AVS Forum banner
1 - 20 of 21 Posts

·
Registered
Joined
·
2,760 Posts
Discussion Starter · #1 ·
Below is an excerpt from my review of the 55EA9800 OLED:


Calibration was performed with a Jeti 1211 spectroradiometer, along with Klein K-10a (thanks Buzz!) and X-Rite i1D3 colorimeters, both profiled off the reference Jeti spectro.


Sources close to the engineering team, who spoke under conditions of anonymity since they admittedly have an overactive imagination, report overhearing the following conversation during the design of the EA9800:
“10 point adjustments are great. They worked good on our LED LCDs.”


“Well, we gotta come out ahead! Let's give 'em 20!”


“Some people didn't like our old partial CMS adjustments, though. They worked OK, but they didn't have enough control to really get it dialed in.”


“I know you think we don't have all the kinks ironed out yet, but put the full CMS adjustment in there too. It's OK, at least we'll have the most adjustments. I pity the fool who tries to use 'em, though! And make sure to give 'em that new space age remote... You know, that wand like thing I seen you waving around. It won't help none when they go to make the adjustments, but it's the coolest thing I ever did see!”



The EA9800 presented some tough calibration challenges. The first issue became apparent soon into my first pass of the 20 point adjustment: if the same shaped object is displayed on the AE9800 for more than a brief period, the TV loses light output, giving way to fatigue. It doesn't matter if the intensity or color is changed; the shape seems to be the trigger. Unchecked, that leads to wildly inaccurate calibration. Stopping the calibration after each control or two and popping up other portions of the menu or other test patterns seemed to be the only solution; and even then, multiple passes were required for any precision. If the problematic 20 point adjustment is ignored and only the simple 2 point adjustment is calibrated, gamma and grayscale tracking have visible errors because of the quirky tracking of the 2 point adjustment.


In addition to panel fatigue, the 20 points progressively mistrack at the high end if the contrast control is not kept very high.


I found that the best way to calibrate the 20 point control was not to try to do it in real time, but rather pass by pass, making adjustments after each pass as needed.


Finally, I discovered that all of the CMS adjustments, but especially the red saturation and luminance, have severe side effects. Making significant adjustments to these controls resulted in patchy looking skin tones, and taken to the extreme they turn Hollywood's most beautiful people into horrifying zombie-like creatures.


The great news is despite these issues, the AE9800 had the most impressive contrast performance I've ever seen, and it did so without resorting to such things as dimming zones or floating blacks. Full black fields were under the .0001 fL threshold of the mighty Klein K-10a.



After calibration, the ANSI checkerboard contrast ratio measured around 70,000:1, with a very small amount of light leakage leading to blacks that measured between .0006 and .0007 fL. Whites measured about 47.1 fL in this test.


There was no real change between different size measurement windows, and there was no color shift when ABL finally kicked in with full fields. There should be no mystery surrounding what size or style of windows to use when calibrating the EA9800, as it appears that just about any window will give similar results. The maximum light output with a 100% full white field was about 24 fL, indicating minimal ABL intrusion. Another test of ABL is the dynamic brightness pattern on the AVS 709 test disc, which showed some brightness limiting but less than what would be seen on a plasma.


White field uniformity was excellent, with no visible changes across the screen. Black uniformity was not an issue since black was totally dark. Off axis, whites take on a more off white, yellowish tone, but otherwise the picture appears to have no change. Images are perfectly watchable at extreme off axis angles, though with the curved screen and white balance shift giving a distinctive character.



Below are measurements of the differences between the color space selections of the EA9800's advanced picture menu.


Standard and BT709:



EBU:



SMPTE:



Wide:



Sony and some others in the industry recommend using spectrometers set to the Judd Vos modified 2 degree CMF for OLED displays. I tried that for the day mode calibration and all the before calibration measurements. For the night mode, I switched back to the standard 1931 CIE 2 degree setting. I found the standard 1931 CIE setting gave the most neutral looking white balance, with the Judd mod setting appearing a bit cool and giving a slightly reddish cast to skin tones.


ISF Night mode after calibration:



ISF Day mode after calibration:
 

·
Registered
Joined
·
12,065 Posts
Good stuff Chad. Besides the white shift did you find the curved screen to display artifacts or diminish the off angle viewing experience ? I have heard differing observations on this.
 

·
Registered
Joined
·
2,760 Posts
Discussion Starter · #3 ·

Quote:
Originally Posted by chunon  /t/1497982/lg-oled-calibration#post_23909937


Good stuff Chad. Besides the white shift did you find the curved screen to display artifacts or diminish the off angle viewing experience ? I have heard differing observations on this.
Thanks! No artifacts from the curve... It messes with the reflections in big ways when directly in front, but it seems to reduce reflections if you're off to the side.

I can see the whole image quite clearly even at extreme off axis angles, though the WB shift is visible. I actually like the curved screen off axis.
 

·
Registered
Joined
·
12,065 Posts
Sounds like a nice set overall, that contrast ratio is unreal. How do the blacks fair in brigher lighting ?
 

·
Registered
Joined
·
364 Posts
Coming from a home enthusiast view point (not being close to a pro calibrator), I see LG is still mucking up things with their calibration firmware. I have had enough frustrations calibrating my old LG plasma with it's 20pt grey scale and now understanding what seems like panel fatigue issues (thanks to Chad for bringing that up in his review). I'm so glad that this is the last day with my LG (as my ST60 will be delivered tomorrow). I hope LG can get this issue straight once and for all.
 

