Max>Zbrush workflow?
polycounter lvl 8
Well, i just started with zbrush, this was my workflow:
-Make Low-Poly model in 3DS Max
-Unwrap in 3DS Max
-Added chamfers to correct smooth (this fucked up the UV)
-Zbrush highpoly, but now i'm stuck because the UV is screwed.
I'm deffenaly doing something wrong, is there any ways of getting the normal map from the highpoly to my original lowpoly model without loosing the uv's? (as it would totaly suck if i had to do them all over again)
Is there any better way of doing this?
Highpoly model (not that good, i know),and original UV layout.
Thanks:thumbup:!
-Make Low-Poly model in 3DS Max
-Unwrap in 3DS Max
-Added chamfers to correct smooth (this fucked up the UV)
-Zbrush highpoly, but now i'm stuck because the UV is screwed.
I'm deffenaly doing something wrong, is there any ways of getting the normal map from the highpoly to my original lowpoly model without loosing the uv's? (as it would totaly suck if i had to do them all over again)
Is there any better way of doing this?
Highpoly model (not that good, i know),and original UV layout.
Thanks:thumbup:!
Replies
I understand if you just want to test the workflow, but since you're stuck at this point anyway, it's a good moment to have another go at the uvs. You'll be surprised what you'll learn if you take the time to look at a few unwraps by skilled professionals, and try to understand why they did things a certain way.
First off, a few tips if you do end up going down that particular road:
-Fill your space as efficiently as possible. In this case, that'd mean rotating the pieces to be straight, and if you have a lot of space left, choose one of these options: mirror less of it / use a rectangular image (512x1024, for instance) / cut your pieces differently so as to more effectively fill the space. If you go with that later option, ofcourse you should be careful you're not shooting yourself in the foot, and you're actually gaining something meaningful.
For organic/character modeling:
1. Create a fairly simple, sculpt friendly, base mesh in 3ds Max -> evenly dense mesh with mostly quads. It should be roughly what you got in mind, in example if your character has a tail or four arms, the base mesh should represent them to be properly sculpted
2. Import the base mesh into ZBrush and do a lot of sculpting even changing drastically the original shape.
3. Export from ZBrush the hi-poly and a mid-poly sculpt as .obj
4. Import the mid-poly in 3ds Max and create the low-poly around it. This is the retopologization step. It is the final mesh you get, respect the shape and the silhouette taking into account character deformations during animation
5. Unwrap the low-poly UVs
6. Import into xNormal the low-poly mesh from 3ds Max and the hi-poly from ZBrush and bake normal maps, and so on.
For Hard-surfaces/props modeling:
1. Create the detailed hi-poly in 3ds Max
2. Adjust the hi-poly to be sculpt friendly -> evenly dense mesh with mostly quads
3. Import the sculpt friendly mesh into ZBrush and add surface details like scratches
4. Export from ZBrush the hi-poly and a mid-poly sculpt as .obj
5. Import the mid-poly in 3ds Max and create the low-poly around it. It can be derived by the original 3ds Max hi-poly reworking the mesh.
6. Unwrap the low-poly UVs
7. Import into xNormal the low-poly mesh from 3ds Max and the hi-poly from ZBrush and bake normal maps, and so on.
-Why do they need to be better packed? I understand its better for game engines and stuff, but MightyPea said if i still got open spaces i should mirror less of the model? Why? I tought it was better to use as many UV space.
-Sometimes you need to create some scratches in zbrush. But they are not hard sculpted, but i think normal mapped(Correct me if I'm wrong)? So if you export the highpoly OBJ, the normal map isnt saved. How do i keep these small details that aren't hard-sculpted?
-Last but not least: Is Xnormal realy the best way to go? I know 3ds max can bake normal maps too. Do they use Xnormal in development studios too?
Thankss again guys!
1. the uvs need to be better packed because they arent using enough of the pixel space also they are unevenly sized so some parts will have loads of pixels to detail them while other just important pieces of your sword will have hardly any pixels and could look a little stretched or warped. so if you show only one side of swords blade in the textures then you get 2x as much space to use on that one half and it could look like double the detail of what you have now if you pull it off nicely.
2.zbrush should be able to export a perfectly good highpoly regardless of how small the details are especially if you manage to decimate nicely as well, only problem is if your computer has trouble handling large files and doesnt want to export/import
3.xnormal is really excellent and used in some studios as far as I can tell, yes max can do it too so if you already know how to do that then you can as long as your happy with the workflow and results.
hope that makes some sense its rather late here
http://eat3d.com/free_zbrush_xnormal_pipe
It's very similar to what SyncViewS mentioned. There is another video there regarding Meshlab and polycruncher, which can connect a few dots, and helped clear a lot of confusion I had previously
Also, when i start xnormal, i get a cant find renderer, using default bucket one error. And also, do you need the uav in Zbrush? I left it out because of that i don't have a polypaint, and only sculpted the highpoly.
some xnormal tutorials. check out the one for normal mapping. its easy, short and gives great results - hassel free!
and you dont need uvs for the high poly if your baking normals, oclussion etc. (if thats what you were asking)