cool thought would be using this as a way to artificially boost OLED light as the emitter material itself degrades.

New age Rejuvination.

Could be speaking out of my rear, but its fun lol

Chamber
Posts: 9
Joined: 08 Jul 2022, 23:17

Re: Isn't QD-OLED perfect for backlight strobing?

Post by Chamber » 17 Jul 2022, 14:44

Chief Blur Buster wrote:
13 Jul 2022, 20:55
Long Term Ultrabight HDR OLEDs + Rolling Scan Strobe

Ultrabright HDR-capable OLEDs, with rolling-scan logic, can do a lagless strobe. Far more complex to strobe than global strobing, but fully lagless like a CRT tube. Requires blue/ultraviolet TFT OLED backplanes with individually controllable subpixels, lighting up simple QD phosphors that is just simple inkjet printed paint onto OLED subpixels (one stage of inkjet printing an OLED -- this is currently starting to make cheaper OLEDs possible, as Samsung uses inkjet printing to print the phosphors onto the OLED subpixels).

The 2000-nit OLED backplanes are now becoming possible with custom formulations in some OLED lab screens, so in theory...

The current-carrying backplane can't HDR-bright all pixels simultaneously (windowing) but if the rolling scan window is tight (e.g. 10% of OLED height), you can use the nit headroom of 2000 nit HDR to get 200 nits of high quality strobing -- while only needing to strobe 2000nit for only 10% of pixels at any one time. The HDR-illumination-power windowing design needs to accomodate the needs of strobing. Done this way, this does not overheat the OLED, as 200nit strobing of a 2000-HDR never (A) exceeds the momentary HDR window due to the rolling nature; and (B) average brightness is only, say 200nits (for a 10% rolling scan window). This of course assumes that you can do 2000nits in a 10% HDR window. Alternatively, you could go 5% HDR window, and settle for 100nits. But regardless, strobed will be HDR-less, despite using the HDR nit surge headroom to keep rolling strobes bright.

This is actually easiest/safest done with built-in automatic pixel-turnoff logic built into each TFT subpixel, so you simply program a pixel-off delay, and now the panel is self-strobing with only a single scanout. Some OLED BFI designs are starting to experiment with TFTs that automatically turns off transistors, without needing a separate de-illumination scanout pass.

Also see Custom OLED Rolling Scan Patterns for related discussion about TFT-driven rolling-scan strobing (for CRT-matching lagless strobing since panel scanout can theoretically be in sync with cable scanout).
I didn't even know that rolling scan strobing for OLED was even a possibility. QD-OLED's weakest point is it's full screen brightness, so being able to utilize it's 10% window brightness for strobing is huge. I know the Samsung S95B can hit 1500 nits for a few seconds on a 10% full white window before dropping to 1000 nits and stabilizing there, but with rolling scan strobe, the dip in brightness won't ever happen considering how brief that 10% strobed window would be. The S95B doesn't even have a heatsink, so 2000 nits at 10% white window should easily be possible with a heatsink + new OLED formulations.

Like you mentioned earlier, this rolling scan strobe requires the blue OLED emitters to be the TFT layer, and we currently don't know that yet. The other possibility (if I understood correctly) is if the TFT layer is separate from the blue OLED emitter layer, then global strobing is possible but rolling scan strobing wouldn't be. I know that the Samsung S95B and Sony A95K QD OLED TV's both support backlight strobing, so the display engineers for those TV's already know whether or not the blue OLED emitter layer is the TFT layer and implemented backlight strobing accordingly. So couldn't we just turn on backlight strobing for one of these TV's and take a slow motion video to see whether it's a rolling strobed or globally strobed? Based on that, I think we'd also figure out whether or not the Blue OLED emitters are the TFT layer.
Chief Blur Buster wrote:
13 Jul 2022, 20:55
Consideration: QD Phosphor Decay Differences + Color Shifts During Strobing

QD is a very fast phosphor so this is usually insignificant. However, at least minor color shifts will occur, because this is where microseconds become human visible because of a twist:

Regardless of global or per-pixel strobing, Strobing a phosphor-excitation layer requires custom color calibration for each pulsewidth. If you use 2ms MPRTs the pulsewidth:phosphordecay ratio is protected more and color distortions are much more invisible. For voltage-boosted strobing (same nits), color distortions at 0.2ms MPRT is 10x worse than color distortions from 2ms MPRT when using strobeable phosphor-excitation layers -- Since at 0.2ms MPRT you need to strobe 10x brighter for same Hz, to get as bright as the same 2ms MPRT, but phosphor decay dramatically worsens despite same average screen brightness.

And if Red or Green decays slower, you will have color tinted ghosts during strobing. Probably not as bad as KSF, but the pulsewidths will need to be carefully calibrated, and then color-recalibrated to compensate for the colorshifts of the different phosphordecay speeds of R/G/B, because colorshifts occur if pulsewidth:phosphordecay ratios varies between the color channels.

