I´ve been working on this project: http://forum.unity3d.com/threads/153689-procedural-character-generation
and would like to know your opinion about what´s more performance efficient: working with transparent textures for facial hair composition, or some kind of blend using a custom shader.
For now, I´m trying the first approach, but would love to hear your opinion.
I’m working on a procedural character generation system, so I need to have possible combinations between head textures and different facial hair textures. Right now I’m testing performance with a second mesh layered above the surface of the first one. The problem with this approach is the aditional rigged polys.
You can see the test here: http://forum.unity3d.com/threads/153689-procedural-character-generation
You could composite the textures at run-time into a unique texture for that particular character. Or have optional slots in the material that get masked on top as part of the shader.
I wouldn’t recommend decals (floating polys) as that gets expensive and tricky to account for in every situation.
For fuller bears and such, an additional model (similar to how you’d do hair) would be required for the best results, I think.
Also, I’d have another look at the muscles of your characters - especially on the bulkier males around the elbows, the brachialis is waaaay too big and bulky.
Hi there Farfarer,
Thanks a LOT for your insights, I’ll give it a try as soon as possible!
About the anatomical considerations, I still have a long road ahead, thanks for the help