2019 OLED Anticiaption Thread - Page 4 - AVS Forum | Home Theater Discussions And Reviews
Forum Jump: 
 70Likes
Reply
 
Thread Tools
post #91 of 136 Old 01-02-2019, 10:33 PM
Member
 
Maybejames's Avatar
 
Join Date: Aug 2016
Location: Hong Kong SAR
Posts: 23
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 16 Post(s)
Liked: 22
Quote:
Originally Posted by unknownuser200 View Post
Vincent was up EARLY this morning to get this one out!
Very happy to see the tick list of improvements that I'd hoped for this year.
VRR being one of the biggest as a Gamer.

Hopefully the staff at AVSforums can get their hands on the sets properly at CES instead of the normal light touch puff pieces.
Maybejames is offline  
Sponsored Links
Advertisement
 
post #92 of 136 Old 01-02-2019, 10:35 PM
Member
 
sonny21's Avatar
 
Join Date: Oct 2007
Posts: 95
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 4 Post(s)
Liked: 11
Quote:
Originally Posted by venus933 View Post
From the above link

"This (2nd generation alpha 9 processor) includes an upgraded black frame insertion system called ’OLED Motion Pro’ that now operates at 100/120Hz (compared to 50/60Hz last year) and with shorter black frame cycle (25% vs. 50% last year). LG says the system eliminates flicker and maintains brightness, which were FlatpanelsHD’s two main concerns with the BFI system in the 2018 LG OLED models. Other improvements include a separate ”smooth gradation” picture setting that no longer reduces resolution."
I was so hyped reading that tidbit. Wow, if this means OLED will finally deliver real motion improvements after years. I don't want to get too excited without seeing people get their hands on and testing it out, but hard not to get excited.
TimRex likes this.
sonny21 is offline  
post #93 of 136 Old 01-02-2019, 11:36 PM
AVS Forum Special Member
 
Join Date: Jan 2015
Posts: 1,044
Mentioned: 51 Post(s)
Tagged: 0 Thread(s)
Quoted: 909 Post(s)
Liked: 946
BFI visibly improves near-black uniformity on my C8 and at 120Hz it should be able to do that without 3:2 pulldown for movie content and without noticeable flicker in bright scenes. Huge selling point for me.
shyguy3763 and video_analysis like this.
jk82 is offline  
Sponsored Links
Advertisement
 
post #94 of 136 Old 01-02-2019, 11:44 PM
Member
 
Join Date: May 2018
Posts: 27
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 21 Post(s)
Liked: 5
Quote:
Originally Posted by BillP View Post
Sorry, but at normal sizes and viewing distances, you are not going to see any improvement with an 8k display over a 4k display. Not to mention there is no 8k material.
Not true, you can get benefit if you have a powerful gaming pc running games in 8k. You realistically would want a sli gtx 1080ti system at least for anything playable with modern games, but it's doable. You can even see differences with older pc games and run them at a solid 60fps... As far as movies goes, a lot of movies won't ever see any decent benefits with 8k over 4k because of how they were shot. The industry will have to change to filimg with 8k native cameras for most movies? Will this even realistically happen within the next 10 years?

Only real benefit for 8k is with pc games and eventually consoles...
unknownuser200 likes this.
smoothbutter is offline  
post #95 of 136 Old 01-03-2019, 01:58 AM
AVS Forum Special Member
 
Join Date: Jan 2016
Posts: 7,626
Mentioned: 18 Post(s)
Tagged: 0 Thread(s)
Quoted: 4298 Post(s)
Liked: 3723
Quote:
Originally Posted by Wizziwig View Post
"There are no major changes to the actual OLED panel, produced by LG Display, but LG explains that will introduce an improved anti-reflective filter this year."

Move along, nothing to see here.
Quote:
Originally Posted by jk82 View Post
BFI visibly improves near-black uniformity on my C8 and at 120Hz it should be able to do that without 3:2 pulldown for movie content and without noticeable flicker in bright scenes. Huge selling point for me.
You were saying, Wizz? This could be the feature that makes the idea of upgrading again a little less dreadful.
video_analysis is offline  
post #96 of 136 Old 01-03-2019, 02:19 AM
AVS Forum Special Member
 
GregLee's Avatar
 
Join Date: Jul 2002
Location: Waimanalo HI
Posts: 4,524
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 1101 Post(s)
Liked: 386
Congratulations to fafrd for a daring and correct prediction.

