hello good people! i need a bit of advice ragarding some more advanced rendering techniques. advanced for me, anyway.
i am trying to build a simple ray tracing application on top of the basic osgviewer program. to start off, i am trying to "look" at the loaded model and decide where to cast rays, e.g. i don't want to cast any rays where the model would be missed. i was thinking of using the stencil buffer and the first thing i need is to draw say white pixels where the model is and say black pixels where the model is not. then display this buffer as the frame buffer, you know what i mean. is this possible? i remember doing something similar with GLUT long, long time ago. //if i was able to do this kind of evaluation on the per-pixel basis, i think, i could add some more advanced features like the color of the first hit object (instead of white) and then maybe continue with some real ray tracing, as i mention in the following lines.// later on, according to where the model would be hit (the white pixels), i would cast rays using line intersectors. haven't thought about details here yet. it would be kind of any of you if you could give me some pointers, especially for the first part with the buffers that i am currently dealing with - is it even possible? is my concept good? - for my knowledge in this area has gotten very poor :( and i am also new to the osg. thank you very much in advance. Andrey ------------------ Read this topic online here: http://forum.openscenegraph.org/viewtopic.php?p=44140#44140 _______________________________________________ osg-users mailing list [email protected] http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
