Syd said: > Just a question to all... > After the issue with the dhc-2 panel , should I be modelling for the > latest plib? > The ' drawing order ' of objects with transparent textures is different > ...so the .ac files have to be modified ... if that makes sense. I > haven't found a way to make the models display correctly between the > different versions. > So should I start using the latest plib and hope everyone upgrades ??? > > P.S. ...redoing the dhc-2 gauges ...it should work with any plib version > .... just more polygons to deal with :( >
This has nothing to do with it. Your gauges look great. The transparencies go on the stack last (or at least after anything that must show behind them) unless there's a sort being done. If putting the transparency last in the file breaks something then there is a bug in plib, not your model. Manual fixed sorting as opposed to dynamically sorting is often problematic. Especially in the cases where you might be looking through multiple layers of alpha textures in different orders depending on viewpoint. If you manually sort and then someone runs a development version of plib that sorts dynamically you should still be fine (unless the plib algorithm is broken...which is not your problem :-)). Basically, the reason the transparencies need to be last (top of the stack) is the rendering starts at the top of the stack, last ac3d object is top of the stack (for that model). The renderer needs to know if there's an alpha object/texture _before_ processing the objects that are behind the alpha object so it can blend their colors in as it goes (which is what gives the "glass" effect). Normal behavior is to assume anything closer to the eyepoint simply obscures what is behind it which is why the guage faces disappeared when the panel glass came after the guages in the process. As I mentioned before, always keep in mind that the order of objects in the ac3d file is the opposite of the order they are rendered. That is why the alphas have to be after the objects behind them in the ac3d file. Best, Jim _______________________________________________ Flightgear-devel mailing list Flightgear-devel@flightgear.org http://mail.flightgear.org/mailman/listinfo/flightgear-devel 2f585eeea02e2c79d7b1d8c4963bae2d
