[Bf-blender-cvs] SVN commit: /data/svn/bf-blender [46974] trunk/blender/source/blender/ compositor/operations: Fix new compositor color to value/ bw conversion working different than it did

[Brecht Van Lommel]

Revision: 46974
          http://projects.blender.org/scm/viewvc.php?view=rev&root=bf-blender&revision=46974
Author:   blendix
Date:     2012-05-24 14:19:20 +0000 (Thu, 24 May 2012)
Log Message:
-----------
Fix new compositor color to value/bw conversion working different than it did
before. It doesn't make much sense to always multiply with alpha, especially
if it's a premul image that already has it multiplied in.

Patch by Irie Shinsuke.

Modified Paths:
--------------
    trunk/blender/source/blender/compositor/operations/COM_ConvertColorToBWOperation.cpp
    trunk/blender/source/blender/compositor/operations/COM_ConvertColourToValueProg.cpp

Modified: trunk/blender/source/blender/compositor/operations/COM_ConvertColorToBWOperation.cpp
===================================================================
--- trunk/blender/source/blender/compositor/operations/COM_ConvertColorToBWOperation.cpp	2012-05-24 14:06:10 UTC (rev 46973)
+++ trunk/blender/source/blender/compositor/operations/COM_ConvertColorToBWOperation.cpp	2012-05-24 14:19:20 UTC (rev 46974)
@@ -38,7 +38,7 @@
 {
 	float inputColor[4];
 	inputOperation->read(&inputColor[0], x, y, sampler, inputBuffers);
-	outputValue[0] = (inputColor[0]*0.35f + inputColor[1]*0.45f + inputColor[2]*0.2f)*inputColor[3];
+	outputValue[0] = inputColor[0]*0.35f + inputColor[1]*0.45f + inputColor[2]*0.2f;
 }
 
 void ConvertColorToBWOperation::deinitExecution()

Modified: trunk/blender/source/blender/compositor/operations/COM_ConvertColourToValueProg.cpp
===================================================================
--- trunk/blender/source/blender/compositor/operations/COM_ConvertColourToValueProg.cpp	2012-05-24 14:06:10 UTC (rev 46973)
+++ trunk/blender/source/blender/compositor/operations/COM_ConvertColourToValueProg.cpp	2012-05-24 14:19:20 UTC (rev 46974)
@@ -38,7 +38,7 @@
 {
 	float inputColor[4];
 	inputOperation->read(&inputColor[0], x, y, sampler, inputBuffers);
-	outputValue[0] = ((inputColor[0] + inputColor[1] + inputColor[2])/3.0f)*inputColor[3];
+	outputValue[0] = (inputColor[0] + inputColor[1] + inputColor[2])/3.0f;
 }
 
 void ConvertColourToValueProg::deinitExecution()