Hi Robert, Thanks for the prompt response. I am not talking about compressed textures. By RGBA8 i mean typically R8G8B8A8. I believe that OpenGL also interprets in this way. Typically if we use an OpenGL App in which we have to read a PNG image that is in RGB8 format, we may use LibPNG library to read the data in the correct format, then it shall give a pointer to the data (let's say *Data).
Then we shall use glTexImage2D(GL_TEXTURE_2D, 0, GL_RGBA8, width, height, 0, GL_RGBA, GL_UNSIGNED_BYTE, Data). So technically OpenGL shall interpret as per input & render. I peeked into the osg::Texture and osg::Image classes, but could not make out whether they handle these data formats properly & pass on to OpenGL in the correct format as the image format (for example RGBA8).Again, OSG shall invoke LibPNG PlugIn for reading the data. It may need to tell that the data is in RGBA8 format. Or I might need to explicitly tell through osg::Image like (image->setImage(width, height, 1 ,GL_RGBA8,GL_RGBA,GL_UNSIGNED_BYTE, (unsigned char*)data,osg::Image::NO_DELETE)). If OSG does handle 8bit/32bit images properly then I should be able to save a lot of texture memory I believe. But I not able to figure out whether OSG does handle this correctly & pass on to OpenGL. Secondly, if it does correctly how shall I confirm this (may be as I posted earlier I need to see using an OpenGL debugger tool for texture memory usage). (I actually am working on OpenGL/OSG as I am a Graphics developer). Thank you! Cheers, umesh ------------------ Read this topic online here: http://forum.openscenegraph.org/viewtopic.php?p=62006#62006 _______________________________________________ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
