Hi all I'm wondering if anyone could give me some advice - I'm working on a 3D viewer application that makes use of OpenSceneGraph. At the moment I'm loading large 2D slices of data into textures, and then using a fragment shader to colourise them.
The slices of data are big (up to 150MB), and consist of 32-bit floating point data. I'd like to split the hunks of data up to reduce the amount of data being paged into the video card when a frame is rendered. I'm wondering what the best approach might be. At the moment I've got a single osg::Geometry created by osg::createTexturedQuadGeometry(). I initially thought I could keep this single geometry and just pass multiple textures through to my shader program, but it seems that sampler2DArray isn't supported on all the cards I need to support. I'm currently thinking of using multiple geometries for say 512x512 sized textures of my data, and overlapping them to avoid interpolation issues on the boundaries. I'm worried about seams appearing and other artifacts though, so I thought I'd ask the mailing list for advice before writing too much code! Thanks for any help, Cheers Grahame _______________________________________________ osg-users mailing list [email protected] http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
