Difference is that in first variant you attached only image to camera, and it getting updated by copying framebuffer object data into image, then texture gets updated from that image, and framebuffer object most likely will be 8-bit per channel. In second variant u have both texture and image attached, camera renders to attached texture, then result gets copyed (with conversion couse texture is 16bit float, and image unsigned byte) to attached image and then texture trying to be updated from that image with same data that it already contains but in another format (not sure about last though, but this can cause that error\warning). And in secont variant framebuffer object will be 16-bit float per channel, then it will be converted to your image internal format.
If u need osg::Image to work with, u need to attach image to camera (not sure about do you need to attach texture, but if you must, then DONT do setImage to your image on that texture). If u need osg::Texture, then dont even create image, texture manages fbos urself Cheers, Sergey ------------------ Read this topic online here: http://forum.openscenegraph.org/viewtopic.php?p=31556#31556 _______________________________________________ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
