Bug #12412

color balance broken

Added by Heiko Bauke over 1 year ago. Updated over 1 year ago.

Patch attached
Target version:
Start date:
Due date:
% Done:


Estimated time:
Affected Version:
git master branch
hardware architecture:


Some of the various sliders in the color balance module have very odd upper and/or lower bounds (e.g. -0.84172). In particular RGBL are broken sliders. The module is currently not usable.


#1 Updated by Heiko Bauke over 1 year ago

Furthermore, every value change triggers an additional entry in the history stack.

#2 Updated by Pascal Obry over 1 year ago

RGBL seems working on my side. The values are -0.5 +0.5 for most. Not tested all. Any more precision?

#3 Updated by Aurélien PIERRE over 1 year ago

Multiple entries in the history stack occur when you have inconsistent modules versions in the XMP stack (for example, going back from version n+1 of the IOP to version n). It's usually solved by compressing the history stack.

The upper bounds in RGB come from the conversions from hue and saturation to RGB at constant luminance. Since RGB and L channels are separated in the UI, the software ensures the sum of RGB parameters = 0 anytime to get a real separation between both. So, when you push the saturation at 1 on a secondary color (yellow, cyan, magenta), it's normal that the RGB bounds are ± 0.85. You are supposed to reach ± 1 only with primary colors.

Keep in mind that the RGB values are actually ratios that will be multiplied by the factor to give the actual parameters of the transfer function.

#4 Updated by Aurélien PIERRE over 1 year ago

Any news on that matter ?

#5 Updated by Heiko Bauke over 1 year ago

I think the problem has not yet been fixed completely. I still observe slider bounds which are asymmetric around 0. This is odd as the sliders color gradient as indicated in the ui is symmetric.

Additinally there is a further issue. In RGB mode sliders come in groups of three. If you push one in one direction the other two go into the opposite direction. This has the consequence that it is very difficult to reset all sliders within a group to zero (via double click). When you have pushed all three sliders away from zero and the try to reset all one after another the first slider jumps away from zero as soon as you reset the second one. A possible solution could be here to reset all three dependent sliders as soon as one is double clicked.

#6 Updated by Aurélien PIERRE over 1 year ago

  • % Done changed from 0 to 70
  • Status changed from New to Patch attached

Also available in: Atom PDF

Go to top