I thought Displacement Maps make those 'deepness'and 'mountains' on your models, i dont understand why Marmoset Toolbag does it so differently (when i put in displacement maps, it made my model explodes', how do i fix this?
looks like all of your UV's and/or edges are currently split. I'd go back into your modelling app and make sure you have smoothing groups set up properly, and that your edges are welded correctly.
looks like all of your UV's and/or edges are currently split. I'd go back into your modelling app and make sure you have smoothing groups set up properly, and that your edges are welded correctly.
THANKS!
is it possible to do this 'smooothing group' in ZBrush? it is currently the only program i know how to use. i did some research on google, seems like ZBrush dont have smoothing-group capability?
zbrush has smoothing , but no proper tangent. dont expect to get good normal map if you use zbrush obj,
either marmoset or pixologic hopefully will fix this issue
now regarding to UV, you need to turn on merge uv in zbrush in tools, ( should be on by default )
zbrush has smoothing , but no proper tangent. dont expect to get good normal map if you use zbrush obj,
either marmoset or pixologic hopefully will fix this issue
now regarding to UV, you need to turn on merge uv in zbrush in tools, ( should be on by default )
thanks xvampire, although it did not help fix my current issue, but i appreciated your feedback!
Unfortunately this isn't something we're able to "fix" on the marmoset end. The reason i say that is because really there's nothing "broken" on our end to fix.
All we do is handle mesh data as it's passed into the program, we can't really make assumptions about mesh data, because that would most likely break things for the majority of users who don't have this issue. As far as i'm aware, Pixologic haven't made their tangent space algorithm public either, this means we can't support their tangent space.
zBrush is a fantastic program, but it should never be considered an "end step" in your creative pipeline. You should really look to import your mesh into a secondary app and fix these issues first. If you can't afford 3dsMax, or Maya, then Blender is a good free opensource alternative.
@almighty_gir: Please correct me if I'm wrong, but when I export an object from Blender in .obj format to use it in marmoset, aren't all UV edges split in the mesh no matter what I do? My understanding was that both hard edges and uv seams cause doubling of vertices/splitting the mesh.
So if I have a cylindrical shape unwrapped, use a hard edge on the seam and crank up the displacement, then the mesh will rip open, right?
If so it may not be "wrong", but for many users this probably is not the desired result. I agree that you can't start making assumptions about the mesh data.
But how about adding an option (defaulting to off) to average normals of vertices that share the same uv coordinate (only for displacement purposes of course)? It could be implemented with a simple checkbox and a float value for a threshold. I'd place it below the "use mipmaps" option and maybe call it "average UV-seam normals" or something like that.
What do you think of this idea?
Are any other users thinking this would help them?
Replies
THANKS!
is it possible to do this 'smooothing group' in ZBrush? it is currently the only program i know how to use. i did some research on google, seems like ZBrush dont have smoothing-group capability?
either marmoset or pixologic hopefully will fix this issue
now regarding to UV, you need to turn on merge uv in zbrush in tools, ( should be on by default )
thanks xvampire, although it did not help fix my current issue, but i appreciated your feedback!
All we do is handle mesh data as it's passed into the program, we can't really make assumptions about mesh data, because that would most likely break things for the majority of users who don't have this issue. As far as i'm aware, Pixologic haven't made their tangent space algorithm public either, this means we can't support their tangent space.
zBrush is a fantastic program, but it should never be considered an "end step" in your creative pipeline. You should really look to import your mesh into a secondary app and fix these issues first. If you can't afford 3dsMax, or Maya, then Blender is a good free opensource alternative.
So if I have a cylindrical shape unwrapped, use a hard edge on the seam and crank up the displacement, then the mesh will rip open, right?
If so it may not be "wrong", but for many users this probably is not the desired result. I agree that you can't start making assumptions about the mesh data.
But how about adding an option (defaulting to off) to average normals of vertices that share the same uv coordinate (only for displacement purposes of course)? It could be implemented with a simple checkbox and a float value for a threshold. I'd place it below the "use mipmaps" option and maybe call it "average UV-seam normals" or something like that.
What do you think of this idea?
Are any other users thinking this would help them?