I have a model that has intersecting meshes (no subdiv levels) and I would like to delete all the polygons that are hidden. So, I was wondering if it was possible to to only mask full and partial polygons that are visible in one go (rather than manually painting everything).
Any ideas?
Thanks.
Cheers,
Revanto 
I donāt think thereās a way to do that. Iām also not sure how much help it would be. You might want to just retopologize the mesh.
It would be a great help if it were possible to do that. By masking only the visible polygons the unmasked non-visible polygons can be hidden and deleted. This would make projectall and retopo projection more accurate, avoiding the ugly flaws that you get with projecting onto the overlapping target geometry.
Since the projection tools donāt consider polygon normals and visible polygons, there is plenty opportunity for ugly meshes as a result.
So, even if I were to retopologise the mesh, I would still be susceptible to projection flaws.
By removing hidden polygons, Iām trying to reduce chances of ugly projection thus speed up my zbrush workflow.
So, in answer to your statement āI donāt know how much help it would beā the above is my answer.
Revanto
But the edges of your deleted polygons are almost never going to match up with the remaining surface. Youāll almost always have parts of polygons that still stick out from the surface or are hidden inside it. Those will still create the problems that youāre referring to.
The best solution is not to have intersecting polygons to begin with. Are you creating the model with ZSpheres? If so, a better understanding of ZSpheres will help you avoid this situation.
Yes, there would still be ājuttingā half-hidden polygons but the chance of projection flaws would be reduced considerably.
Iām trying to use this technique http://www.youtube.com/watch?v=2oQzY1xRAVg which provides me mith more control of my initial base design when Iām creating a model. Using separate meshes gives me more freedom in what I am doing. Working with one mesh doesnāt always work for me so if I can use that method of creating my design, I find it is faster and more fitting for me. The problem is the projection part which gives flaws when I create a retopologised mesh. Creating a boolean union with multiple meshes in Lightwave to provide a better target mesh is a nightmare considering that the meshes I want to combine can be in the hundreds of thousands polygon-wise.
Itās a matter or trying to work in the way I feel comfortable faster and more efficiently.
Revanto