A 0.01ms phosphor decay divergence in a 1ms strobe -- that's 1% of 1ms. That is bigger than the color tint difference of RGB(128,128,128) and RGB(128,128,130). TEN microseconds, but it's 1% of the photons! Because of tiny ratios involved, this is where microseconds become human visible (a famous example of The Amazing Human Visible Feats of The Millisecond)
Wow, so it turns out that tuning QD-OLED for backlight strobing is much harder than I imagined. They say good things don't come easy, but if QD-OLED supports rolling scan strobe, I think it's worth pursuing. If 0.2ms MPRT proves to be too hard or even impossible to be tuned, then I think it's fine to settle at a higher MPRT like 0.4ms or 0.5ms, which is still substantially better than any consumer monitor on the market. I'm quite mind blown by the fact that the human eye can see microseconds of phosphor decay, every time you post, I always learn something new from you Chief :D.
Chief Blur Buster wrote:
13 Jul 2022, 20:55
Consideration: OLED Brightness Limitation & Bright Strobe Pulses
Talbot-Plateau Law strikes back again with its evil red light saber!

Another major problem with blue OLED is brightness compared to the brightest LCD backlights. Using blue OLED light excitation sources for QD layers (whether it be phosphor-on-TFT-OLED, or an LCD-driven phosphor layer), is that voltage boosting the OLED layer can degrade their lifetimes faster than an LCD backlight, if you want short MPRTs without losing brightness. So you need to use very durable blue OLEDs that are burnin and wear resistant, in order to brightly strobe them without being too dark.

Hope this answers the technical questions of strobing a multilayered screen technology involving OLED in one of its layers...
Luckily Blue Phosphorescent OLED's will be commercially available in 2024 and seem to be massively more efficient, durable, and probably brighter. One study found that blue phosphorescent OLEDs are 10x more durable than blue fluorescent OLEDs at the same brightness. Samsung Display has already been researching Blue Phosphorescent OLEDs and is looking to replace the blue fluorescent OLED emitters in QD-OLED with Blue Phosphorescent OLED. Phosphorescent OLED material currently has 100% luminance efficiency whereas Fluorescent OLED material only has 25% luminance efficiency. I've seen some sources claim brightness gains, but none have given concrete numbers. It does contain phosphor so phosphor decay might be an issue and tuning for all three color primaries might be necessary :twisted:. But red and green OLEDs are already Phosphorescent based in most OLED displays. In the further future, the blue OLED emitter material can be replaced with Galium Nitride blue nanorod LEDs (Samsung QNED) which is extremely burn in resistant, promises much greater brightness, and perfect panel uniformity.
Chief Blur Buster wrote:
13 Jul 2022, 20:55
Various Ways of Layering a Screen That Includes An OLED and Includes Quantum-Dots
Various Ways of Layering a Screen That Includes An OLED and Includes Quantum-Dots
  • When QD is involved with OLED, there are many combinations where OLED and LCD can help each other, or can be pure OLED:
    Global OLED light, TFT-matrix phosphor (OLED-backlit phospor-pixel LCD).
    OLED is just a global light source for an LCD that uses QD-phosphor-subpixels instead of normal color filters. Better color gamut.
  • TFT-matrix OLED light, non-TFT phosphor (Blue-OLED excited non-TFT phosphor pixels).
    OLED blue subpixels lights up phoshors. Fastest GtG but hardest to strobe. No LCD is involved.
  • Locally dimmed OLED light, TFT-matrix phosphor (TFT OLED-backlit FALD for phosphor-pixel LCD).
    OLED is zone-controlled light source for an LCD that utilizes QD-phosphor-subpixels.
  • Etc.
    There are other engineering permutations possible of a hybridized OLED / OLED+LCD
    Including unreleased new layering formulations -- the hybridization possible of OLED+LCD working together, makes discussion a little complicated, and requires a deep-dive on how the display is manufactured, and manufacturers sometimes are tight-lipped at the beginning, especially when they're only showing off prototypes.
Hybrid displays seem fascinating, but future self emissive displays are already promising to fix burn in, boost brightness, and support wider color gamuts.NanoLED/QDEL (a display technology with self emissive quantum dots and no backlight) is set to be commercially available between 2025 to 2026 and is more than 50x brighter than OLED while apparently being cheaper to produce than OLED and burn in resistant. Samsung's QNED was planned to release around the same time between 2024 to 2025 but apparently Samsung's going to postpone it now.
QD-microLED is set to release between 2023 to 2024 but it's probably going to be too expensive for most, as there's no mention of prices or mass production.

The last Hybrid display, the Hisense U9DG (A Duel Cell LCD TV that had 1080p IPS white and black display behind a 4k IPS colored screen) didn't last very long and has already been discontinued, so display manufacturers are going to be more reluctant to develop hybrid displays. I'm sure there would be a lot of added cost to hybridizing OLED+LCD and with NanoLED being cheaper to produce than OLED, I just can't see it working out.

Post Reply