PMOLED/CRT (VFD) vs OLED by Sony

Advanced display talk, display hackers, advanced game programmers, scientists, display researchers, display manufacturers, vision researchers. The masters on Blur Busters.
Post Reply
nuninho1980
Posts: 65
Joined: 26 Dec 2013, 09:49

PMOLED/CRT (VFD) vs OLED by Sony

Post by nuninho1980 » 05 Feb 2014, 09:19

Welcome to Blur Busters Forums after Blur Busters Blog. :)

For gaming/multimedia, PMOLED/CRT (VFD due to some VFD graphics) haven't motion blur while OLED by Sony little motion blur.

PMOLED is passive matrix OLED but currently it has only small screen.

Have you read my message about "Why Do Some OLEDs Have Motion Blur?"?
PC specs:
CPU: i7-4790K
RAM: 2x8GB DDR3@2400MHz
MB: ASUS Z97-Pro Gamer
GPU: EVGA GTX 980 Ti SC+ :D
Opt. disc: LG BD-RE writer BH16NS40
HDDs: SATA both 1TB+3TB
SSD: OCZ RD400 512GB
PSU: AEROCOOL 1kW 80+ Gold
Disly: new LCD soon...

spacediver
Posts: 505
Joined: 18 Dec 2013, 23:51

Re: PMOLED/CRT (VFD) vs OLED by Sony

Post by spacediver » 05 Feb 2014, 16:10

I think there may be a language barrier here nuninho. I'm not sure what the point of your post is.

User avatar
Ahigh
Posts: 95
Joined: 17 Dec 2013, 19:22

Re: PMOLED/CRT (VFD) vs OLED by Sony

Post by Ahigh » 05 Feb 2014, 16:32

http://en.wikipedia.org/wiki/Crystal_LED

Sony has demonstrated displays with no motion artifacting at CES before. Crystal LED was one of the most compelling example.

Other OLED displays also have the capability to demonstrate displays that have no motion artifacts from their nature being self-illuminating and quick response time .. but they don't always intend to do this. Sometimes the display maker will leave the LED's or OLED's lit for the entire duration of the frame for improved brightness.

In any case, Sony has demonstrated a knowledge of eliminating motion artifacting with their LED and OLED screens. In addition, they have demonstrated "Sony MotionFlow" backlight strobing for computer generated and/or HDMI input content at 60fps. Something that is relatively unique among LCD manufacturers.

I don't know if this is the same thing the OP is trying to convey, but possibly it might help to add these comments in to the thread.

nuninho1980
Posts: 65
Joined: 26 Dec 2013, 09:49

Re: PMOLED/CRT (VFD) vs OLED by Sony

Post by nuninho1980 » 05 Feb 2014, 19:31

@spacediver: My any phrase is bad!? But sounds good.

NOTE: Sorry for bad english.


@Ahigh: Ok. But OLED by Sony (impluse-driving - TV OLED 11" and more professional TV-monitors only) is different to "OLED by Samsung" (sample-and-hold - like smartphones, PS Vita...).
OLED is NOT equal to LED for avoid confusion. Not exist LED pixels for shorter screen than ~50" but LED pixels need GIGANT screen. ;)
EDIT: Ah!? New CLED technology?! :) But does it get flicker as CRT and PMOLED?
PC specs:
CPU: i7-4790K
RAM: 2x8GB DDR3@2400MHz
MB: ASUS Z97-Pro Gamer
GPU: EVGA GTX 980 Ti SC+ :D
Opt. disc: LG BD-RE writer BH16NS40
HDDs: SATA both 1TB+3TB
SSD: OCZ RD400 512GB
PSU: AEROCOOL 1kW 80+ Gold
Disly: new LCD soon...

nuninho1980
Posts: 65
Joined: 26 Dec 2013, 09:49

Re: PMOLED/CRT (VFD) vs OLED by Sony

Post by nuninho1980 » 09 Feb 2014, 18:08

@Chief Blur Buster: please should write here.
PC specs:
CPU: i7-4790K
RAM: 2x8GB DDR3@2400MHz
MB: ASUS Z97-Pro Gamer
GPU: EVGA GTX 980 Ti SC+ :D
Opt. disc: LG BD-RE writer BH16NS40
HDDs: SATA both 1TB+3TB
SSD: OCZ RD400 512GB
PSU: AEROCOOL 1kW 80+ Gold
Disly: new LCD soon...

