Simon Hammett <[email protected]> wrote: > > Well .lwo was the first format I found that worked and it's binary so > the models load quickly. > The apps I work on, load lots of stuff outta dbs, run third party dlls > & black boxes etc > on start up, so they are already irritatingly slow to start up.
Well, you export to .osg and then you can always convert to binary .ive once you are happy with the model. > There's also the problem with setting up new machines to run a project; > It takes ages to join the code co-op projects, install boost & other > libraries, black boxes, etc, etc. Adding another step of installing > blender and then getting > the export script as well just makes things take longer. Hmm, how is that different from going the .lwo route? You need Blender installed and export to .lwo as well, no? > Nice, thanks for the info, I'll look in to that. > > I'd prefer to load models directly out of the .blend file, > it would save mucking about with exporting. You can actually automate the export - e.g. the Soya3D engine does it in that way that whenever the user requests a .blend file to be loaded, it spawns Blender in the background calling it with a command line argument to load the model and call the export script. The result is then cached and the model is re-exported only when the .blend file is newer than the cached data. The user never has to tinker with the exporter, they just notice Blender starting once when the model is exported for the first time. Jan
signature.asc
Description: This is a digitally signed message part.
_______________________________________________ osg-users mailing list [email protected] http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
