On 19 Jul 2012, at 14:32, Tim Moore wrote: >> Do we really not run the terrain fragment shader when the terrain is seen >> through the cockpit floor (my system seems to slow down even though no >> terrain is seen in the event and the framerate responds to my shader quality >> settings)? > > That would be nice, but the near/far camera defeats that optimization :(
Which is very unfortunate, since in the common case it would allow a huge amount of tile/terrain pixels to be dumped. Could we use a buffer copying to trick to initialise a stencil plane on the far camera based on the near-camera Z-buffer? I've no idea what the penalty of a z-buffer read and stencil buffer write on the two camera would look like, but my impression is 'out of the window' pixels are waaaay more expensive in terms of shaders and fog / distance effects than in-cockpit pixels. > That's a different issue. Earth centric coordinates can't be stored in > single precision floating point at the required precision, so each > tile has a local coordinate system. A "world wide" noise basis would > be a good project. Right, we need distinct coordinates per tile I'm afraid, no working around that. > > The depth-pass-only pass is a well known optimization, but the fact it > is not helping implies that our bottleneck is not in fragment > processing. I've suspected for years that it lies on the CPU, and have > been trying to optimize our scene graph a bit... I have some plans in that direction post 2.8, especially to flatten the LOD quadtrees and transforms of the tiles. Each tile will get some top-level LOD groups for all objects (shared and static). I'm hoping in combination with the LOD-scale function in OSG, this will mean we can automatically drop random tress / building and STG objects to keep frame-rate up. (as an optional mode of course!) Another goal is to add more node bits (and a GUI dialog to control them) so various categories of objects can be disabled during the update pass. This will mean the direct hit of, say, AI models vs particles vs random trees can be measured. Of course it won't account for resources (memory, textures) burned by such things, but would still help different people identify slowness on their setups. James ------------------------------------------------------------------------------ Live Security Virtual Conference Exclusive live event will cover all the ways today's security and threat landscape has changed and how IT managers can respond. Discussions will include endpoint security, mobile security and the latest in malware threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/ _______________________________________________ Flightgear-devel mailing list Flightgear-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/flightgear-devel
