BrightContrast not working correctly with negative contrast
ClosedPublic
Actions

Authored by Philipp Oeser (lichtwerk) on Aug 1 2019, 7:15 PM.

Details

Reviewers

Sergey Sharybin (sergey)
Brecht Van Lommel (brecht)

Maniphest Tasks

T67808: Bright/Contrast node is broken

Commits

rBb54528fa1ed4: BrightContrast not working correctly with negative contrast

Summary

followup to rB8dd95abb2ff9 (which fixed this for the Compositor node),
turns out this was also wrong for the VSE modifier and in vertex color
operator.

also adjust min/max for VSE modifier
also guard against division by zero

Diff Detail

Repository: rB Blender

Event Timeline

Philipp Oeser (lichtwerk) created this revision.Aug 1 2019, 7:15 PM

Harbormaster completed remote builds in B4252: Diff 16777.Aug 1 2019, 7:15 PM

Brecht Van Lommel (brecht) requested changes to this revision.Aug 2 2019, 4:07 AM

Brecht Van Lommel (brecht) added inline comments.

source/blender/blenkernel/intern/seqmodifier.c
555	Better would be to do this. a = 1.0f / max_ff(a, FLT_EPSILON); Otherwise contrast values > 100 would start reducing the contrast, and values of `0 < a < FLT_EPSILON` would give more contrast then `a == 0`.
560–561	If we're changing code here anyway, this should be changed too since the formulas don't make sense for contrast < -100. a = max_ff(1.0f - delta * 2.0f, 0.0f);

This revision now requires changes to proceed.Aug 2 2019, 4:07 AM

Philipp Oeser (lichtwerk) added inline comments.Aug 2 2019, 9:35 AM

source/blender/blenkernel/intern/seqmodifier.c
555	Otherwise contrast values > 100 would start reducing the contrast Thats why I tweaked the valid range with `RNA_def_property_range`, but yeah, makes sense, will update...
560–561	this should be changed too since the formulas don't make sense for contrast < -100 Thats why I tweaked the valid range with `RNA_def_property_range`, but yeah, still makes sense, will update...