Re: [darktable-devel] Basecurve Revisited

2013-09-06 Thread Torsten Bronger
Hallöchen! David Vincent-Jones writes: > [...] > > The image that I used probably had a better grey distribution than > samples previously offered on this subject ... however I will > remake based on a much more normal scene. I have never smoothed, yet my plots only looked this way for rare grey

Re: [darktable-devel] Basecurve Revisited

2013-09-06 Thread David Vincent-Jones
On 13-09-06 07:49 PM, Torsten Bronger wrote: > Hallöchen! > > David Vincent-Jones writes: > >> [...] >> >> A plot that is typical of my findings are attached. Note: the >> absence of some of the 256 values causes the variation and I have >> not attempted any smoothing. > Your analysis that some va

Re: [darktable-devel] Basecurve Revisited

2013-09-06 Thread Torsten Bronger
Hallöchen! David Vincent-Jones writes: > [...] > > A plot that is typical of my findings are attached. Note: the > absence of some of the 256 values causes the variation and I have > not attempted any smoothing. Your analysis that some values occur very rarely is probably correct. This *may* ma

[darktable-devel] Darktable app icon proposal

2013-09-06 Thread Jakub Steiner
Howdy Darktable devs, application icons are shown in larger sizes on today's systems than in the past. Within GNOME we present apps in 96px+ size and with the new Software app, we'd like to present Darktable in the featured section[1][2]. The current app icon is lacking though, so I took the libert

[darktable-devel] Basecurve Revisited

2013-09-06 Thread David Vincent-Jones
I decided to revisit the basecurve question prior to making any modifications on my system. The tests were made using a Canon 50D camera shooting a raw/jpg combination. I did have full dynamic range in the raw file. I extracted 256 test values (0 -255) from the raw file and compared them to t

Re: [darktable-devel] cancel export - 2nd try

2013-09-06 Thread johannes hanika
imo this is a bug in views/view.c which cleans up all the modules when switching views, regardless whether or not they will be present in the next view. should take an hour of concentrated code reading to change that.. until that is done you'll have to use kill -9 or something mean like that. -jo