I had a plan to take a low poly character and add normal maps out of z_brush. I imported the 500 poly mesh and divided it up and added details. I then thought I’d use zapp link to apply the textures from photo ref. This all works fine in Z-brush the problem happens when I go to apply the colour map to the original mesh. It seems to me that because the map was being created on top of the high res mesh z-brush distorts it correctly to fit. This of course seems logical but my problem is because I’m not using a displacement map but in fact a normal map on my low res mesh the texture is all distorted still and the normal map can’t correct this. So basically I end up with a low res mesh that has a great normal map but a wonky distorted colour map.
Does anyone have a good pipeline for creating normal maps for low res characters so that the colour map also matches when applied back onto the low res mesh?
The only thing I can think of doing is creating the colour map first in another package and then using that as I guide to where to displace the mesh. Saying that I haven’t actually tried it yet so not sure if that will work either
Edited to say that if I import my original mesh back in at level 1 and then generate a normal map the normal map is way off the mark in regards to where the colour map is, they simply don’t match anymore. But if l leave level 1 as the adjusted version that z-brush has generated throughout the iterations of divides and then create a normal map it matches the colour map perfectly. One real pain in the butt for me is the fact that the model has quite a few triangles in it and this just causes the normal map to go way off in some places.
I’ve got some pics attached to kind of describe what’s going on, as I probably don’t make much sense.
One final note, I find that the distortion in the model from a low poly character to be really bad. What I mean by this is because the polys are so few, dividing the mesh into more just causes it to totally round itself meaning it’s very hard to imagine what the final normal map will look like on the original mesh. I’ve never really encountered this problem before simply because I’m used to higher res meshes which retain their shape during the divide. How do people compensate for this. Do you move the points of the mesh around in Z-brush to try and re-shape it back into the more bulky original? Or do you import a denser mesh with the same UV layout, as it’s lower res counterpart. The only problem I can see with this is that I may get the whole problem with normal maps not matching but I have yet to try it.
Really would appreciate some feed back on this as it’s a total un-known for me.
Pete B
Attachments