Greg Lee
GregLee is offline  
post #97 of 136 Old 01-03-2019, 01:33 PM
AVS Forum Special Member
 
Wizziwig's Avatar
 
Join Date: Jul 2001
Location: SoCal, USA
Posts: 3,544
Mentioned: 23 Post(s)
Tagged: 0 Thread(s)
Quoted: 2238 Post(s)
Liked: 1520
Quote:
Originally Posted by video_analysis View Post
You were saying, Wizz? This could be the feature that makes the idea of upgrading again a little less dreadful.
Have you seen a 2018 OLED with the BFI activated? Vertical banding in the usual test clips we discuss in the uniformity thread is still plainly visible to me. Maybe if you won the lottery and your set is already in the 99th percentile, then BFI can hide the remaining issues. Wish it was that simple but it's not.

BFI also introduces additional issue when the BFI rate doesn't match the content rate. You just get image duplication and ghosting. If they force it to 120Hz like many recent LCDs, then it will be essentially useless for the majority of people watching 60Hz or less content. As I posted in the other thread, they should instead offer a mode where the black duration is increased from 8ms to 12ms which would actually help to finally reach 1080 lines of motion resolution on 60Hz content.
reallynotnick likes this.
Wizziwig is offline  
post #98 of 136 Old 01-03-2019, 02:49 PM
AVS Forum Special Member
 
NintendoManiac64's Avatar
 
Join Date: Feb 2012
Location: Northeast Ohio
Posts: 2,011
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 1027 Post(s)
Liked: 456
Quote:
Originally Posted by smoothbutter View Post
Not true, you can get benefit if you have a powerful gaming pc running games in 8k. You realistically would want a sli gtx 1080ti system at least for anything playable with modern games, but it's doable
The bigger issue is that even the newer and faster 2080Ti only supports HDMI 2.0 (8k 30Hz 4:2:0 8bpc) and DisplayPort 1.4 (8k 60Hz 4:2:0 8bpc), and at such high resolutions you'd ideally want a variable refresh rate...which Nvidia does not support over HDMI.

On a similar note, I would imagine that AMD's upcoming 7nm Navi GPU architecture (we may get a product announcement at AMD's CES keynote on the 8 9th) to be the best chance at being the first GPU to include support for HDMI 2.1, especially since that's the architecture that was developed in tandem with Sony for use in the PS5.

Last edited by NintendoManiac64; 01-04-2019 at 12:29 AM.
NintendoManiac64 is offline  
post #99 of 136 Old 01-03-2019, 03:13 PM
AVS Forum Addicted Member
 
fafrd's Avatar
 
Join Date: Jun 2002
Posts: 13,314
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 14 Post(s)
Liked: 4878
Quote:
Originally Posted by jk82 View Post
BFI visibly improves near-black uniformity on my C8 and at 120Hz it should be able to do that without 3:2 pulldown for movie content and without noticeable flicker in bright scenes. Huge selling point for me.
Thanks for confirming a theory I've been asking about.

With 50% BFI, a 5% grey field is actually the average of a 0% black field and a 10% field, and since the 0% field has 'perfect' uniformity (off), 5% w/ BFI should have uniformity at least twice as good as 5% w/o BFI.

Would love to see comparison pics if you could post them (5% w/ BFI at your preferred OLED Light level
and 5% uniformity w/o BFI @ ~half that OLED Light level.

Improved near-black uniformity through use of [email protected] is one of the huge sleeper features of the 2019 WOLEDs...
stl8k and TimRex like this.
fafrd is offline  
post #100 of 136 Old 01-03-2019, 03:26 PM
AVS Forum Addicted Member
 
fafrd's Avatar
 
Join Date: Jun 2002
Posts: 13,314
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 14 Post(s)
Liked: 4878
Quote:
Originally Posted by Wizziwig View Post
Have you seen a 2018 OLED with the BFI activated? Vertical banding in the usual test clips we discuss in the uniformity thread is still plainly visible to me. Maybe if you won the lottery and your set is already in the 99th percentile, then BFI can hide the remaining issues. Wish it was that simple but it's not.
I just added a post on the theory - near-black nonuniformity w/ BFI should be reduced by at least 50%. Pics need to be taken with a long exposure time and a tripod (and in the pitch-black except for the TV). However bad the 'vertical banding' looks with BFI, it should look twice as bad without (~140% as bad based on perception).

OLED Light levels need to be adjusted for ~equal luminance (meaning ~twice as with BFI as without).


Quote:
BFI also introduces additional issue when the BFI rate doesn't match the content rate. You just get image duplication and ghosting. If they force it to 120Hz like many recent LCDs, then it will be essentially useless for the majority of people watching 60Hz or less content. As I posted in the other thread, they should instead offer a mode where the black duration is increased from 8ms to 12ms which would actually help to finally reach 1080 lines of motion resolution on 60Hz content.
Support for VRR should mean 96Hz can be supported through HDMI, which would allow a 'double-shutter' projector-emulation of 24fps content presented as a 96Hz stream of frames repeating twice with black frames inserted between presentations of the same frame as well as subsequent frames...

