Hello, I have a question about getting PBR material detail onto a low poly mesh.
I have a high poly mesh that is an accessory to my character model. I sculpted all the details I wanted, and then retop’d it and baked the high poly to the low poly. However, I am trying to fit multiple objects onto one UV map. which means I can’t apply a PBR material to the low poly as it will take away the material with the maps baked from the high poly.
Does it make sense to apply the PBR to the high poly before baking to the low poly to get that addition detail from the PBR? Is there a better way to handle this situation in general?
Also, when creating characters does it make sense to have the body of that character have its own UV map, and then have all of the accessories on another map?
That’s a lot of questions, and I hope I articulated them well enough. Any insight provided would be very appreciated!
Adding PBR to a HP mesh is a total waste of time ( unless you are using the HP only for rendering) as you end up if you Re-topo down to a Low Poly you lose all the work you did on the HP…
As @oo_1942 said the normal process is to bake the Normal Map and Displacement if used ( after you have created the LP and Un-wrapped it) so you can BAKE the DETAILS DOWN to the LP mesh, then you work only on the LP to create the rest of your PBR maps…
Sections that have little to no detail can be packed into their UV, and then parts that have High Detail into another (I use UDIMs for this) and it is the basic work-flow you will see in something Like Substance Painter) Imagine a Gage Dial, where you need the numbers and name of the dial and you want to add some grunge to it, if Baked into the little detail UV this would make them un-readable, on its own 2048 UV then it would keep all those details…
This brings up another point, if you need to or want to bake all the files to their own UV map, then increasing the size can allow you to do so if you don’t have very fine detail since you can double the area of your UV and Maps, say 1024 > 2048 > 4096, etc.
Thanks for the reply! When you say displacement is that the same as a cavity map? Where the cavities of the mesh are darker and the flat parts are whiter?
Thanks for the reply! I’ll have to look into UDIMs. What does working with the LP to create PBR maps look like? The LP already has a normal map(and other maps) from the HP so how can I add another normal map to that same mesh?
Very interesting, I didn’t know about this. So you can use both a normal map, and a displacement map on the same mesh? Would you use a normal map for large details and a displacement map for finer details?
Also, most modern renderers have an ‘autobump’ feature built into displacement, if you enable it. It adds microbump detail on top of the displacement map, without extra maps. Both bump and displacement come from just the displacement map.
Yes, it’s the opposite in all cases. A normal map is nothing more than a shading trick, an illusion. It doesn’t affect the silhouette, whereas a displacement map is actually affecting the geometry and changing the silhouette.
Well said, Musashidan. I’m sometimes timid (not always!) in my maturing years at making absolute declarations. But I can’t think of a single case where this isn’t the guiding rule.
Hehe. I’m quite the opposite as more and more of my best years pass behind me. I go the direct route. We have earned that right as we approach ‘grumpy old men’ status.
I see. So you would use a displacement for high poly modeling to get really fine details because you have a lot of geometry. Then that detail would be reflected on a normal map when you bake to low poly.
Yes and no. A displacement is capturing mid to high frequency details, specifically the silhouette changes, which normal maps cannot. The normal map is capturing high frequency/microsurface from the surface shading. Technically, it captures XYZ shading data and stores it in RGB - red/green/blue channels of the texture map. The displacement map is capturing vertex position/height data.
Technical stuff aside, the simplest way to think of it is in terms of silhouette. Both of these maps can capture the same high frequency detail, but only the displacement can capture the silhouette.
No. This is the other distinction in usage. If you are talking about baking to low poly then you wouldn’t use a displacement map. This is pretty much the reason that normal maps exist - to fake high poly detail on a low poly mesh that cannot use displacement.
Not self reflections. For the others, that only affects light transport (on the micro/nano level), not what you can physically see (on the macro level). Not sure if normal maps on a regular glossy are a viable solution to doing anisotropic shading these days, sure didn’t use to work back in the day.
I go the direct route. We have earned that right as we approach ‘grumpy old men’ status.