Compound Steel Helmet

Cactus on Fire

Hey. I've been working on this for some time, and I finally finished the lowpoly baking. There are just some issues left that need fixin' like the seam cut on the forehead and the gap behind the jaw.

Is there any way to bake the floating geo on irregular surfaces ? I guess I can add bumps and normals on the baked maps with mudbox but they will be excluded in the ambient occlusion baking process. So how do you deal with microdetails on rough surfaces

Comments & Suggestions are always welcome.

AtlusZMH

add some more polys to that thing. The silhouette is very jagged and doesn't really do the model justice. Smooth out the top at least.

As for the smaller detail, I would suggest something like nDo2.

But again, I would go back and throw a lot more polys at this thing to make everything have a nicer silhouette before moving forward too much.

Cactus on Fire

AtlusZMH wrote: »

add some more polys to that thing. The silhouette is very jagged and doesn't really do the model justice. Smooth out the top at least.

As for the smaller detail, I would suggest something like nDo2.

But again, I would go back and throw a lot more polys at this thing to make everything have a nicer silhouette before moving forward too much.

Okay thanx. I'll start increasing the head polys where its more spherical and give it a shot in nDo2

BARDLER

You have some normal map issues, and some highpoly problems as well.

It looks like a combination of some problems with your sub-d model, and the way you baked your normal maps. Could you post your highpoly mesh with and without wireframe? Also did you do an exploded bake and use a cage when you baked your normal map?

Cactus on Fire

@BARDLER

The highpoly has pinches as well unfortunately The cylinderical objects in the back are prebaked because Max was giving memory errors when baking that many polys.
As for the baking process, everything was put together just like the highpoly to get an accurate AOS.

J0NNYquid

Yeah, the high-poly definitely needs work, I'd recommend spending a lot of time in this thread: http://www.polycount.com/forum/showthread.php?t=56014 Specifically withing the last 20-30 pages, Perna gives some incredible advice/tips. I think you'd definitely benefit from it. We'd also probably need to see some wires as well, would really help determine where the pinching is coming from.

BARDLER

J0NNYquid wrote: »

Yeah, the high-poly definitely needs work, I'd recommend spending a lot of time in this thread: http://www.polycount.com/forum/showthread.php?t=56014 Specifically withing the last 20-30 pages, Perna gives some incredible advice/tips. I think you'd definitely benefit from it. We'd also probably need to see some wires as well, would really help determine where the pinching is coming from.

I agree with this as well. Also for hard surface stuff you need to do exploded bakes in order to avoid shells bleeding into one another. You can do a couple things to get an accurate AO with an exploded bake. I personally use a program called Knald, it is super fast and gets awesome results. You can also follow this process by Earthquake http://wiki.polycount.com/AmbientOcclusionMap.

As for the projection issues you need to get a better lowpoly mesh, and you need to use a cage. Checkout these threads by Earthquake that cover good normal map techniques http://www.polycount.com/forum/showthread.php?t=107196 http://www.polycount.com/forum/showthread.php?t=81154.

Last piece of advice is check out xnormal, it uses a lot less ram to bake maps than any other software I have seen. It can handle any size map and pretty much any polycount you throw at it.