In addition, I am hopring that the reference to '25% BFI' means they also support 75% BFI for 60Hz content (which is 12 ms black duration you are asking for).

A hard-coded solution that only supports 25% BFI would be hugely dissapointing (though hopefully upgradeable through FW).
video_analysis likes this.
fafrd is offline  
post #101 of 136 Old 01-03-2019, 04:04 PM
AVS Forum Special Member
 
Wizziwig's Avatar
 
Join Date: Jul 2001
Location: SoCal, USA
Posts: 3,544
Mentioned: 23 Post(s)
Tagged: 0 Thread(s)
Quoted: 2238 Post(s)
Liked: 1520
Quote:
Originally Posted by fafrd View Post
I just added a post on the theory - near-black nonuniformity w/ BFI should be reduced by at least 50%. Pics need to be taken with a long exposure time and a tripod (and in the pitch-black except for the TV). However bad the 'vertical banding' looks with BFI, it should look twice as bad without (~140% as bad based on perception).

OLED Light levels need to be adjusted for ~equal luminance (meaning ~twice as with BFI as without).
Even if the BFI helped uniformity (it didn't in my testing of a couple scenes from Arrival on the A1E or C8), what are you going to do for HDR? Even in LCD land where brightness is abundant, nobody uses BFI because of the massive loss in brightness.

As 2016 owners, your uniformity is most likely already better than any of the 2017/2018 sets, but if you or video_analysis want to conduct an experiment, find the brightest pair of sunglasses in your home and then raise your OLED light until the 5% slide through the glasses matches what you saw without glasses at normal OLED light settings. Do you see an appreciable loss in vertical banding during the camera pans of the many torture test scenes? Don't really care what it does in a static photo of a slide - especially with unknown camera exposure and quality. FYI, most of the recent sets I've seen have visible vertical banding all the way up to 30%.

From the flatpanels article:

"This includes an upgraded black frame insertion system called ’OLED Motion Pro’ that now operates at 100/120Hz (compared to 50/60Hz last year) and with shorter black frame cycle (25% vs. 50% last year). LG says the system eliminates flicker and maintains brightness, which were FlatpanelsHD’s two main concerns with the BFI system in the 2018 LG OLED models. "

You still think it's gong to reduce image persistence on 60hz content vs. their 2018 implementation when not using 120hz interpolation?
Wizziwig is offline  
post #102 of 136 Old 01-03-2019, 04:27 PM
AVS Forum Special Member
 
GregLee's Avatar
 
Join Date: Jul 2002
Location: Waimanalo HI
Posts: 4,524
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 1101 Post(s)
Liked: 386
Quote:
Originally Posted by Wizziwig View Post
Even in LCD land where brightness is abundant, nobody uses BFI because of the massive loss in brightness.
Not so. I use it on my Samsung Q9FN, and I have brightness to spare, by turning up Backlight and Contrast.

Greg Lee
GregLee is offline  
post #103 of 136 Old 01-03-2019, 05:09 PM
Senior Member
 
mikeyf's Avatar
 
Join Date: Mar 2006
Location: Rochester, New York
Posts: 211
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 10 Post(s)
Liked: 16
Quote:
Originally Posted by davewolfs View Post
I've had a Kuro and Panasonic plasma for around a decade. Hoping to jump on the OLED train soon! Is 2019 the year to do it or better off waiting until 2020?
You and me both. My Kuro is eleven years old and going strong. I've been holding out recently to see if HDMI 2.1 is closer to happening than we think. I was concerned about motion detail with OLED but it sounds and reads like that is less of an issue especially with the A9F.
All thoughts and experiences are welcome.....
mikeyf is offline  
post #104 of 136 Old 01-03-2019, 05:13 PM
Member
 
Blaine Doss's Avatar
 
Join Date: Sep 2002
Location: in MO.
Posts: 187
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 51 Post(s)
Liked: 22
Great news I am definitely purchasing a C9 this year I am consurned with a new process starting next year and the bug associated with the first run of a new process..

Sent from my Pixel XL using Tapatalk
Blaine Doss is offline  
post #105 of 136 Old 01-03-2019, 05:34 PM
AVS Forum Addicted Member
 
fafrd's Avatar
 
Join Date: Jun 2002
Posts: 13,314
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 14 Post(s)
Liked: 4878
[quote=Wizziwig;57372038]Even if the BFI helped uniformity (it didn't in my testing of a couple scenes from Arrival on the A1E or C8), what are you going to do for HDR? Even in LCD land where brightness is abundant, nobody uses BFI because of the massive loss in brightness.

