AVS Forum banner

12221 - 12240 of 12733 Posts

·
Registered
Joined
·
29 Posts
For OLEDs there is NO difference between BT.1886 and gamma 2.4.

I was surprised about this comment, but after some testing, I can see that when your 0% black level = 0 nits, BT1886 becomes Gamma 2.4. If 0% is not 0 nits, you will get the logarithmic curve. Very interesting!
 

·
Registered
Joined
·
18 Posts
Since the corection files is TV brand specific and even with differences from year to year I choosed to stick with a neutral profile since I have an Philips. Even if the panel is LG Philips have another Picture processor. But i will test the LG OLED correction file an se how it goes.

I agree about the reference curve, it is strange that it is flat.

You are right about Philips having a different processor. But the meter needs to know what tech it's looking at. The RGB Gain/Bias controls are what fix the colour nuances of the processors. Maybe I will eat my words since I don't have an OLED, but I'll stand by my claim lol.
Things like this makes me so confused... Calibrate against another brands correction file and set every taget right against that corection or just use a neutral value and every target right against the neutral file. I want Philips specific correction files!! 😃 My tv uses 2016 LG panel.
 

·
Registered
Joined
·
8,836 Posts
Thanks for the answer, that only confirms that it probably is a good way to go and use BT. 1886. My only concern is the peak up to about 2,5 at 90-100% white, but maybe it isn't a problem?
Small gamma errors at high input levels are totally inconsequential. A gamma value of 2.5 (instead of 2.4) at 90% translates to a level error of -1.1%, or a delta-E of 0.4.
 

·
Registered
Joined
·
8,836 Posts
Things like this makes me so confused... Calibrate against another brands correction file and set every taget right against that corection or just use a neutral value and every target right against the neutral file. I want Philips specific correction files!! 😃 My tv uses 2016 LG panel.
The meter doesn't care whose sticker is on your TV. It sees the R/G/B components and "interprets" them. If the TV uses 2016 LG panel then indeed that would be the right correction file to use.
 

·
Registered
Joined
·
8,836 Posts
Hey y'all - after A LOT of reading and help from this forum I was able to calibrate my Vizio PX75-G1 for SD and HD content using HCFR. The calibration process is starting to make sense, and with the calibration settings on my TV I'm able to tweak quite a bit. Thanks for all the help!

HDR calibration baffles me still unfortunately. Is anyone aware of an updated HCFR "dummies" guide that covers a typical HDR calibration workflow? I'm running HCFR on a Macbook Pro using bootcamp, and have a i1 Display Pro retail probe.

A few follow-on comments/questions.

I have HDR test patterns that send the proper metadata to my TV, but the techniques I use to dial-in white balance and color don't seem to do anything (in HCFR and on the TV). My TV has an 11 point RGB white balance calibration mode (5% to 100%) but changing the RGB for a given % doesn't move the needle on correcting the D65 RGB errors. I'm not even sure if I'm even supposed to be correcting for D65 in HDR.

I've seen forum references to a PQ curve but I don't know how to bring that up in HCFR, nor do I know how to interpret it or change it's values based on my TV calibration white balance/color settings. Again, tweaking RGB for white balance and color doesn't seem to affect anything.

Thanks for your patience...I'm trying to read all I can but there's so much out there and it's hard to weave it all together into a sensible HCFR calibration workflow.

Dan
The process of calibrating HDR is not all that different. The main differences are:
- The EOTF ("gamma curve") is absolute and not relative to the 100% point; e.g., regardless of whether your TV has 500 nits or 1000 nits peak white, the 50% input point should read 94.4 nits.
- The colour space is BT.2020 instead of BT.709. Since no consumer TV can reach the full gamut boundary, you can only calibrate the TV using patterns up to the limits of your TV, e.g., using Rec2020/P3 or Rec2020/Rec709 patterns, or use 50% saturation points of Rec2020.
- Related to the above two, the amplitudes of the test patterns will be at 50% input.
 

·
Registered
Joined
·
18 Posts
After a new look in HFCR i can se there are two choices related to Oled.
- LED OLED (EDR for OLED created with primaries R, G, B, W)
- WRGB OLED (LGB7) (2017))


Anyone who knows the difference between these two? Because my screen is the 2016 model, i assume I should use the first alternative?
 

