GameDev Art Pipelines
Probably the most important back-end part of GameDev for me would have to be the art pipeline! At present my role is not so much creating the art (models & textures) but being the techy that gets the finished art into the engine & makes sure its stable, optimal & engine ready.
This involves material setup, preempting sorting issues, billboards if any, animation sequence setup, LOD (level of detail) setup, placement of the instance into the world & finally a optimisation pass.
It all sounds so simple when written, and most times it is, for something small like a telephone box or arcade machine which just has a flickering computer monitor. But as soon as you beginning dealing with large worlds, and your art pipeline is not specifically setup to deal with large custom worlds it can become a nightmare. Cutting up a mesh that spans across a quarter of your world into 15, or more, smaller pieces presents huge art resource management issues, especially when each section has to be stored in its own 3D Max file due to exporter specifics. Thats the first part, secondly when loading these into the game engine, each piece has to be loaded, placed, rotated, setup to take the correct lighting, and finally checked, one at a time.
Now…. we’ve cut a larger mesh into smaller meshs for optimisation reasons, it becomes an alignment/placement problem (or management nightmare), lighting is sometimes affected and the gaps between meshes can become apparent.
As a ‘Game Techy Artist’ desperate to feel more like an ‘Artist’, the art to game engine pipeline can and normally does suck up most of my time. Time spent going back and forth, checking LOD setting, checking lighting, confirming collision polys haven’t freaked out and moved randomly. The slicker the pipeline, not only the better for the artist to work, but it would allow for better art creation, because time could be spent refining the art instead of fighting to get it to work in the engine itself.
Related Posts:
Random Rantings
Leave a Reply