At least on WOLED, the loss of brightness associated with BFI is unnecessary (and I'm hoping LG will eventually figure that out).

The aging associated with WOLED is a function of photon emission. Ruduce the time of photon emission by half, you can double the intensity.

If LG is leaving the instantaneous ABL limit at 150cd/m2 despite engaging 50% BFI, that's poor engineering - it should increase to 300cd/m2 (and the 'meaning' of OLED Light should double).

Hopefully, they will get there eventually (maybe with NHKs help ).

Quote:
As 2016 owners, your uniformity is most likely already better than any of the 2017/2018 sets, but if you or video_analysis want to conduct an experiment, find the brightest pair of sunglasses in your home and then raise your OLED light until the 5% slide through the glasses matches what you saw without glasses at normal OLED light settings. Do you see an appreciable loss in vertical banding during the camera pans of the many torture test scenes? Don't really care what it does in a static photo of a slide - especially with unknown camera exposure and quality. FYI, most of the recent sets I've seen have visible vertical banding all the way up to 30%.
That is an excellent suggestion and I will do that and report back. Using a filter on a camera is a way to get a more recordable/objective comparison, but next time I see DSE on a dark pan, I'll try throwing on my glasses, increasing OLED Light to come close to apparent brightness, and repeat the scene.

Quote:
From the flatpanels article:

"This includes an upgraded black frame insertion system called ’OLED Motion Pro’ that now operates at 100/120Hz (compared to 50/60Hz last year) and with shorter black frame cycle (25% vs. 50% last year). LG says the system eliminates flicker and maintains brightness, which were FlatpanelsHD’s two main concerns with the BFI system in the 2018 LG OLED models. "

You still think it's gong to reduce image persistence on 60hz content vs. their 2018 implementation when not using 120hz interpolation?
So first of all, any discussion of improved motion performanc really only makes sense in the context of interpolation. If you are only interpolating to 60Hz, it appears that LG has decided to prevent 75% and possibly also 50% BFI options, so in that case, you'd be correct, [email protected] will take a step backwards versus 2018 (in terms of persistance, but loss of brightness will improve).

If you are interpolating to 120Hz, you'll get persistance reduced to 4.2ms instead of 8.3ms and with reduced flicker at 120Hz instead of 60Hz - big improvement.

More importantly, VFR support will hopefully allow 24fps content to be displayed at 96Hz-double-shutter-single-frame-repeat mode (with the appropriate support on HTPC or whatever). This will reduce persistance by 25% from 41.7ms to 31.25ms. 50% BFI can hopefully be engaged on top of that which would reduce persistance by another 5.2ms to 26.0ms (and all without interpolation).

And as far as 60Hz content, the same HTPC solution would allow it to be presented as a 120Hz w/ 50% BFI HDMI stream and the TVs native 50% [email protected] input will allow an effective 75% BFI @ 60Hz to be delivered (despite LGs likely decision to decide that the flicker is so bad, they want to eliminate that option).

In addition, the VFR features (whatever they are) may offer additional options/benefits.
fafrd is offline  
post #106 of 136 Old 01-03-2019, 05:34 PM
AVS Forum Special Member
 
Join Date: Jan 2015
Posts: 1,044
Mentioned: 51 Post(s)
Tagged: 0 Thread(s)
Quoted: 909 Post(s)
Liked: 946
Quote:
Originally Posted by fafrd View Post
Would love to see comparison pics if you could post them (5% w/ BFI at your preferred OLED Light level
and 5% uniformity w/o BFI @ ~half that OLED Light level.

Here is a long exposure (10 seconds, ISO 50) comparison picture. I adjusted oled-light so that both presets (bfi on and off) measured at exactly 0.05 nits on the 5% slide. (is that even nits? the Y value in HCFR? Sorry new to calibration)

Please note that my smartphone always makes the center brighter on low light scenes and it also added some color banding and tinting that isn't nearly as bad looking with my eyes.





Also I was too lazy to remove the i1Display Pro for the photo. Just got it today and it's probably gonna stick on the TV for the next couple of days.
Attached Thumbnails
Click image for larger version

Name:	bfi_off.jpg
Views:	29
Size:	307.6 KB
ID:	2505322   Click image for larger version

Name:	bfi_on.jpg
Views:	36
Size:	274.0 KB
ID:	2505324   Click image for larger version

Name:	0.5_nits_bfi-on_bfi-off.gif
Views:	142
Size:	492.9 KB
ID:	2505326  
Wizziwig, spear, tigertim and 1 others like this.

