Original Post
If I want to have the same 2D picture on the screen in different locations, what''s faster: creating new transformed vertices for each one, or translating the same untransformed and lit vertices?
I hate all these tutorials that tell you how to do things, but not how to best do things.