ZBrushCentral

Why do we project detail?

I have a question regarding DETAIL PROJECTION, MAP EXPORTING amd RETOPOLOGY.
So let’s say I built a high poly, dense model in Zbrush and now I wanna make it ready for animation so I need a low poly model with a good topology.
So If i export high poly model to Maya and retopo it and import it back to zbrush I know I can project detail but than I have another high poly model. So what am I missing ? I thought omce its retopologized in Maya I can just export maps out of Zbruah and apply it to my new low poly model in Maya. Why do people use Projection and basically end up with same high dense mesh?

Thank you

This is only required for baking maps inside of Zbrush, as it essentially uses the difference in subdivision levels on a single mesh to create its maps rather than use separate/different models.

If you bake your maps externally with a tool that casts rays to create its maps (xnormal, Maya, Substance Painter, etc), then you can use low/high res meshes that have different topologies and as a result you won’t have to do any detail projecting.

Thanks for clarifying it.
So it doesnt matter what program I baje maps in? They all deliver same results? Or if I made it in Zbrush as it was a native software if origin it would bake most precise maps?
Also I assume both meshes (low and high poly) must have same UV’s.
Thank you

The program that you bake in does matter to a degree since:

  • It must fit nicely into your production pipeline/workflow
  • If you are baking tangent based normal maps then the texture baker needs to use the same tangent-basis as the software that will be rendering it in the end.

Up until a few years ago, there was no standard for the tangent basis. It was a wild west where different programs would use different math (and in some cases, programs like 3dsMax would even calculate it differently depending on if it was baking or rendering). A normal map that looked right in one program might look wrong in another program, leading people to think their map was baked wrong. Nowadays a lot of programs and engines seem to be settling on using Mikk, but that’s not always the case. One of the reasons I prefer baking in Xnormal is because it will let you change which tangent basis is used so that you can sync it to whatever the renderer/shader is using.

I don’t like baking in zbrush because

  • You can't use meshes with differently topology (which also means you can use tricks like floating geometry)
  • There's no way to specify the tangent basis being used
  • There's no way to control the vertex normals (something else you'll usually wan to have synced between the baking and rendering process)

Also I assume both meshes (low and high poly) must have same UV’s.

Only the low-resolution mesh requires UVs.

Hey Cryrid thank you so much. I am learning as I go and some of these things still confuse the hell out of me. I really appreciate you taking time to explain it and explain it well :slight_smile:
So I guess one thing I can take home from this conversation would be to bake maps in a software that will be doing final rendering as it will match/sync map baking and rendering algorythm. (Because I dont own Xnormal just yet)