Hi, this is a request for comments on some changes I am considering to work on.
Concretely I'd like to change the way softproofing and gamut checking (SP/GC) is handled on the GUI side and in the pixelpipe as at the moment they are not very convenient IMHO. And being at that I'd also like to change the way how the display profile and display intent can be selected by the user. With the current status I see the following issues: A) SP/GC can only be activated by keyboard shortcuts (you need to remember those). B) Only if the GUI of the "output profile" module is in focus you can see if SP/GC are active. C) Selecting the softproofing profile is a per-image setting. You either need to select the profile individually for each image or you need to generate a preset/style which is automatically applied to images - inconvenient if you want to quickly test different profiles or different images. D) As SP/GC are part of the "output profile" module, both actions happen late in the pixelpipe but not at the end - there are some modules after that (namely the channel mixer) which can render both actions useless. E) SP/GC are done on Lab data. Very often the Lab values of dark shadows tend to be unbounded. The gamut checking functionality of lcms2 is very fussy and will mark those areas as out-of-gamut although in fact those values are not of a big concern once converted to RGB and clamped to the permissible range - the sheer level of false alarms tends to hide the real problems. F) For the display profile and the display intent the same issues apply as under (C). My idea is to do the following: 1) Find a way to activate SP/GC in a way similar as we do with over-/underexposure, i.e. we should have an activation button at the lower edge of the image pane (this addresses A and B). The button should allow to toggle the three display features (normal/softproof/gamut-check) and change its appearance accordingly, like [ ] .. [S] .. [G] 2) Right clicking the button should open a menu to select the softproofing profile as well as the display profile and display intent (this should address C and F). One of the combobox options for display profile and display intent could be "image settings" to keep compatibility with legacy history stacks (similar to what we do in the "export selected" panel). It's understood that on systems with a proper color management setup, manually changing the display profile is normally not needed. 3) SP/GC functionality in the pixelpipe should be moved into a separate hidden module very much at the end of the pixelpipe. We would deal with this in a similar way as we handle over-/underexposure. We would apply SP/GC on the final output RGB data rather than the intermediate Lab color space (this addresses D and E). That's the idea. Any comments, concerns and proposals? Best wishes Ulrich ------------------------------------------------------------------------------ Dive into the World of Parallel Programming The Go Parallel Website, sponsored by Intel and developed in partnership with Slashdot Media, is your hub for all things parallel software development, from weekly thought leadership blogs to news, videos, case studies, tutorials and more. Take a look and join the conversation now. http://goparallel.sourceforge.net/ _______________________________________________ darktable-devel mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/darktable-devel