User avatar
Chief Blur Buster
Site Admin
Posts: 6509
Joined: 05 Dec 2013, 15:44

Re: PMOLED/CRT (VFD) vs OLED by Sony

Post by Chief Blur Buster » 10 Feb 2014, 19:20

AMOLED, PMOLED, and Crystal LED have their pros/cons.

PMOLED is generally not practical at large screen sizes due to the incredible brightness needed at sub-microsecond levels (when hitting high dotclock frequencies). You can impulse-drive an AMOLED by using a rolling-scan algorithm, see http://www.blurbusters.com/high-speed-v ... ster-oled/

A japanese paper showed this diagram:
Image


Discrete LED (Also known as Crystal LED in the miniaturized TV-sized versions) is typically used by Times Square jumbotrons, stadium jumbotrons, and the big screens seen outdoors throughout Las Vegas. Most of them are sample-and-hold because they need to keep LEDs on for long periods to have maximum brightness (no strobing). So they will have motion blur, an image persistence equal to frame length (e.g. 60Hz = 16.7ms persistence = 16.7 pixels of motion blurring during 1000 pixels/second).

All of them (PMOLED, AMOLED, and discrete LED) can be strobed to reduce motion blur. PMOLED by its own behaviour is very flickery. However, it's not easy to scale PMOLED reliably to huge resolution sizes at high refresh rates because as you raise the dotclock, there is less time to light up each pixel. The on-time for each pixel is shorter because there are more pixels to illuminate, so you have to flash each pixel much brighter to compensate for the short time period of the flicker. PMOLED at 1920x1080p could easily achieve microsecond-league persistence, but the image might be too dim to be useable. If a whole scanline is illuminated at a time, you are still using strobe lengths in the tiny fraction of a millisecond league (135KHz horizontal scanrate at 1080p@120Hz suggests a 1080p120 PMOLED persistence of roughly 1/135,000th of a second = 74 microsecond persistence. Ouch. Where are you going to get the light output for a bright 1080p120Hz image on a PMOLED? At 0.074ms persistence, you would need to be roughly 112 times brighter than a 120Hz sample-and-hold AMOLED, in order to stay as bright as a non-strobed AMOLED. (8.3ms frame cycle at 120Hz divided by 0.074ms persistence = 112 times the light output if you flash for only 1/224th the amount of time).

Much easier (for law of physics) is a rolling-scan algorithm, and achieving more practical persistence of about 0.1ms to 2ms.

Currently, 2ms pixel illumination length is probably a practical goal to aim for initially (that appears to be common strobe length for LightBoost/ULMB/BENQ Blur Reduction/Turbo240) as it requires an OLED only ~4x brighter at 120Hz, to achieve same brightness as non-strobed. However, shorter is even better where possible -- my eyes are able to tell apart 0.5ms, 1.0ms and 2.0ms persistence based on my testing on fixed-firmware XL2720Z's (adjustable persistence all the way to 0.5ms) during fast panning motion at ~3000 pixels/second at http://www.testufo.com/photo and other fast-panning/strafing/turning tests running at framerate==refreshrate. There are strong points of diminishing returns occuring at 0.5ms persistence. If rolling-scan AMOLED displays can add a persistence(brightness) adjustment similar to the LightBoost % settings, a good adjustment range would be 0.5ms to 4ms (or even all the way to sample-and-hold).

To achieve 0.5ms persistence using a 120Hz refresh (an 8.3ms refresh cycle), you will need ~17x brightness to keep the screen as bright as non-strobed (8.3ms/0.5ms = 16.6). So you can see 0.5ms and 1.0ms is still quite aggressive, given OLED still has some difficulty pushing out enough brightness to achieve short persistence needed. However, the manufacturers are working on solving the OLED light output issue, and we should eventually be able to get the light output necessary for CRT-clarity AMOLEDs -- probably more easily than bright large-format PMOLED.

