Dynamic Dimming Gamma Correction - AVS Forum | Home Theater Discussions And Reviews
Forum Jump: 
 2Likes
  • 2 Post By harlekin
 
Thread Tools
post #1 of 5 Old 11-23-2019, 11:28 AM - Thread Starter
Advanced Member
 
bdht's Avatar
 
Join Date: Aug 2016
Posts: 631
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 298 Post(s)
Liked: 116
Dynamic Dimming Gamma Correction

If a display has a dynamic lamp dimming function, but there is no internal processing for gamma/grayscale correction, is it possible to perform this function on the PC with a 3d lut and madvr?

http://www.cine4home.de/Specials/Viv...20Preview3.htm
"The adaptive lighting control worked effectively on our tester, but was not free of errors. Especially in predominantly dark pictures a color pumping into the reddish was visible in places, which is probably due to a not perfectly adjusted RGB-correction. In the first stage, the adaptive lighting control is largely reliable, in modes 2 and "Infinity", however, the provoked artifacts are too large."

http://www.cine4home.de/tests/projek...2Mico_Test.htm
"An adaptive lighting control works only as well as the corresponding real-time gamma compensation allows. This is responsible for ensuring that we do not perceive annoying brightening in light to dark or vice versa transitions. The system works very reliably with the Mico 50. The adaptive light control of the LEDs harmonizes with the digital gamma correction"
bdht is online now  
Sponsored Links
Advertisement
 
post #2 of 5 Old 11-25-2019, 08:36 PM
Advanced Member
 
harlekin's Avatar
 
Join Date: May 2007
Posts: 740
Mentioned: 69 Post(s)
Tagged: 0 Thread(s)
Quoted: 317 Post(s)
Liked: 150
No, basically.

Dynamic brightness means, that the gamma curve fluctuates all over the place. You cant compensate for that dynamically.

Normally, as brightness doesnt impact color error that much (at smaller variability changes), it should be possible to calibrate a grayscale at a constant (use tricks to get it? ) brightness level, and then the deviation shouldnt be that high.

But simply put, if display gamma isnt stable - lut wouldnt be able to fix it. Those corrections are all static and expect a stable display.
You wouldnt even be able to obtain a proper lut correction, if the display varied brigthnesslevels (and therefore gamma) on the fly.

Last edited by harlekin; 11-25-2019 at 08:42 PM.
harlekin is online now  
post #3 of 5 Old 11-25-2019, 08:50 PM - Thread Starter
Advanced Member
 
bdht's Avatar
 
Join Date: Aug 2016
Posts: 631
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 298 Post(s)
Liked: 116
Quote:
Originally Posted by harlekin View Post
No, basically.

Dynamic brightness means, that the gamma curve fluctuates all over the place. You cant compensate for that dynamically.

Normally, as brightness doesnt impact color error that much (at smaller variability changes), it should be possible to calibrate a grayscale at a constant (use tricks to get it? ) brightness level, and then the deviation shouldnt be that high.

But simply put, if display gamma isnt stable - lut wouldnt be able to fix it. Those corrections are all static and expect a stable display.
You wouldnt even be able to obtain a proper lut correction, if the display varied brigthnesslevels (and therefore gamma) on the fly.
Thank you for the explanation. So how bout making a 3d lut for rec2020/p3 to use with madvrs hdr to sdr, if the projector does compensate gamma like the mico50 or runco q750?
http://www.cine4home.de/tests/projek...o/Gamma22b.jpg
bdht is online now  
Sponsored Links
Advertisement
 
post #4 of 5 Old 11-26-2019, 01:39 AM
Advanced Member
 
harlekin's Avatar
 
Join Date: May 2007
Posts: 740
Mentioned: 69 Post(s)
Tagged: 0 Thread(s)
Quoted: 317 Post(s)
Liked: 150
Same issue essentially. Calibration, or 3D LUT calibration expects a stable display/image characteristic to which it will correct towards. If thats not there - it breaks calibration as a concept.

I have a similar issue with my 2016 LG OLED, which has dynamic dimming based on the average brightness level of the image (so had plasma TVs in the past (less aggressively)) - there is no way to correct that with normal calibration, or 3D LUTs.

But - the "feature" here at least gets disabled, once you put 10% color windows on screen for calibration - so you are calibrating a stable display state - and the deviation (in brightness) only then sets in (based on the scenes average brightness level (if it gets too high)) afterwards.

As this "dynamic adjustment" only is present in maybe 2% of the scenes in an average movie calibration still makes sense though. (I also have a 3D LUT box, and I can do jack about it - so I'm speaking from practical experience. )

If you've got a "dynamic adjustment" based on roomlight f.e. (as lets say a Macbook with an ambient light sensor has), the measured gamma (SDR) varies by a significant amount even while measuring a 10 step grayscale. So you absolutely have to disable that (which you can on a Macbook) before even attempting to calibrate (or profile the display).


A 3D Lut is basically a correction based on 1000s of individual measurements and not 10 as in the example above, so it amplifies the problem.

You have to understand, that what a 3D Lut does is, that it moves individual color points 'more' than f.e. just a normal greyscale (whitepoint) and CMS (primary colors at 100% saturation) calibration. And it does so not in one direction (vector) but in thousands.

And your fundamental problem with "dynamic brightness adjustment" is, that one and the same color, will be displayed differently based on roomlight, or average picture brightness level (didn't read up on your projector and what it does). So the same color gets displayed different based on criteria you dont control.

You absolutely cant calibrate this away. (Think about it for a minute.. ).

In closing - again, most displays/projectors that use a form of dynamic brightness adjustment (if its based on average image brightness level), will disable it, when you calibrate with 10% windows (size of the calibration pattern), or 10% windows with APL (average picture level) compensation (basically if bright window, make boarder dark, if dark window make border brighter (grey) - so average image level doesnt fluctuate as much). And then you can calibrate.

But if you see gamma curving away from target at the lower end - always even in your default 'cinema modes' presets, on an SDR calibration - it likely means, that dynamic brightness adjustment is impacting your calibration - and thereby ruining it. And deviations from target would be visible at 20% and 30% grey as well. (Not just 10% where gamma can break depending on display characteristic alone.)

If your gamma is on target - meaning, following it roughly (like you've posted in the last graph) - it means, that at least during calibration "dynamic brightness adjustment" is disabled. Which means, you can calibrate (or even 3D LUT calibrate) the thing on a base level (then ignore the error that sets in, when dynamic brightness adjustment sets in on real life content).


Basically calibration, or 3D LUT calibration doesnt deal with a temporal (time as a dimension ) element. It wants and needs stable readings throughout the process.
Dominic Chan and bdht like this.

Last edited by harlekin; 11-26-2019 at 01:56 AM.
harlekin is online now  
post #5 of 5 Old 11-26-2019, 12:48 PM - Thread Starter
Advanced Member
 
bdht's Avatar
 
Join Date: Aug 2016
Posts: 631
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 298 Post(s)
Liked: 116
Gotcha, so actually correcting the dynamic dimming function needs to be done by the manufacturer. If it is and done well then try the 3d lut with the function off.

Quote:
Originally Posted by harlekin View Post
If your gamma is on target - meaning, following it roughly (like you've posted in the last graph) - it means, that at least during calibration "dynamic brightness adjustment" is disabled. Which means, you can calibrate (or even 3D LUT calibrate) the thing on a base level (then ignore the error that sets in, when dynamic brightness adjustment sets in on real life content).
For that graph, the dynamic dimming was on for the measurement.
bdht is online now  
Sponsored Links
Advertisement
 
Reply Display Calibration

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