The necessity of a clean mesh cannot be overstated. Simple put… a clean mesh is easier to work with for everyone down the pipeline from you. It also makes for more fluid characters when animating whether it be in a game or a production.
In a production such as a video-based animation, you can have high-polycount characters depending on the scene and their locations within the scene.
In games, a leaner mesh is what counts. You can’t have a 1 million or in some cases even 100K polycount characters running around otherwise that running around will be more like crawling around. We all know about Level of Detail (LOD) and a lot of 3D sculptors struggle with reducing their original meshes to work with LOD depending on the distance to the camera. To top that off you need at least 3 LODs on each character and it could be more.
Decimating characters down to needed levels really screws up a mesh. Those nice and neat little quads can become a jumbled mesh of triangles that in some cases won’t bend properly if there aren’t enough tris in a given area for it to bend.
From the Manual Demonstrating the Cleaner Remesh of an Object
You also don’t want to make an enemy of the character rigger on down the pipeline. Sending them a jumbled mesh will not put you on their Christmas list. You don’t want to be the one that elicits groans from the riggers when your name is mentioned. As a grunt, I rigged characters at the first of my career and I dealt with many a mesh that I wanted to stick where the sun didn’t shine next time that character developer showed up for a pow-wow.
A clean, quad mesh is the holy grail of the pipeline.
Personally, I think they are Unicorns… as much as I want to believe… they just don’t exist. At least I’ve never seen one yet but the goal is to get as close as possible.
This why that dreaded word… dare I utter it… retopology… exists in our vocabulary. Like rabies and plague it's a word we all could have done without but alas… it is real even though it sucks major wind.
To me… retopology is one of the worst assignments I could imagine. It’s right up there with watching wine ferment and paint dry. There are no crowds in retopology (unless you are topologizing a crowd) as no one I know really wants to do it. Period.
Decimation screws things up so what are we left with?
A method created by a group of people you’ve never met, probably will never meet but owe a debt of gratitude for giving you a decent mesh from that monstrosity you were sent or put together yourself.
Hey… with a Remesher… no one knows how bad the original mesh was and that can cover and correct a multitude of 3D mesh sins.
The downside to a Remesher (you’re in 3D… you KNEW one was coming) is the fact that you are at the mercy of how that particular tool handles the mesh. The remesh algorithm was written in a place far, far away by those that lurk in the shadows of 3D development.
Remesher 3.0 on Organic Model with MashUp Mesh First (Left), Initial Remesh (Middle) Followed by Final Remesh (Right)
If you flip their switches, twirl their dials and it doesn’t work for you then you are out of luck and maybe looking at a long and tedious retopology session.
This was pretty much the case until the Remesher in ZBrush 2018 come out. It had better methods of remeshing than the original version as the company had time to process user feedback and respond.
Now, with Remesher 3.0, there is even more control and its in the areas that some of us need the most.
As much time as previous Remeshers have saved us they literally destroyed hard edges. They were great for organic models, but hard surfaces suffered.
Now we have Edge Detection which needs no explanation and even better we have KeepCreases which respects any creases in the mesh when optimizing. Combined these two features can produce a much cleaner mesh that respects hard surfaces.
To keep users happy, they have kept the old Remesher under the Legacy button, which will allow it to be used on more organic models while the new method can be used on hard surfaces without a lot of damage.
Just as Remesher 2.0 was a much-needed improvement, its baby brother is already proving to be a powerful new kid on the block that provides cleaner quads while maintaining creases and hard lines.
It’s just another reason that the 2019 release is a major step forward in the evolution of ZBrush.
M.D. McCallum, aka WarLord, is an international award-winning commercial graphics artist, 3D animator, published author, project director, and webmaster with a freelance career that spans over 20 years. Now retired, M.D. is currently working part-time on writing and select character development projects. You can learn more about MD on his website.