Hi, On Saturday, March 17, 2012 07:45:50 Renk Thorsten wrote: > > Before going into more detail, you might want to check with Mathias' > > recent work to integrate space-view into FlightGear. Otherwise you'll > > probably be doing duplicate work. > > Well, if Mathias could simply comment on the issue? What can we expect at > what timescale? (And how do you know these things - I've never read about > any plans to include orbital rendering here).
Once you understand how the scattering integrals work it's natural to compute them also from the outside of the atmosphere. There are several publications out there which descripe more or less reproducible how this works. Most of them sadly less reproducible. The best one (IMO) is currently one from a guy at inria who also provided a glsl demo with his publication. The demo is pretty raw, but can serve well as a reference how to do that. This demo relies on geometry shaders for almost no reason. Plenty of that stuff could be done in a way that also runs on older hardware compared to the demo. Also the demo relies on some undefined nonstandard behaviour of the nvidia driver. I have already found and fixed that here on my disk. Integrating this kind of stuff into a viewer is done by something very similar than Frederics work does. Render the scene into a fbo in this case without any fog and scattering and compute the atmosphereic scattering stuff in the final composition step. So it's just what I told in some mails: We need to do fog in a post processing step on the final image. You will need to render the scene and depth into an fbo and compute the scattering by evaluating the color and depth values for each fragment. As a simple consequence of this, you will just get spaceflight for free since you can mostly omit the sky dome (except may be the stars) with all the nice side effects you were working around. So martin probalbly followed this kind of mails to know this. My plans *were* to start going in rembrandts direction with this simple pulled out fog computation at first. Then once the fog is pulled out of the leaf models, replace this with a more sophisticated scattering computation. And then way later investigate what to do with what Frederic is working on. The plan was also to do that without disturbing the current fgfs implementation just in fgviewer. This area would have provided the playground to investigate how to cope with all the fbo multipass/compositing problems Frederic needs to tackle now. And in my plan, once this code is mature we might think about what to do with that in an end user distribution. Currently I for myself am more working on pulling out different components from the simulation. As a side effect I am doing that work on fgviewer which should do a standalone viewer component. I cannot recommend to use fgviewer for anything today because plenty of stuff is missing. But that was the place that I thought would be a good start to implement this kind of doing fog/atmospheric scattering. Now with rembrandt, lets see what happens and in which order. Also, since this would just happen at some time as a side effect of the central thing that I am working on, there is no schedule. Appart form that, since i am not working full time on this project, I cannot promise anything in terms of time. Greetings Mathias ------------------------------------------------------------------------------ This SF email is sponsosred by: Try Windows Azure free for 90 days Click Here http://p.sf.net/sfu/sfd2d-msazure _______________________________________________ Flightgear-devel mailing list Flightgear-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/flightgear-devel