Last edited by jk82; 01-03-2019 at 05:39 PM.
jk82 is offline  
post #107 of 136 Old 01-03-2019, 05:52 PM
AVS Forum Addicted Member
 
fafrd's Avatar
 
Join Date: Jun 2002
Posts: 13,314
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 14 Post(s)
Liked: 4878
Quote:
Originally Posted by jk82 View Post
Here is a long exposure (10 seconds, ISO 50) comparison picture. I adjusted oled-light so that both presets (bfi on and off) measured at exactly 0.5 nits on the 5% slide.

Please note that my smartphone always makes the center brighter on low light scenes and it also added some color banding and tinting that isn't nearly as bad looking with my eyes.





Also I was too lazy to remove the i1Display Pro for the photo. Just got it today and it's probably gonna stick on the TV for the next couple of days.
That is incredible - thanks!

These pics are difinitive enough that I consider the theory proven, but since I didn't realize that you have an i1DisplayPro, if you want to take some more pics that are even more difinitive, here are some things you could try in a second round:

1/ Try 4% or even 3% instead of 5% - whatever near-black field shows the worst vertical banding.

2/ Find a location for your i1DisplayPro that matches the average intensity of the field. Especially on the pic without BFI, the center of the screen where you have the i1DisplayPro appears to be a 'bright spot' so much of the screen appears darker and none of the more uniform picture with BFI gets darker as a result. By fnding a location closer to the average intensity (such as halfway to either edge), the second pic with BFI should show equal areas getting brighter and darker.

For reference, here's what the 4% field on my 65C6P looks like (without BFI):
Attached Thumbnails
Click image for larger version

Name:	image.jpg
Views:	36
Size:	123.6 KB
ID:	2505344  
fafrd is offline  
post #108 of 136 Old 01-03-2019, 06:47 PM
AVS Forum Special Member
 
Wizziwig's Avatar
 
Join Date: Jul 2001
Location: SoCal, USA
Posts: 3,544
Mentioned: 23 Post(s)
Tagged: 0 Thread(s)
Quoted: 2238 Post(s)
Liked: 1520
If the exposure is fixed, why is one of the photos darker than the other?

For a 100% peak of 120 nits, as is usually the case for SDR calibration, your 5% slide should be at 0.09 nits on an OLED tracking 2.4 gamma.

Assuming it's properly calibrated for both with/without BFI, do you actually perceive some improvement in uniformity on any of the camera panning scenes? For me the banding was still visible so I wrote off this entire approach. Not to mention the other side effects of running everything with BFI enabled (too dim for HDR, 3:2 judder, flicker for some, etc.).
Wizziwig is offline  
post #109 of 136 Old 01-03-2019, 07:04 PM
AVS Forum Special Member
 
Wizziwig's Avatar
 
Join Date: Jul 2001
Location: SoCal, USA
Posts: 3,544
Mentioned: 23 Post(s)
Tagged: 0 Thread(s)
Quoted: 2238 Post(s)
Liked: 1520
Quote:
Originally Posted by GregLee View Post
Not so. I use it on my Samsung Q9FN, and I have brightness to spare, by turning up Backlight and Contrast.
Engaging a 50% BFI as discussed here would cut your max HDR brightness by 50%. So unless your Q9FN can do 8000 nits in HDR without BFI, you're going to be compromising the HDR performance for any 4000 nit titles. Even more so for games which routinely use up to the 10,000 nit max. For SDR, I'm sure you have ample brightness in reserve to make it a non-issue.

Looks like Samsung still does a proper 60Hz BFI with 50% light loss.

Most FALD LCDs use a scanning backlight with an adjustable "black" time period. The unlit black band scanning down the screen can be adjusted in height for fine tuning the amount of acceptable light loss. At max setting, the black band could be the size of the entire screen but many manufacturers no longer offer that at 60Hz because of flicker complaints. You end up in a blur regression as we saw on the 2016 vs. 2018 Sony. You might end up in a similar regression on the 2019 OLEDs if they only allow 120Hz BFI. Time will tell but their press info is not encouraging.

Last edited by Wizziwig; 01-03-2019 at 07:11 PM.
Wizziwig is offline  
post #110 of 136 Old 01-03-2019, 07:30 PM
AVS Forum Addicted Member
 
fafrd's Avatar
 
Join Date: Jun 2002
Posts: 13,314
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 14 Post(s)
Liked: 4878
Quote:
Originally Posted by Wizziwig View Post
If the exposure is fixed, why is one of the photos darker than the other?
Because he normalized/calibrated on a spot on the screen that happened to be a 'bright spot' (lighter than average).

