Hello,
I know my question has been treated before, but the answer did not satisfy me completely.
I am grading my films in Resolve with the color management activated, with an output color space of rec709 gamma 2.4.
Since recently, i did not give a second tought when doing a DCP with DCP-o-matic : it recognized my source as being rec709.
But on the rare occasions when i saw the results in the theater, it seemed a little bright. Then i discovered you can choose an entry gamma for rec709 2.4 instead of 2.2, which converted the setting in BT 1886.
My question : i have now many films, graded very low-key in resolve. I know a shift at the bottom of the luminance could spoil everything for the audience, and i am not sure i will be able to check the DCP in a proper theater.
My question : is the 2.4 of BT1886 behaving exactly like the gamma 2.4 of rec709, expecially in the dark tones where it is crucial?
Are the graphs of gamma 2.2 and gamma 2.4 normalized and constant data, or are there differences between srgb, rec709, BT 1886 ? Is gamma 2.4 the same courb between rec709 and BT1886? Is gamma 2.2 the same courb between sRGB and rec709?
What choices did DCP-O-matic made when coding?
I am asking that because in the color settings box of DCP-o-matic, i do not know how to interpret some choices.
There is a choice for the input gamma between "simple gamma, or simple gamma linearised in the dark tones, or... What changes does it cause? In the dark tones?
There is also a number for entry power, i do not know its influence.
Also for the YUV to RGB conversion, at REC709, does it stay at rec709 for BT1886 ?
Sorry for my approximative english by the way, and thanks in advance for your help !
Using BT 1886 for rec709 gamma 2.4 sources
-
Carsten
- Posts: 2804
- Joined: Tue Apr 15, 2014 9:11 pm
- Location: Germany
Re: Using BT 1886 for rec709 gamma 2.4 sources
Can you show us your export settings in Resolve? Do you use data or video levels?
On what type of machine are you using Resolve?
Are you sure your monitor/viewer is actually calibrated? What type of monitor do you use?
In general, the difference between the rec.709 and BT.1886 preset in DCP-o-matic should only be 2.2 vs. 2.4
You can see the difference when opening the waveform monitor with some relevant content frame in the viewer. Then switch between rec.709 and BT.1886.
BMD will tell you that you need a Decklink/Monitor card and a calibrated monitor to have proper color space monitoring.
- Carsten
On what type of machine are you using Resolve?
Are you sure your monitor/viewer is actually calibrated? What type of monitor do you use?
In general, the difference between the rec.709 and BT.1886 preset in DCP-o-matic should only be 2.2 vs. 2.4
You can see the difference when opening the waveform monitor with some relevant content frame in the viewer. Then switch between rec.709 and BT.1886.
BMD will tell you that you need a Decklink/Monitor card and a calibrated monitor to have proper color space monitoring.
- Carsten
-
wragge75
- Posts: 3
- Joined: Tue Feb 16, 2021 9:12 am
Re: Using BT 1886 for rec709 gamma 2.4 sources
My export setting in Resolve is quicktime prores 4444.
Data levels are auto.
Color space tag and gamma tag are the same as project.
The project is color managed, and the three spaces (input, timeline, output) are rec709 gamma 2.4.
My monitor is an eizo CG246 calibrated for rec709 - gamma 2.4 -100cd/m2 - 6500K .
When I re-import in resolve the DCP-BT1886 , interpreting it as being in DCI X'Y'Z' , the images of the DCP and the prores are just the same.
If i do the same with a DCP-Rec709, the DCP is brighter than the source in Resolve, which is coherent with the fact it is interpreted as gamma 2.2 (classic rec709) in DCP-o-matic.
So my workflow seems now coherent to me, i just wanted to check with you if i am doing right, if the image graded in Resolve will look exactly the same when the DCP-BT1886 is projected, especially in the dark tones.
My other question is :
- what is the difference between gamma simple, gamma linearised for dark tones, and S-gamut3 for the BT1886 setting? how does it influence the image?
What is "input power" for (it is changeable for srgb)?
Data levels are auto.
Color space tag and gamma tag are the same as project.
The project is color managed, and the three spaces (input, timeline, output) are rec709 gamma 2.4.
My monitor is an eizo CG246 calibrated for rec709 - gamma 2.4 -100cd/m2 - 6500K .
When I re-import in resolve the DCP-BT1886 , interpreting it as being in DCI X'Y'Z' , the images of the DCP and the prores are just the same.
If i do the same with a DCP-Rec709, the DCP is brighter than the source in Resolve, which is coherent with the fact it is interpreted as gamma 2.2 (classic rec709) in DCP-o-matic.
So my workflow seems now coherent to me, i just wanted to check with you if i am doing right, if the image graded in Resolve will look exactly the same when the DCP-BT1886 is projected, especially in the dark tones.
My other question is :
- what is the difference between gamma simple, gamma linearised for dark tones, and S-gamut3 for the BT1886 setting? how does it influence the image?
What is "input power" for (it is changeable for srgb)?
-
Carsten
- Posts: 2804
- Joined: Tue Apr 15, 2014 9:11 pm
- Location: Germany
Re: Using BT 1886 for rec709 gamma 2.4 sources
Not sure if I can follow, but as far as I know, Prores 4:4:4:4 uses data/full levels. This could result in lifted blacks when this file is interpreted as video levels, as is 'normal' for other video files.
These gamma settings are universal, not just für BT.1886. The correct setting for BT.1886 should be gamma 2.4, linearized for dark tones:
https://kb.portrait.com/help/bt-1886-10 ... 10-answers
Read page 35 here: https://tech.ebu.ch/docs/tech/tech3320.pdf
The proper BT.1886 gamma is not just a straight gamma, but one that deals with dark tones in a special way. It's a combination of a linear and power curve.
S-Gamut3 is a special gamut used in certain (Sony) cameras. While these files should normally go through a color grading process before being converted to a DCP, someone convinced Carl that it could be handy to use S-Gamut3 Files directly within DCP-o-matic.
Techwise, I guess we should not use the term rec.709, but, there are historical reasons to keep the rec.709 preset with it's 2.2 gamma. All modern cameras will work in BT.1886, but will often still quote rec.709 in manuals/specs.
- Carsten
These gamma settings are universal, not just für BT.1886. The correct setting for BT.1886 should be gamma 2.4, linearized for dark tones:
https://kb.portrait.com/help/bt-1886-10 ... 10-answers
Read page 35 here: https://tech.ebu.ch/docs/tech/tech3320.pdf
The proper BT.1886 gamma is not just a straight gamma, but one that deals with dark tones in a special way. It's a combination of a linear and power curve.
S-Gamut3 is a special gamut used in certain (Sony) cameras. While these files should normally go through a color grading process before being converted to a DCP, someone convinced Carl that it could be handy to use S-Gamut3 Files directly within DCP-o-matic.
Techwise, I guess we should not use the term rec.709, but, there are historical reasons to keep the rec.709 preset with it's 2.2 gamma. All modern cameras will work in BT.1886, but will often still quote rec.709 in manuals/specs.
- Carsten