|
|
I'm using After Effects for making a tiff sequence, but again, the problem is the wrong colorspace.
What are the correct settings to get the CIE XYZ as output?
If thats what I need?
Just read the walkthrough...this is seriously one of the coolest posts I have ever read on REDUSER Leerafel. Well done.
I especially like the fact that I have everything I need at hand beside the Easy DPX license.
Now, to find a booth manager in London whose feeling generous...
Doing here my first conversions to DCP for a local indie feature.
Got my first pakage with Opencinematools today.
Wolfgang, Pavel, Chris etc have given an invaluable information here.
But one aspect worries me.
With the ImageMagick sample the conversion is from sRGB to X'Y'Z' P3.
-gamma 0.454545 -recolor "0.4124564 0.3575761 0.1804375 0.2126729 0.7151522 0.0721750 0.0193339 0.1191920 0.9503041"
What coeficents shoud I use if the source is Rec.709?
Got the original as BlackMagic 10-bit uncompressed video and converted to DPX files with Combustion. Gamma was left as original.
Thanks
Priit
priitp,
see http://github.com/wolfgangw/digital_...-xyz_matrix.rb for the exact numbers and source code to check those numbers.
Thanks Wolfgang.
Have see it earlier but forgot already.
What about gamma?
As Rec.709 gamma has a linear part it is not so easy to deal it with Imagemagic convert -gamma.
Let's say we use one value for gamma but what might it be?
I remember Michael Most wrote somewhere that "Rec709 is described as having a specific gamma that is somewhere around 2.3 or so (but not exactly, the lower end of the curve is a bit different)"
So will this be the best bet? -gamma 0.434783
It seems to me it would be good to have a 1D LUT for linearisation.
I have to check the -clut possibilties.
Priit
I might have said that, and it's not inaccurate. The real problem with Red709 is that there is no definition for a reference display, in fact, there is no display definition at all, as opposed to, say, the DCI P3 specification. I'm told Charles Poynton is currently working on trying to rectify that, among other things.
Thanks Michael for the input.
Unfortunatelly ITU Rec.709 standard is not freely available.
From Cine-tal forum here found this formula:
-----------------------------------------------------------------------
Overall opto-electronic transfer characteristics at source:
V = (1.099 * L^0.45) - 0.099 for 1 ≥ L ≥ 0.018
V = 4.5 * L for 0.018 > L ≥ 0
where:
L: luminance of the image 0 ≤ L ≤ 1
V: corresponding electrical signal
-------------------------------------------------------------
Not sure how reliable is this source.
Meaning to make a -clut image based on it.
Priit
| « Previous Thread | Next Thread » |