If he chose a calibration spot that was darker than average, that same image would be lighter than the other.

By selecting a calibration spot that matches the average luminance, part if the second image will be brighter and part of the second image will be darker (and the average of both will be roughly equal).

Quote:
For a 100% peak of 120 nits, as is usually the case for SDR calibration, your 5% slide should be at 0.09 nits on an OLED tracking 2.4 gamma.
I'm guessing he activated BFI before taking the first pic and used measurement of that screen to determine the OLED Light level to use.

For the purpose of this comparison, whatever luminance level demonstrates the worst streaking / nonuniformity is best.

Quote:
Assuming it's properly calibrated for both with/without BFI, do you actually perceive some improvement in uniformity on any of the camera panning scenes? For me the banding was still visible so I wrote off this entire approach. Not to mention the other side effects of running everything with BFI enabled (too dim for HDR, 3:2 judder, flicker for some, etc.).
This is a key point and perhaps we all need to share our favorite near-black torture scenes with jk82 .
fafrd is offline  
post #111 of 136 Old 01-03-2019, 08:15 PM - Thread Starter
AVS Forum Special Member
 
venus933's Avatar
 
Join Date: Apr 2012
Posts: 4,405
Mentioned: 10 Post(s)
Tagged: 0 Thread(s)
Quoted: 2365 Post(s)
Liked: 1942
Based upon my own panel and comments from reviewers/calibrators of their 2018 samples I thought the 2018 panels at least got back to 2016 levels.
Attached Thumbnails
Click image for larger version

Name:	5.jpg
Views:	48
Size:	676.4 KB
ID:	2505382  

LG 65B8 - Primary
LG 55EG9100 - Bedroom
Sonos Playbase
MS Xbox One S (using strictly as a disk player)
venus933 is offline  
post #112 of 136 Old 01-03-2019, 08:35 PM
Member
 
Join Date: Aug 2018
Posts: 77
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 57 Post(s)
Liked: 40
Quote:
Originally Posted by fafrd View Post

This is a key point and perhaps we all need to share our favorite near-black torture scenes with jk82 .

Well, think about when individuals are told that they have their OLED light set too high to judge their panel's uniformity. That is the beginning and end of the effect that BFI has on uniformity; you get the uniformity benefit of a higher OLED light setting at a lower light output.


So what your uniformity looked like at 10% grey without BFI(@50%) is what your 5% grey will look like with BFI activated, when calibrated to the same light output. Your essentially moving your blemishes further on down the line. It's also why BFI can't help uniformity of HDR content, as OLED light is maxed out here (not taking into account any dynamic tone-mapping shenanigans).


So yes, there is very clearly an improvement to uniformity for a set light output with BFI engaged, including during troubling panning scenarios.


Did that make sense?
JPbuckwalter is offline  
post #113 of 136 Old 01-03-2019, 08:39 PM
Senior Member
 
Jin-X's Avatar
 
Join Date: Apr 2006
Posts: 301
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 141 Post(s)
Liked: 146
Quote:
Originally Posted by Wizziwig View Post
"There are no major changes to the actual OLED panel, produced by LG Display, but LG explains that will introduce an improved anti-reflective filter this year."

Move along, nothing to see here.
Plenty to see if you play video games, pretty much all the new features are focused in that area. Also plenty of us do use BFI on the LG 8 series, it works great on SDR gaming as you can make up the luminance drop by raising OLED light, and input lag stays the same. As for flicker, I don't see it, image looks the same with it on and off, just better motion resolution with it on.
Cam1977 likes this.
Jin-X is offline  
post #114 of 136 Old 01-03-2019, 08:51 PM
Member
 
Join Date: Aug 2018
Posts: 77
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 57 Post(s)
Liked: 40
Hopefully the BFI is adjustable and not set at 25% only, as that would just be a straight downgrade on both motion resolution and uniformity fronts for those of us who can tolerate the flicker at 50%.


Also, lemme get a 75% option while you're at it. The more options, the better!
JPbuckwalter is offline  
post #115 of 136 Old 01-04-2019, 12:08 AM
AVS Forum Special Member
 
unknownuser200's Avatar
 
Join Date: Sep 2008
Location: Scranton , PA
Posts: 2,294
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 1445 Post(s)
Liked: 763
Quote:
Originally Posted by JPbuckwalter View Post
Hopefully the BFI is adjustable and not set at 25% only, as that would just be a straight downgrade on both motion resolution and uniformity fronts for those of us who can tolerate the flicker at 50%.


