Robert (and others), I'm using the HDR loader to load HDRE images; if you ask it convert to floating point, it converts to 4 byte floats - i.e. RGB32F.
Unfortunately, I want/need linear filtering on the texture, so need to "drop" this image down to RGB16F. I've tried texture->setSourceFormat(GL_RGB32F), and texture->setInternalFormat(GL_RGB16F), when attaching the image, but still no LINEAR filtering. To tell you the truth, I'm still pretty puzzled about what texture->setSourceFormat actually does. This may be due to my card being old and infirm (does anybody know if an ATI Radeon 9600 supports GL_LINEAR on GL_RGB16F)? Lastly, I have to do some image manipulation on the source image, so need to do read/write half floats, something like (for reading) osg::Vec3 colour; half* ptr = (half*) image->data(col,row); colour[0] = (float) *(ptr++); colour[1] = (float) *(ptr++); colour[2] = (float) *ptr; and a similar operation in reverse for writing. Am I missing some easy trick here? David
_______________________________________________ osg-users mailing list [email protected] http://openscenegraph.net/mailman/listinfo/osg-users http://www.openscenegraph.org/
