The question came up on SLU. Feeling good about what I said, I’m gonna copypasta it here too. =^-^=
Nearly all of the mesh objects I’ve made so far yield a fractional PE. While rounded to 1 on their own, linking them begins to reveal their actual value. (^_^)
Approaching mesh with that in mind, I see a large opportunity in the building components market. Things like chairs, doors, stairs, fences and such which have benefited from ‘precise’ sculpties in the past will make for very effective mesh items. (^_^)
I cite an example of a flight of 16 steps I’ve made. They were formed from a plane, no rear faces, one UV map face, and a physical bounding box of a cube which follows the stair path. (^_^)
These stairs, alone, are 1 PE. Already, a sculpt solution would need 2 to be cleanly physical. (I say cleanly, because I know how to manipulate the sphere. Still, it’s a sphere and not a flat box surface). I modeled 3 LOD levels and let the client junk up the lowest one. So, in fact, they’re actually 0.5 PE, or something of the sort. (^_^)
So… I’ve linked two together and placed regular prim boxes on the sides in order to visually close in the assembly. Yield: 1 object, 4 pieces, 2 prims. PE, being a reflection of the amount of lag the item contributes, has determined that the assembly is negligible in part that the sum of the whole assembly only relates to 2 prims in server, render, physical, and bandwidth impact. (^_^)
No, mesh won’t replace anything. What it will do is provide people a tool with a broader range of detail and efficiency for various items. They match up with prims almost perfectly. And, when used properly, can and will contribute to better looking builds and reduced lag. (^_^)
The proof is sitting in my front yard @ Morbidette. (^_^)y
