I seem to see a lot more continuous meshes (B) rather than multiple element designs (A.) Is there reason to the madness? As far as I can see as long as the texture is mapped correctly, (A) would use less polygons and be preferable to (B.) Which should be used in a portfolio? (Yes this is for my port, I'll be sure to pimp…
I think you're modules are too small and you need to think about entire sections and floors of the building as a module. I think Guriamo is closer but I think the overall shape is not really set up to handle even sized modules.
sprunghunt: Thanks, that explains a lot. I thought the problem, if any, would be engine related. So deleting a poly that doesn't affect the vert count is not effective? Guriamo: I don't think I understand your suggestion. Are you suggesting that geometry B get cut across to make the tris more consistent and proportioned?…
Sprunghunt, actually no. If the mesh was smooth shaded then that would be true, but since this is all hard edges you actually save quite a few vertices by doing like in figure A. Vertex lighting would be f-cked, true, but that is not used as much in current gen games though. On the other hand, you would probably get…
you should also consider adding a span to connect the seperate window columns, you got very thin and long tris there which is bad... and it wouldnt really add too many polygons.
Figure A is a false economy. Game engines don't store faces they store vertices. So you could bridge across the vertices without affecting how big your model is in memory. Also a long face like in figure A will light badly (when using vertex lighting) due to the lack of vertex splits to define shadows. I suggest you try…