I haven't really seen any guidelines about efficient shader coding, but I've come across a few statements here and in the forum now and then, which I so far assumed to be true. I've now spent the last few days benchmarking my lightfield/terrain haze framework to see if I can't squeeze a bit more performance out. Since the results somewhat contradicted things I have assumed to be true before, I thought I might share my observations.
1) vertex shader is fast and efficient, fragment and geometry shaders are comparatively slow and cost performance I don't know who stated that here, but I intuitively assumed that this must be right. You can imagine that I was surprised that when I trivialized the fragment part of my code, the performance didn't change at all, but when I trivialized the vertex part, the performance doubled. So it seems like the whole performance drain of my code is caused by an overworked vertex shader. I then transferred the haze color computations to the fragment shader, which improved overall performance in my benchmarks by 20-40%. It seems the optimal results appear for a shared workload between vertex and fragment shader. Since for instance 3d clouds run almost exclusively via vertex shader, this may not be an otherwise irrelevant observation. 2) vertex count doesn't matter I've read that in the forum as advice to model-builders quite often. For all I can test, that's wrong. In order to get the same framerate in default scenery and custom Italy CORINE scenery, I have to set the visibility range ~a factor 10 different. That's consistent with a hundred times higher vertex count in the CORINE scenery, or with a factor 10 higher linear resolution of landcover and elevation data, which seems about right. So the vertex count more or less directly sets my framerate. This is also relevant for models, because having a cluster of ~ 5 AI planes (which happened to be stuck into each other) in my view or not caused a 25% change in framerate, so the vertex count of models matters compared with the vertex count of scenery and is not some insignificant correction. 3) shifting constant-per-frame computations out of the shader doesn't improve performance significantly. I suggested that here as an idea to improve performance of the cloud shaders. I was now able to construct a test case for the idea. The computation in question involved getting a per-frame constant from other constants, i.e. to compute sqrt(2 * hazeAlt * EarthRadius) where hazeAlt is the altitude up to which the haze extends which is passed as a uniform. Computing two of such square root expressions inside the vertex shader or not computing it makes the difference between 21 and 22 fps in a high vertex count benchmark test, i.e. 5%. Which is not that much in itself, but if we would find, track and eliminate all these occasions, the sum of such small improvements might well go into the 20-30%. I don't know how much of all that is specific for my system (NVIDIA GeForce 8600M GT) and how much is general, but my conclusion is that these things should be tested systematically, and I very much invite others to test the before/after version of the haze shader code. (I know that Vivian and Emilian are still working on separating the fog and light function out, but that's going to take, and if anyone could help me to get the correct xml conditionals so that the whole framework can be switched on and off along with the skydome shader, so that the whole thing can be committed and tested easily, that'd be terrific. I tried, but I keep messing the effect file up...) Cheers, * Thorsten ------------------------------------------------------------------------------ Try before you buy = See our experts in action! The most comprehensive online learning library for Microsoft developers is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3, Metro Style Apps, more. Free future releases when you subscribe now! http://p.sf.net/sfu/learndevnow-dev2 _______________________________________________ Flightgear-devel mailing list Flightgear-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/flightgear-devel
