On Sunday, 2 January 2005 18:03, Christian Mayer wrote: > I see no benefit in adding an dependancy to a library that effectively > can do the same as OpenGL - but only in software.
The difference is a powerful text and vector library vs OpenGL primitives. Have you ever tried rendering true type fonts in OpenGL? It's a pain in the ass! Under windows you have to use wiggle functions and I can't remember how it's done under other OS's. > If OpenGL is too complicated for some cases, we can encapsulate the > necessary functions in C/C++ code and offer that function. I think that would be a good option. I think a panel designer should be given a canvas/texture that they can "paint" on with easy to use text and vector functions. The canvas and painting should be defined in 2D pixel co-ords. MSFS actually do it this way using GDI+ (software rendered canvas) and although I don't feel we need to be copycats this method works well and keeps it simple for panel designers. > An glass cockpit can be implemented by rendering the display content to > an texture and using that dynamic texture in the 3D cockpit. Yeah I know about off screen rendering to textures but I don't know of anyone who is willing to implement it for us. There are several ways of doing it which nvidia have documented here : http://developer.nvidia.com/object/gdc_oglrtt.html Paul _______________________________________________ Flightgear-devel mailing list Flightgearemail@example.com http://mail.flightgear.org/mailman/listinfo/flightgear-devel 2f585eeea02e2c79d7b1d8c4963bae2d