ZBrushCentral

Woman with 25K polys: how many polys in head?

ZBrush Usage Questions
#1

I’m struggling to make a relatively low-poly model that has reasonable geometry at the facial features (not to mention hair, hands and toes). Is there a rule of thumb on what percentage of polys should be devoted to which body part? I know large relatively flat areas get fewer polys, and detailed areas like the eyes and mouth get more. But how many more? Does the head get as many polys as the rest of the body put together? That’s what I seem to need!

Can anyone point me to a guide or a “rule of thumb” about this? Many thanks.

#2

Hello @Grotius ,

There is no rule that I’m aware of that applies to every situation. As with most things, it depends on your eventual output goals. Which areas of your model are likely to receive the most scrutiny? Which areas of the mesh will require the finest detail?

Even if an area is flat, if you’re going to be sculpting high res detail on it, it will need sufficient polygons to capture that detail when subdivided. If you’re going to be covering every surface of the mesh with fine detail (like extra fine fabric detail), then there may be no area that you can skimp on polygons.

Another thing to keep in mind is that optimal topology for sculpting and painting in ZBrush is evenly distributed square polygons. You should aim towards this if you’re going to be working in ZBrush, then consider retopologizing at some point if your final output requires more optimized topology. Animation in particular requires that topology be drawn in specific ways for the best results, and this may not be the same way that provides the best results for sculpting.

Obviously faces are going to be important, as they are a major focal point, and will receive a lot of detail. You’ll want your topology here to be as clean as possible, and the polygons to be smaller than other low-visibility areas. Low-scrutiny areas like the bottoms of feet, under arm cavities, etc, can be good places to tie off edge loops in an uglier way if necessary. Avoid poles where you can, and try to place them in low visibility areas.

Whatever you do at the base level of topology will be multiplied many times when subdivided to high resolution. With careful planning, you may be able to create an optimized mesh that saves you from having to subdivide it an additional time in order to get enough resolution in a crucial area. This can make a huge difference in how easy a mesh is to work with.

However, ZBrush also gives you the freedom of working on a mesh without a fixed topology until the final stages of work if you wish. Some tools in the ZBrush toolset can’t be used without changing your topology. This gives you the freedom to change the geometry on the fly without worrying about preserving a base topology. It’s simply another option.

Good luck! :slightly_smiling_face:

#3

Thanks very much for your helpful reply! Lots of good food for thought.

Is it uncommon to use ZBrush’s sculpting tools to produce a “base mesh” of about 25,000 polys, suitable for animation and maybe games? I suppose I could try using zModeler, but I find zBrush’s traditional sculpting tools easier to work with. For the same reason, I don’t love box/extrude modeling in Blender, say.

#4

This is not uncommon. It is less common however to do this at the start of a project than it used to be.

ZBrush didn’t used to have all the mesh creation tools it does now, and users would commonly create a base mesh in another application, subdivide and sculpt on that mesh without ever changing the topology. This is less common than it used to be but still done, especially for projects where the mesh may need to match external files.

These days, ZBrush has very robust mesh creation abilities in itself, as well as very fast auto-retopo options that allow you quickly resurface a mesh. Tools like Sculptris Pro and Dynamesh allow you to sculpt out complex shapes from a simple primitive without being tied to or constrained by topology.

If you tried to do this with a fully modeled mesh with topology that cannot change, you will inevitably stretch the polygons out of shape with any sort of drastic change to the form. Working this way, the user generally doesn’t worry about topology until their form is stable, and then they construct an export quality stable topology for the purpose of posing, sculpting high res detail, UV unwrapping and texture creation. Depending on your output needs, for instance for 3d print, you may not care much about your topology at all, and simply decimate for print.

Both approaches have pros and cons, and may serve one situation better than another. It’s entirely up to what works best for you.

:slightly_smiling_face:

1 Like
#5

Thanks! Very helpful. I appreciate your taking the time to give me such a thorough reply! Thanks again.