Game Development Reference
In-Depth Information
mon sense” about what it is trying to do to avoid having to do an exhaustive search for
solutions to an existing problem. They are algorithms not based on rigid mathematical
formulas but more like neural networks; they must be trained to do what you want them
to. These algorithms decide which elements of the scene are eye dependent and which
are not in a process that occurs entirely in the render pipeline.
It is possible for the programmer to defeat the “common sense” rules the computer is
using in the pipeline so the method is not entirely fire-and-forget but does reduce a lot
of the workload for development. One of the biggest benefits for larger game studios is
that it avoids anyone having to reprogram existing game engines. By the same token, it
allows existing games to be easily played in stereoscopic 3D. All of the preceding rec‐
ommendations apply to passive stereoization except that there should be no user-
adjustable settings in the game. Passive stereoization relies on third-party profiles that
help the GPU do the work. The user will have set up a profile with whatever stereoization
software he or she is using, such as NVIDIA's 3D Vision. Other recommendations may
be specific to the stereoizer, and manufacturers usually publish a best practices guide.
The NVIDIA one is very helpful, and we recommend you read it if you are interested
in using stereoization in your games.
Search Nedrilad ::




Custom Search