If most of what you're doing is facial closeups, consider deleting the rest of the figure. . .that is, you can have a "head, shoulders and hair" model for the closeups, and then use the larger model for the full shots. Poser's geometry processing isn't particularly optimized, so that in a scene where the bulk of the body isn't visible, the geometry is still calculated. If you think about it, movie cinematographers think like this. . .detailed makeup is unnecessary in a crowd scene, mandatory for a "two shot" Think of it another way: if the amount of image area that the rendered image is going to cover is only 100 pixels square, then having 200 pixels square of texture is unnecessary and wasteful. you can downsample your textures in photoshop and have a Steff HiRes Stef MedRes and Stef LoRes. . . use them in different shots depending on how close the camera will be