Native Poser characters are often over 300,000 polygons: too high for most current real-time rendering VR platforms.
In order to explain this task properly, I will take the most complext position that we have chosen to have separate meshes for all body parts.
The fewer number of meshes your character is comprised of, the easier but less flexible this task will be.
Identify the priority of your model. For Sydney, her face and eyes are the most important, so these meshes must exhibit maximal quality (while maintaining visual consistency with the rest of the model and operating within technicl constraints).
This takes some time but it is worth it in the end. The basic process for each mesh is as follows:
Keep an eye on the polygon counter (Utilities tab -> More -> Polygon Counter). My overall target is around 80,000.
How I disected Sydney:
Next, apply a 'MeshSmooth' modifier to each mesh. This will smooth the mesh and increase the visual quality at the expense of polygons.
Select the 'Quad Output' subdivision method for best results.
After that, you will have to decrease the MultiRes percentage radically. For instance, after I applied the MeshSmooth modifier to the face, I then decreased the relevant MultiRes from the above stated 45% to 15%. This maintains a similar number of polygons while increasing the visual quality of the character. Experiment to find the highest visual quality while keeping polygons to a maximum of around 80,000 for characters to be used in real-time applications in the CAVE.
There seems to be no benefit from applying the MeshSmooth modifier to the eyes: there is no significant increase in visual quality, and the polygon count goes through the roof.
