Offering up a patch for review that fixes images with premultiplied alpha being incorrectly rendered in the UV Image Editor window.
Discussed at length with Lukas Tonne. It seems that a CompBuf doesn't retain any information regarding the type of alpha being passed through it. As a result, it is impossible to detect currently. Even then, it is likely prudent to leave this decision up to the artist. The patch includes a toggle that appears when an artist chooses to display the image with alpha in the UV Image Editor. It has a separate button with a custom designed icon. The patch turns on the proper glBlendFunc mode to GL_ONE for associated alpha images and GL_SRC_ALPHA for unassociated alpha. It should also be noted that for some legacy ATI card implementation reasoning, that by default the viewer only displays 8bpc sources. The code is more or less intact but commented out for float support. It may be worth considering either offering up a WITH_FLOAT_GL in the compile options or a GL depth selection between 8bpc and 32bpc float. With respect, TJS http://projects.blender.org/tracker/index.php?func=detail&aid=28782&group_id=9&atid=127
_______________________________________________ Bf-committers mailing list [email protected] http://lists.blender.org/mailman/listinfo/bf-committers