PMOLED flicker by necessity, while AMOLED can optionally be made to flicker (for the purposes of lowering persistence -- motion blur reduction). Technologically, it is much easier to make an AMOLED flicker, than to scale PMOLED brightly to current HD sizes. The question is: Convincing display makers to release a 120Hz rolling-scan AMOLED for the gaming market -- at affordable prices.
Head of Blur Busters - BlurBusters.com | TestUFO.com | Follow @BlurBusters on Twitter

       To support Blur Busters:
       • Official List of Best Gaming Monitors
       • List of G-SYNC Monitors
       • List of FreeSync Monitors
       • List of Ultrawide Monitors

nuninho1980
Posts: 65
Joined: 26 Dec 2013, 09:49

Re: PMOLED/CRT (VFD) vs OLED by Sony

Post by nuninho1980 » 11 Feb 2014, 07:18

Ok but I don't need more than 60Hz for PMOLED. ;)

On CRT, I'm playing games at 60Hz because CPU/GPU hardly hold 75@75 to 120fps@120Hz. And framerate is capped (Konami, Sega, emulators (MAME, PS2...)). While I'm watching multimedia (except Blu-Ray at 60 Hz) at 50Hz. :D


CRT, PMOLED and VFD don't get motion blur - 100% sure! Also CLED? But not LightBoost, Plasma nor OLED by Sony due to SOFTLY flicker.
PC specs:
CPU: i7-4790K
RAM: 2x8GB DDR3@2400MHz
MB: ASUS Z97-Pro Gamer
GPU: EVGA GTX 980 Ti SC+ :D
Opt. disc: LG BD-RE writer BH16NS40
HDDs: SATA both 1TB+3TB
SSD: OCZ RD400 512GB
PSU: AEROCOOL 1kW 80+ Gold
Disly: new LCD soon...

User avatar
Chief Blur Buster
Site Admin
Posts: 6509
Joined: 05 Dec 2013, 15:44

Re: PMOLED/CRT (VFD) vs OLED by Sony

Post by Chief Blur Buster » 11 Feb 2014, 13:59

CRT, PMOLED and VFD don't get motion blur - 100% sure! Also CLED? But not LightBoost, Plasma nor OLED by Sony due to SOFTLY flicker.
LightBoost flickers a lot less than PMOLED and a lot less than 60Hz CRT.

I have six monitors here including CRT, LightBoost, ULMB, Turbo240, and BENQ Blur Reduction. Although strobe backlights have a harsher on-off cycle than CRT phosphor decay, this is more than overcome by the massively higher refresh rate (e.g. 120Hz squarewave strobing-flicker cycle versus 60Hz phosphor-decay-flicker cycle).

On the other hand, if you play emulators and use the 60Hz strobing tricks (black frame insertion combined with LightBoost) the 60Hz strobing does indeed flicker noticeably more than a 60Hz CRT (apples-versus-apples refresh rate comparision).
Head of Blur Busters - BlurBusters.com | TestUFO.com | Follow @BlurBusters on Twitter

       To support Blur Busters:
       • Official List of Best Gaming Monitors
       • List of G-SYNC Monitors
       • List of FreeSync Monitors
       • List of Ultrawide Monitors

nuninho1980
Posts: 65
Joined: 26 Dec 2013, 09:49

Re: PMOLED/CRT (VFD) vs OLED by Sony

Post by nuninho1980 » 11 Feb 2014, 17:24

Chief Blur Buster wrote:LightBoost flickers a lot less than PMOLED and a lot less than 60Hz CRT.
Is LB harder flicker or faster refresh rate than CRT?
I have six monitors here including CRT, LightBoost, ULMB, Turbo240, and BENQ Blur Reduction. Although strobe backlights have a harsher on-off cycle than CRT phosphor decay, this is more than overcome by the massively higher refresh rate (e.g. 120Hz squarewave strobing-flicker cycle versus 60Hz phosphor-decay-flicker cycle).

