[解決済] what format do I export my RAW images to if I will combine with AVCHD format (.mts)
I haven't had a chance to use Cinelerra for a while and it is my first time that I shot images in RAW format and edited them in Darktable. I am wondering if exporting my pictures to TIFF 16 bit rec.709 is the best choice to conserve details and colourspace and add compositing effects in Cinelerra.?
I could choose rec 2020 or sRGB colourspace but I do not want to compress to jpeg at this stage. I have heard that PNG 16 bit has a much smaller file size than TIFF and is less lossy than jpeg, so it would be good to edit with in Cinelerra?
I wonder if I should work in RGBA float (with proxy files), would that match my sources better overall?
Is there other settings for the colour management that I should chance from default? I am not sure if choosing jpeg vs mpeg colour extension is an option for RGBA and which would be best?
The real color expert is @DeJay. This is an interesting discussion of his with user Glitterball3:
All the information on how CinGG relates to colors has been collected in the manual:
In theory to maintain as much color fidelity as possible you should use RGBA-FLOAT with Rec2020 and Color Range jpeg (full), but in practice you will not always get the correct results.
The Color Range should be varied depending on the source image, if you create your tiffs make sure they are Full Range and then set CinGG in jpeg color range.
I believe the main difference between png and tiff is the greater support for color models and the greater color depth achievable. The handling of metadata for printers and multipage, which contribute to the larger size of tiff, are useless for the purposes of CinGG.
Finally, it should be kept in mind that CinGG always shows images in the compositor in sRGB. If you can do a good conversion of the raw to this color space, then you might want to use it for the whole workflow in CinGG. At this point you are fine with PNGs that are 8 bits per channel and sRGB color space.
If you can get good results let us know we need many different experiences to improve the manual.
You have to decide whether to do a high quality workflow or reduce everything to avchd quality (which should be YUV 8 bits and 4.2.0 chroma subsampling). In the first case you have to transcode everything to a high quality intermediate (for example DNxHR 444 or image sequences) and use Tiff. Otherwise, png and sRGB are fine. Just be sure to keep the color spaces of the various media homogeneous, using the "ColorSpace" plugin if there is a need to conform them.