Also, lemme get a 75% option while you're at it. The more options, the better!
or native 120hz panels lol
unknownuser200 is offline  
post #116 of 136 Old 01-04-2019, 03:19 AM
AVS Forum Special Member
 
Join Date: Jan 2015
Posts: 1,044
Mentioned: 51 Post(s)
Tagged: 0 Thread(s)
Quoted: 909 Post(s)
Liked: 946
Quote:
Originally Posted by Wizziwig View Post
If the exposure is fixed, why is one of the photos darker than the other?

For a 100% peak of 120 nits, as is usually the case for SDR calibration, your 5% slide should be at 0.09 nits on an OLED tracking 2.4 gamma.

Assuming it's properly calibrated for both with/without BFI, do you actually perceive some improvement in uniformity on any of the camera panning scenes? For me the banding was still visible so I wrote off this entire approach. Not to mention the other side effects of running everything with BFI enabled (too dim for HDR, 3:2 judder, flicker for some, etc.).

It is as fafrd already said, I didn't use correctly calibrated presets. I just adjusted oled-light until both were reading 0.05 nits at 5% for comparisons sake. The lowest I can measure is about 0.03 and I wanted to be close to that.


I sometimes play sidecroller games (Hollow Knight for example) at a low peak brightness on the C8. It definitely helps uniformity in that case and I don't notice the BFI flicker at a low oled-light setting.



I also noticed on the picture that my panel has these weird half-round shapes at the bottom that I have seen on other panels in the uniformity thread before.

The 10 seconds exposure picture exaggerates everything..
jk82 is offline  
post #117 of 136 Old 01-04-2019, 03:31 AM
AVS Forum Special Member
 
Join Date: Jan 2015
Posts: 1,044
Mentioned: 51 Post(s)
Tagged: 0 Thread(s)
Quoted: 909 Post(s)
Liked: 946
Quote:
Originally Posted by fafrd View Post
That is incredible - thanks!

These pics are difinitive enough that I consider the theory proven, but since I didn't realize that you have an i1DisplayPro, if you want to take some more pics that are even more difinitive, here are some things you could try in a second round:

1/ Try 4% or even 3% instead of 5% - whatever near-black field shows the worst vertical banding.

2/ Find a location for your i1DisplayPro that matches the average intensity of the field. Especially on the pic without BFI, the center of the screen where you have the i1DisplayPro appears to be a 'bright spot' so much of the screen appears darker and none of the more uniform picture with BFI gets darker as a result. By fnding a location closer to the average intensity (such as halfway to either edge), the second pic with BFI should show equal areas getting brighter and darker.

For reference, here's what the 4% field on my 65C6P looks like (without BFI):

If I pick a darker spot or darker slide, I don't think the i1DisplayPro would even read anything. So I would have to crank up oled-light.


BFI for some reason seems to make the low end brighter, which makes comparisons difficult and which is also the reason why I measured the 5% and not the peak luminance of the presets.

The TV also seems to compensate for the brightness loss at lower oled-light settings. For example at oled-light 0, BFI off/on seems to have almost similar peak brightness when it actually should be half. The higher I go with oled-light the more difference in peak brightness there is.


edit:
Just out of interest I measured peak brightness with BFI off and on at different oled-light settings. As I suspected at oled-light 0 it's almost identical and at 100 it's like half.

Peak brightness BFI off / BFI on

oled 0
48/45

oled 25
116/82

oled 50
186/110

oled 75
254/137

oled 100
328/166


Also thanks LG for reducing sdr peak brightness in a fw update.
cockatiel likes this.

Last edited by jk82; 01-04-2019 at 04:08 AM.
jk82 is offline  
post #118 of 136 Old 01-04-2019, 08:05 AM
AVS Forum Addicted Member
Industry Insider
 
Cleveland Plasma's Avatar
 
Join Date: Jun 2005
Location: Cleveland, Ohio
Posts: 24,563
Mentioned: 81 Post(s)
Tagged: 1 Thread(s)
Quoted: 6676 Post(s)
Liked: 6670
Cleveland Plasma is online now  
post #119 of 136 Old 01-04-2019, 08:57 AM
AVS Forum Addicted Member
 
fafrd's Avatar
 
Join Date: Jun 2002
Posts: 13,314
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 14 Post(s)
Liked: 4878
Quote:
Originally Posted by jk82 View Post
If I pick a darker spot or darker slide, I don't think the i1DisplayPro would even read anything. So I would have to crank up oled-light.


BFI for some reason seems to make the low end brighter, which makes comparisons difficult and which is also the reason why I measured the 5% and not the peak luminance of the presets.