On the other hand, if you play emulators and use the 60Hz strobing tricks (black frame insertion combined with LightBoost) the 60Hz strobing does indeed flicker noticeably more than a 60Hz CRT (apples-versus-apples refresh rate comparision).
Ok. :) But does Crystal LED by Sony never get motion blur at 50 or 60 Hz?
PC specs:
CPU: i7-4790K
RAM: 2x8GB DDR3@2400MHz
MB: ASUS Z97-Pro Gamer
GPU: EVGA GTX 980 Ti SC+ :D
Opt. disc: LG BD-RE writer BH16NS40
HDDs: SATA both 1TB+3TB
SSD: OCZ RD400 512GB
PSU: AEROCOOL 1kW 80+ Gold
Disly: new LCD soon...

User avatar
Chief Blur Buster
Site Admin
Posts: 6509
Joined: 05 Dec 2013, 15:44

Re: PMOLED/CRT (VFD) vs OLED by Sony

Post by Chief Blur Buster » 11 Feb 2014, 23:41

nuninho1980 wrote:
Chief Blur Buster wrote:LightBoost flickers a lot less than PMOLED and a lot less than 60Hz CRT.
Is LB harder flicker or faster refresh rate than CRT?
We have to consider the "apples versus apples" factor.

There's roughly a 25% increased visibility of flicker on a strobe backlight (human dependant). So that means 75Hz strobe backlight flickers more than 75Hz CRT, but once you push refresh at least 25% higher, it begins to flicker less. Which means 120Hz strobe backlight flickers a lot less than 60Hz CRT.

CRT 75Hz flicker visibility < LCD strobe 75Hz flicker visibility -- BENQ Blur Reduction minimum stroberate
CRT 85Hz flicker visibility < LCD strobe 85Hz flicker visibility -- NVIDIA ULMB minimum stroberate
CRT 100Hz flicker visibility < LCD strobe 100Hz flicker visibility -- NVIDIA LightBoost minimum stroberate
CRT 120Hz flicker visibility < LCD strobe 120Hz flicker visibility -- Common maximum stroberate of most strobe backlights

But...
CRT 60Hz flicker visibility > LCD strobe 100Hz flicker (less visible)
CRT 60Hz flicker visibility > LCD strobe 120Hz flicker (less visible)

Same refresh rate, strobing flickers more than CRT. However, you can overcome this by an overkill amount of refresh rate. So LightBoost 120Hz flickers a LOT less than CRT 60Hz.
Ok. :) But does Crystal LED by Sony never get motion blur at 50 or 60 Hz?
No, Crystal LED is currently very crappy in motion blur. It's the same problem in Why Do Some OLEDs Have Motion Blur?. You can strobe Crystal LED, but the prototype is not strobed, and during tests, it had some motion clarity issues during fast motion.

Less flicker almost always means more motion blur. It's a law-of-physics effect if you're maintaining refresh rate (e.g. 60Hz). You have to raise your refresh rate to compensate for the flicker, but video is 60fps, and there's no global standard for native 120fps television video at the moment.

Now, some strobe backlights are able to flash as short as 1ms per refresh cycle -- similar to a CRT phosphor.. That's pretty good for motion clarity. The only way to do 1ms persistence is to make each frame last 1ms long -- you can either do strobing (flash frames for 1ms), or fill all timeslots (do 1000fps at 1000Hz). Doing 1ms persistence without strobing/phosphor/light modulation is not currently yet possible.

Flicker visibility can be softened by using a simulated phosphor decay effect in future strobing/scanning backlights, at a cost of flicker-versus-motion-clarity. You can workaround some of this (e.g. by scanning instead of full strobing), but generally, flicker is unavoidable for zero-motion-blur 60Hz.
Head of Blur Busters - BlurBusters.com | TestUFO.com | Follow @BlurBusters on Twitter

       To support Blur Busters:
       • Official List of Best Gaming Monitors
       • List of G-SYNC Monitors
       • List of FreeSync Monitors
       • List of Ultrawide Monitors

Post Reply