Chris Burel
39 lines
2.3 KiB
Plaintext
39 lines
2.3 KiB
Plaintext
We used the sRGB values from 'Table 2: R'G'B coordinates of the ColorChecker' from this PDF:
|
|
http://www.babelcolor.com/index_htm_files/RGB%20Coordinates%20of%20the%20Macbeth%20ColorChecker.pdf
|
|
|
|
Specifically the second table labeled 'BabelColor Avg.' as this is a spectral average of 20 charts.
|
|
|
|
Those are the values typed into the color picker in sRGB swatches (stored in the material as linear RGB values)
|
|
|
|
Note: the colors displayed in that pdf table (on the left) do not seem to represent the actual sRGB values of the table (I found that odd)
|
|
|
|
since we are entering 1...255 values, instead of normalized 0...1 values, or 16-bit values, there is actualy some minor rounding errors.
|
|
|
|
The texture: ColorChecker_sRGB_from_Lab_16bit_AfterNov2014.tif is 16-bit color
|
|
But photoshop will quantize to 15+ bit color
|
|
Also photoshops color picker doesn't allow you to enter 16 bit color
|
|
|
|
So the avg sRGB values in the above pdf,
|
|
the values in the 16-bit macbeth texture,
|
|
and values color picked are slightly different
|
|
(but may or may not be noticable.)
|
|
|
|
That macbeth chart came from this more comprehensive site (which covers a lot of the variance):
|
|
http://www.babelcolor.com/colorchecker-2.htm#CCP2_images
|
|
|
|
The texture was converted to 1024x1024 based on the sRGB image 16-bit TIF in the section:
|
|
From X-Rite L*a*b* D50 (formulations AFTER Nov. 2014)
|
|
|
|
So, for now, that 16-bit .tif should be the most accurate - and in the future we may be able to enter more precise numeric values. This is after all meant to be a baseline and validate raw color versus color from texture.
|
|
|
|
Also i note, that the color swatch versus color texture material variants have a slight visual difference I am unable to
|
|
currently reconcile - could be a number of things, like a color transform somewhere, mathematical rounding error, etc.
|
|
|
|
Will need to come back and figure out where it is exactly.
|
|
|
|
Additional information about colors are here, along with what seems to be a pretty (not entirely) accurate set of macbeth charts for sRGB and ACES AP0 color spaces (they are crappy jpeg, so the AP0 isn't even a linear floating point image.)
|
|
|
|
http://www.nukepedia.com/gizmos/draw/x-rite-colorchecker-classic-2005-gretagmacbeth
|
|
|
|
Note: picking against the visual macbeth chart image versus the values in the original pdf there is a slight difference. I have leaned on the side of the spectral average in the original pdf.
|