The TV also seems to compensate for the brightness loss at lower oled-light settings. For example at oled-light 0, BFI off/on seems to have almost similar peak brightness when it actually should be half. The higher I go with oled-light the more difference in peak brightness there is.


edit:
Just out of interest I measured peak brightness with BFI off and on at different oled-light settings. As I suspected at oled-light 0 it's almost identical and at 100 it's like half.

Peak brightness BFI off / BFI on

oled 0
48/45

oled 25
116/82

oled 50
186/110

oled 75
254/137

oled 100
328/166


Also thanks LG for reducing sdr peak brightness in a fw update.
OK, this is helpful. If you have the energy to try again, here is what you need to do:

With BFI engaged, set OLED Light so peak brightness measured full-field is 120cd/m2 (well below ABL limit of 150 cd/m2) - probably to 50-60 based on your prior measurements - and adjust Brightness so 0% (video level 16) is true black (invisible) and 1% (video level 18) is just barely visible to dark-adjusted eyes. You should be able to achieve 0% true black and 1% visible, but if not, click brightness down until 0% is invisible, even if 1% gets crushed (and 2% is the lowest % your eyes can percieve). Probably best to use BT.1886 gamma.

Now with BFI off, perform the same calibration to 120cd/m2 peak by reducing OLED Light (probably to 25-30 based on your prior measurements). Hopefully the Brightness setting will not need to change, but if it does, you will need to maintain different settings of OLED Light and Brightness for the two measurements.

Now you are on the non-BFI screen which should have the worst uniformity so you need to display a full-field pattern at 5% or 4% or 3% (whichever you feel is worst and is most likely to be visible on content) and take your picture. You can also measure cd/m2 if you want but should do so by moving your meter around to take a 3x3 or 4x4 matrix of measurements including both darkest areas and lightest areas of the screen.

Now switch back to your BFI settings and take a picture of the same % field. Repeat the matrix of measurements. If the average of the measurements is very different, adjust OLED Light up or down until the average of the measurements is close to what they were for the non-BFI field and take a third picture.

Those 2 or 3 pics at 4% or 5% should tell the story in terms of what your eyeballs will be able to perceive but should not require your 10-second exposure times. If you want your longer exposure time, you can try to perform the exercise at 3% (and possibly even 2%, at least if you are using gamma of 2.1).

Those 3x3 or 4x4 matrices of measurements will allow us to quantify the improved uniformity that our eyes can see from the pics.

I'm assuming you have access to pluge patterms but checking/adjusting brightness with full-field near-black patterns (0%-5%) is the best way to dial-in Brightness setting on an OLED.
fafrd is offline  
post #120 of 136 Old 01-04-2019, 09:08 AM
AVS Forum Addicted Member
 
fafrd's Avatar
 
Join Date: Jun 2002
Posts: 13,314
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 14 Post(s)
Liked: 4878
Quote:
Originally Posted by JPbuckwalter View Post
Well, think about when individuals are told that they have their OLED light set too high to judge their panel's uniformity. That is the beginning and end of the effect that BFI has on uniformity; you get the uniformity benefit of a higher OLED light setting at a lower light output.


So what your uniformity looked like at 10% grey without BFI(@50%) is what your 5% grey will look like with BFI activated, when calibrated to the same light output. Your essentially moving your blemishes further on down the line. It's also why BFI can't help uniformity of HDR content, as OLED light is maxed out here (not taking into account any dynamic tone-mapping shenanigans).


So yes, there is very clearly an improvement to uniformity for a set light output with BFI engaged, including during troubling panning scenarios.


Did that make sense?
No, it's actually better than that (twice as good).

Lets assume uniformity at 10% is better than your uniformity at 5%, and for sake of argument, lets say 10% is half as nonumiform as 5%.

With OLED Light increased to match luminance and BFI engaged, 5% is now going to include 50% of the time displaying the non-BFI 10% (which is half as nonuniform) and 50% of the time diaplaying perfect black (off) which has NO nonuniformity. So the average of those two will reduce the non-uniformity in half, meaning one-quarter the nonuniformity ofnthe non-BFI field (or half the nonuniformity of the 10% field.

Because of gamma and other nonuniformities, it's not exactly true that 10% w/ BFI = 5% w/o BFI but what is true is that:

-nonuniformity is absolute (that is why it is easier to percieve at lower intensity levels than highe intensity level).
-50% BFI only displays that fixed nonuniformity for one-half of the time.
-hence 50% BFI reduces intrinsic nonuniformity by half (at any underlying luminance level).
robert0507 likes this.
fafrd is offline  
Sponsored Links
Advertisement
 
Reply OLED Technology and Flat Panels General

Thread Tools
Show Printable Version Show Printable Version
Email this Page Email this Page


Forum Jump: 

Posting Rules  
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off