·
Registered
Joined
·
48 Posts
The process of calibrating HDR is not all that different. The main differences are:
- The EOTF ("gamma curve") is absolute and not relative to the 100% point; e.g., regardless of whether your TV has 500 nits or 1000 nits peak white, the 50% input point should read 94.4 nits.
- The colour space is BT.2020 instead of BT.709. Since no consumer TV can reach the full gamut boundary, you can only calibrate the TV using patterns up to the limits of your TV, e.g., using Rec2020/P3 or Rec2020/Rec709 patterns, or use 50% saturation points of Rec2020.
- Related to the above two, the amplitudes of the test patterns will be at 50% input.
Thanks Dominic - I think I’m starting to get it. HCFR reads the HDR greyscale patterns just fine (I’m using the free HDR patterns linked on this website), I just can’t seem to adjust/move the delta-E using the RGB controls on my TV like I did in SDR mode. In other words, adjusting the TV’s whitespace RGB controls for any of the % live readings doesn’t fix the D65 errors for that given pattern. The %s don’t budge. Do I have to limit the input or something?
 

·
Registered
Joined
·
8,836 Posts
Thanks Dominic - I think I’m starting to get it. HCFR reads the HDR greyscale patterns just fine (I’m using the free HDR patterns linked on this website), I just can’t seem to adjust/move the delta-E using the RGB controls on my TV like I did in SDR mode. In other words, adjusting the TVs whitespace RGB controls for any of the % live readings doesn’t fix the D65 errors for that given pattern. Do I have to limit the input or something?
That could be a limitation of your specific TV.
 

·
Registered
Joined
·
48 Posts
That could be a limitation of your specific TV.
What’s more likely is I’m not doing it right. :p To be frank, I don’t have a fundamental understanding of a proper HDR calibration workflow. I know this is for an OLED (I have the Vizio PX75-G1 LED) but do you think this would be a good place to start? https://www.lg.com/us/support/products/documents/UHDA Calibration Procedure.pdf

Edit: and instead of using the OLED light adjustment, I use the backlight setting to set the initial luminance?
 

·
Registered
Joined
·
18 Posts
Hello to the expertice! :)


I´m going nuts….


I have done the calibration and temperature, greyscale and gamma lines up beutifully.


But it seems that i can´t get low Delta E on the colors... Maybe It´s my TV or maybe I´m doing something wrong in my workflow?


The TV is an Philips 901F.


When i did the grayscale run i had to make some sacrifice at the 10% level to get the lowest possible DeltaE at the other levels. But despite the sacrifice seend to the RGB curve i get an low DeltaE even att 10% grey. Se the attached file. Can the compromise at the lower end affect my colors?


In my TV i have an color adjustment that affects the luminance for all the colors, and I have an HUE adjustment and a saturation adjustment for every color.


What I have been reading on the web I Think the best way to go when you just have one luminance adjustment that affect all the colors is to aim for a luminace of red about 21% lower than the luminace value that have been read for white?


I have only seen the target at 21% for red. Is there some target % for the other colors aswell? Due to my luminace contol that affects all colors, I assume my way to go is to do luminance adjustments till I get the best compromise luminance value between red, green and blue?


The 21% Y-target for red was related to 100% color pattern, but i assume that this 21% value works the same way to the 75% pattern? Take a reading of the Y-value for 75% white and then the Y-value for red 75% should be 21% lower than the value for white?
 

Attachments

·
Registered
Joined
·
8,836 Posts
Hello to the expertice! :)


I´m going nuts….


I have done the calibration and temperature, greyscale and gamma lines up beutifully.


But it seems that i can´t get low Delta E on the colors... Maybe It´s my TV or maybe I´m doing something wrong in my workflow?


The TV is an Philips 901F.


When i did the grayscale run i had to make some sacrifice at the 10% level to get the lowest possible DeltaE at the other levels. But despite the sacrifice seend to the RGB curve i get an low DeltaE even att 10% grey. Se the attached file. Can the compromise at the lower end affect my colors?


In my TV i have an color adjustment that affects the luminance for all the colors, and I have an HUE adjustment and a saturation adjustment for every color.


What I have been reading on the web I Think the best way to go when you just have one luminance adjustment that affect all the colors is to aim for a luminace of red about 21% lower than the luminace value that have been read for white?


I have only seen the target at 21% for red. Is there some target % for the other colors aswell? Due to my luminace contol that affects all colors, I assume my way to go is to do luminance adjustments till I get the best compromise luminance value between red, green and blue?


The 21% Y-target for red was related to 100% color pattern, but i assume that this 21% value works the same way to the 75% pattern? Take a reading of the Y-value for 75% white and then the Y-value for red 75% should be 21% lower than the value for white?
It would be easier for others to help with troubleshooting if you attach the HCFR data file (*.chc, zipped to upload).

For accurate colours you need CMS controls for each colour. If the TV only has a global colour/tint control you indeed have to compromise between colours. However, with some OLED TVs it it recommended not to adjust the colours.

