What’s the best practice for light mapping very large objects, like buildings?
At the moment I’m thinking that I’ll just have to have the artists break the objects up into chunks, and have each chunk be a separate object in the scene (so it can receive its own lightmap). But this is kinda inelegant… is there a better way?
Why do think it would be better to break things up manually? Beast makes sure that everything gets the correct density (quality) for the lightmaps.
No, it doesn’t. Unity and Beast won’t pack a single object into more than one lightmap, so if I’m using 1024x1024 lightmaps, then I have only about a million texels to paint the whole object with. With large objects like buildings, this looks very bad.
There was an editor script… just a single line of code i think… that made it so u can specify the actual size of the max size of a lightmap… So you arent limited to 1024… i honestly dont remember the code or where on hte forums it is… but it is there somewhere. I warn u though… i do not beleive the android tablets can go any further then 2048x2048 for a texture.
We already upped all our lightmaps to 2048. So that is half as crap as using 1024s, but half of extremely crap is still pretty crap.
Seriously, has nobody here actually tried to lightmap a large object (like a building)? Are all your environments just broken up into really small pieces or something?
Thanks, don’t care. This is a desktop project.
Well in the real world you would be breaking it up in a 3d package… even if u generate the lightmap in a 3d package. So if its 1 big mesh in max or maya… its gonna be 1 big mesh in unity… What do you expect unity to do for an art pipeline that wasnt planned out? Unity is not going to modify your art for you and make magic happen and suddenly make it a well put together model. If you want seperate lightmaps your going to have to break it up so it uses seperate areas.
Yes, that’s what we’re doing. I was wondering if there’s a better / automated way.