·
Registered
Joined
·
6,723 Posts
Greetings


Blacks always look blacker in brighter lighting. One of the benefits of adding backlights ...


regards
 

·
Registered
Joined
·
5,916 Posts
This picture is a vast exaggeration of what Chad and I actually viewed - blame the camera. Those brown curves are curtains hanging dead verticle. The curved screen distorted the reflection. I found it quite distracting even though it wasn't as bad as it looks.

 

·
Registered
Joined
·
5,916 Posts

Quote:
Originally Posted by realzven  /t/1497982/lg-oled-calibration#post_23910940


It's the display pro in the middle ?

That's Chad's D3.

Quote:
Originally Posted by realzven  /t/1497982/lg-oled-calibration#post_23910945


Btw, this display is like mirror...
Quote:
Originally Posted by buzzard767  /t/1497982/lg-oled-calibration#post_23910774

This picture is a vast exaggeration of what Chad and I actually viewed - blame the camera.
 

·
Registered
Joined
·
511 Posts
i know buzz but from this photo the brown curtains is quite visible... in a batcave the set will look gorgeous with this mll and fantastic contrast :b)
 

·
Registered
Joined
·
5,163 Posts

Quote:
Originally Posted by buzzard767  /t/1497982/lg-oled-calibration#post_23910774


This picture is a vast exaggeration of what Chad and I actually viewed - blame the camera. Those brown curves are curtains hanging dead verticle. The curved screen distorted the reflection. I found it quite distracting even though it wasn't as bad as it looks.

Buzz, what did you think of profiling your K10-A to the Jeti.??


Did Chads Jeti 1211 have any sync problems.??


From looking at your picture and Chads charts, it looks like OLED has a way to go before it ready for prime time.


ss
 

·
Registered
Joined
·
2,760 Posts
Discussion Starter · #13 ·
The Jeti wouldn't sync to it, but I turned sync off and took continuous readings at various levels and repeatability was very good.


Yes, the reflections were terrible sitting spot on in front of the TV, but that huge sliding glass door was right behind the viewer and in front of the TV. I think it would fare much better even in just as bright of a room if it were just positioned so that the door was not directly in front of it.
 

·
Registered
Joined
·
5,916 Posts

Quote:
Originally Posted by sillysally  /t/1497982/lg-oled-calibration#post_23911216


Buzz, what did you think of profiling your K10-A to the Jeti.??


From looking at your picture and Chads charts, it looks like OLED has a way to go before it ready for prime time.

Hi ss - I wasn't there when the K10 profiling was accomplished. Chad and I both arrived at Chris's house around 9PM Saturday night and all I had time for was taking luminance measurements relating to contrast. We old guys go to be early ya know.


The picture was taken Sunday mid morning. The curtains were covering a sliding glass door and there was more light then we would have liked. The picture itself greatly enhanced the actual reflection that we saw. Lets just say that I don't know much about using my Canon.... Had there been a wall behind me it would have been fine.


The OLED is drop dead gorgeous but Chad had his hands full dealing with the typical LG CMS drawbacks. He did a great job though, and the charts show very good linearity.
 

·
Banned
Joined
·
3,587 Posts
Contrast ratio will become a "non-spec" if OLED or regular LED TVs ever replace other display technologies. Once black is black, you can't calculate a contrast ratio because you would be dividing by "zero" to calculate the ratio. You could call the contrast ratio with truly black blacks infinite without much argument. If every display has an infinite contrast ratio, there's nothing left to talk about and contrast ratio will disappear as a video display specification.
 

·
Registered
Joined
·
5,916 Posts

Quote:
Originally Posted by Doug Blackburn  /t/1497982/lg-oled-calibration#post_23915022


Contrast ratio will become a "non-spec" if OLED or regular LED TVs ever replace other display technologies. Once black is black, you can't calculate a contrast ratio because you would be dividing by "zero" to calculate the ratio. You could call the contrast ratio with truly black blacks infinite without much argument. If every display has an infinite contrast ratio, there's nothing left to talk about and contrast ratio will disappear as a video display specification.

We were measuring ANSI, Doug. An all black pattern cannot be measured by the Klein - 0.0000 FtL. A 4x4 ANSI pattern, however, shows that some light is being transported through the glass and although the OLED pixels were turned off there was still measurable light in the glass that covers them and it's coming from adjacent white squares..
 

·
Registered
Joined
·
14,830 Posts

Quote:
Originally Posted by Doug Blackburn  /t/1497982/lg-oled-calibration#post_23915022


Contrast ratio will become a "non-spec" if OLED or regular LED TVs ever replace other display technologies. Once black is black, you can't calculate a contrast ratio because you would be dividing by "zero" to calculate the ratio. You could call the contrast ratio with truly black blacks infinite without much argument. If every display has an infinite contrast ratio, there's nothing left to talk about and contrast ratio will disappear as a video display specification.

If that happens, that marketing guys will be very disappointed.
 

·
Registered
Joined
·
14 Posts
Hi Guys,
in this year I suppose that you viewed so many LG OLED so ...
Do you ever seen something like "clouding" in OLED LG panels?
Yesterday I calibrated a 55EA980 (european model) and when it reproduce a full black pattern (or also partial black) it is cleary visible some big area not black like others....
I repeat something like clouding in LCD panels.....

On the other side I can confirm that it's impossible to have a linear and stable resposne from the pannel ... the TV arbitrary dimm and increase light emission ...
so It's impossibile to do a decent gamma calibration.

thank all.
 
1 - 20 of 21 Posts
Top