This is more theoretical than an actual problem in need of a solution.
I'm making a 3D surface out of quads based on an array of height data. One way I can do it is to have each quad be its own object. The other way is to make a single huge object with strips of quads. I suspect the second method is faster. My question is: how much of a performance hit would I take by using the first method? Or is it all pixels by the time it gets to the scenegraph and the canvas with no difference at all?
Yes, I could test it, but I expect someone else already knows the answer. It seems to me method one simplifies location selection, texture application, and shape modification (digging holes and such).