There’s no need to manually calculate the 21% red etc as they are all factored in HCFR’s target values.
 

·
Registered
Joined
·
18 Posts
It would be easier for others to help with troubleshooting if you attach the HCFR data file (*.chc, zipped to upload).

For accurate colours you need CMS controls for each colour. If the TV only has a global colour/tint control you indeed have to compromise between colours. However, with some OLED TVs it it recommended not to adjust the colours.

There’s no need to manually calculate the 21% red etc as they are all factored in HCFR’s target values.
Thanks for the reply! I have a global colour control but separate tint and saturation Controls. The datafile is attached.
I did the run with 100% color patterns and tint/saturation controlls untouched.


Maybe the best thing is to back down color a bit to get the best average value between red, gren and blue and just leave it there. And don´t touch the tint and saturation controllers for each color?
 

Attachments

·
Registered
Joined
·
8,836 Posts
Thanks for the reply! I have a global colour control but separate tint and saturation Controls. The datafile is attached.
I did the run with 100% color patterns and tint/saturation controlls untouched.


Maybe the best thing is to back down color a bit to get the best average value between red, gren and blue and just leave it there. And don´t touch the tint and saturation controllers for each color?
Is the TV in standard (non-HDR) mode? The colours are slightly oversaturated.

If you look at the green and cyan delta-L, they are 40% too high. I would reset all colour controls.
 

Attachments

·
Registered
Joined
·
18 Posts
Is the TV in standard (non-HDR) mode? The colours are slightly oversaturated.

If you look at the green and cyan delta-L, they are 40% too high. I would reset all colour controls.
Thanks for your time aqnd effort to help me! :)


Yes I calibrate in SDR mode. I want to get this mode right before i goes in to calibrate the HDR mode.
I did a new run now with everything at set to factory default. Global colour control at 50 (goes from 0-100), hue and saturation set to 0 for every colour (both controls goes from -15 to +15).


I did one run with 100% colour and one run with 75% color. Both files attached. Colors seems to be off even with this settings.
 

Attachments

·
Registered
Joined
·
8,836 Posts
Thanks for your time aqnd effort to help me! :)


Yes I calibrate in SDR mode. I want to get this mode right before i goes in to calibrate the HDR mode.
I did a new run now with everything at set to factory default. Global colour control at 50 (goes from 0-100), hue and saturation set to 0 for every colour (both controls goes from -15 to +15).


I did one run with 100% colour and one run with 75% color. Both files attached. Colors seems to be off even with this settings.
Which pattern disc did you use? AVSHD709 does not use 75%/75%, IIRC.

For initial set up I would actually use HCFR's internal pattern, as that ensures there's no mismatch between the patterns and what HCFR expects.
 

·
Registered
Joined
·
18 Posts
Thanks both Dominic Chan and SirMaster for your help! :)

I did a new run today with HCFR at pattern generator and now everything lines up much better (even if not all values are on the green side), so i think there was a little mismatch between my USB patterns and what HCFR expected. See the attached file.

Some of the colours could line upp better with som adjustments in tint/saturation, but I feel a bit unsecure about touching the tint/saturation buttons or not. Because I have heard that touching these buttons sometimes can introduce other problems on the Oled.

So maybe this is as good as it gets?
 

Attachments

·
Registered
Joined
·
4,091 Posts
HDR calibration on this TV using HCFR is driving me nuts.



Does anyone have a secret to getting the gray scale RGB calibration dialed in? The TV CMS 11 point calibration to HDR pattern mapping jumps all over the place depending on whether the CMS screen is up, and even then much of the time it doesn’t map to the HCFR reference patterns. This makes HCFR calibration super confusing.



For example, here’s how the HDR patterns (streamed through a Firestick 4k via Kodi) map to the reference patterns on HCFR.



HDR patterns HCFR reference patterns

5% 0%

10% 10%

20% 30%

30% 45%

40% 55%

50% 70%

60% 80%

70% 100%

80% 100%

90% 100%

100% 100%



But in order to get the probe to respond to the TV’s CMS RGB changes for a given test pattern, the following mappings apply.



CMS HDR Patterns

5% 25%

10% 35%

20% 45%

30% 55%

40% 65%

50% 70%

60% 75%

70% 80%

80% nothing works

90% nothing works

100% 100%



Having to deal with two sets of mappings is making proper calibration almost impossible, at least with the approach I’m attempting in HCFR. It’s very possible I’m doing something fundamentally wrong as this is the first time I’ve attempted calibrating an HDR TV.
Please discuss the topic in this thread.
 
12221 - 12240 of 12733 Posts
Top