I haven't done any modeling in a while so I chose this to get me back on track. I used to always try modeling high poly shit but always tried doing things out of my league like guns and vehicles before I really knew what I was doing. So far I have done the high poly, low poly, and UVs. I tried baking in xNormal but that got totally fucked. I pretty much have no idea what I'm doing as far as the normal map stuff goes so here are pics of my high and low meshes and my normal map. I think I know what is wrong but I wanna get your guys' input before I try to change anything. Thanks in advance.
High:
Low:
Normals (sized down from 2048x2048 to 512x512):
By the way, I also baked the entire thing at once as the low poly is combined so I could put them on one UV sheet. Is there an alternative way so I do not have to attach them?
Replies
Looking at your bake it looks like you have some errors to fix, and stuff that should not have unique texture space (repeating textures should overlap in the uv's).
I actually find baking in one pass much quicker. Explode (separate them) out the low poly bits you want to bake, attach them as one mesh, then add as many high poly bits as you need and bake just once.
Baking lots of little low res bits separately just means composting time in Photoshop.
Just a thought!
Honestly this is the worst, least productive workflow you could possibly do. I always cringe whenever anyone suggests this, or when i hear someone say something like "oh here is part of the bake, still need to bake the other half". Every additional bake you have to setup, debug and render is that much more wasted time.
What you should do is keyframe your objects(high and low), set a frame where they are all "exploded"(manually separted so you dont get any bake errors from intersecting meshes). And bake all at once. You can quickly bake an AO map from just the lowpoly(put back together) and add that in your texture with the AO from the exploded high.
Hell even baking it all as one solid mesh, and doing the touch up work in photoshop to fix errors would be better than doing separate bakes for each mesh chunk, but i would never recommend that method either.
It shouldn't matter if the high is inside the low or vise-versa. What matters is that the ray distance, or cage, depending on which you are using, is larger than the lowpoly. The low(cage or otherwise) casts IN, not out.
In max cage is the default behaivor, and the cage settings by default suck. You need to go into the cage modifier and hit "reset" and then play with the push amount(or whatever it is called) until it is just covering the highpoly mesh.
Alternatively you can go into the bake options, and click "use offset" then set an offset distance.
The main difference between these two methods are:
With cage in max, if you have any hard edges(smoothing groups) they will be welded, and produce an averaged result, this is great when you want to get the detail on the those hard edges without missing any informationg, but may cause some details to be more "skewed" because of the averaged calcualtion.
With offset, the hard edges will stay hard edges in your bake mesh, so you will lose detail around those areas as the rays will shoot straight out and miss the corner detail.
If you dont have any hard edges, the biggest difference is that with cage, you can visually see how it works, and even go in and edit/tweak the verts of your cage to fix things.
For complicated assets with many hard edges i find it can be best to render both your AO and normal maps once with cage, and once with offset, then composite them in photoshop to get the best of both worlds. You can also try manually tweaking the cage, or adding more edges to the lowpoly mesh to fix many issues as well. I try to stay away from editing the cage manually, because if you need to bake in another app(XN or maya for instance) you're going to have trouble getting the same results.
[edit] also, post some more images dude, the more images, the easier it is for us to help with problems!