On Sunday, 16 August 2015 22:29:42 CEST, Pedro Côrte-Real wrote: > And now you have a bunch of output files, some straight from the > camera, some processed from raw. Colors/curves/noise/etc will not > match so this is only really workable if you don't need consistency in > your output. > > Since to be able to process raws you need to be happy with how the raw > conversion output looks most people will just get that working well > and then ignore what the camera does by default.
A big +1 here from my side to what Pedro said. I was shooting RAW+JPEG, too, back in the times when I was still using Canon's DPP. It made some sense during these times because the output of the in-camera JPEG and the output of the vendor's software were pretty close to each other. My reasoning was "let's just have the RAW as a backup in case I screwed my settings", and I would typically use the JPEG in the most cases where I was hapy with the unchanged output. The thing is, any third-party RAW convertor will typically produce output which *is* different from the in-camera conversion, even after you spent hours of your time adjusting the curves and what not. This is inevitable because there's no such thing as a "correct" conversion, and each software package for RAW conversion works in a slightly different way. If you like your camera's JPEGs, by all means try to tweak your Darktable's styling to *approach* what the built-in software in the camera outputs, but be advised that you can only get more or less close approximation, not a 1:1 match. Since these times when I was happy with whatever the camera generated, however, my perceived-quality requirements have changed. During the time, I found out that I rely on the RAW data more and more -- if only for tweaking the subtle white balance changes which *always* happen for me when I'm shooting outdoor events where action moves between areas where lightning is a tiny bit different, or when my subjects just move their heads a little bit. I also like drawn masks and use them to bring up details and shadows in slightly shaded faces, something which I get quite a lot due to my photojurnalistic style. That all means that while I could still probably work with JPEGs in maybe the majority of my shots, the increased complexity (and time!, and storage costs) do not make such a workflow compeling for me anymore. It seems that this is actually also what most of the photographers who shoot RAW are doing these days. So I simply rely on the RAWs, and I treat all of my pictures the same. I use DT's views to perform selection of the keepers, and when I'm happy with what I have, I press "export". There is no need to keep a mental track of "do I use the original JPEG, or the developed version?", because I develop -- or rather I let the computer develop -- everything in the same way. My suggestion to the original poster is to follow suit here. In my experience, it has made the whole workflow easier for me. I was also able to get better performance out of my cameras now that they are writing less data during action moments -- yes, I use bursts every now and then. Hope this helps, Jan -- Trojitá, a fast Qt IMAP e-mail client -- http://trojita.flaska.net/ ------------------------------------------------------------------------------ _______________________________________________ Darktable-users mailing list Darktable-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/darktable-users
