This is a follow-up to my earlier point about higher joint counts.
Are there any plans to support higher-resolution cages in the future?
If Roblox is serious about supporting a realistic clothing or merchandise economy within the metaverse, then higher-resolution cages are a must.
Right now, subdividing a cage doesn’t actually improve visual fidelity. It simply doesn’t work.
When I created my first UGC bundle, I expected that subdividing the cage would enable interpolated smoothing and result in a more accurate shape, but it just threw an error.
A cage’s UVs act like its DNA, so Roblox’s system can detect which regions correspond to the chest, hands, and so on.
I believe subdivision could allow for more flexible, higher-fidelity cages through UV-based interpolation.
As Roblox continues pushing toward more realistic experiences, it needs to support increasingly complex character shapes.
This is especially noticeable when trying to create items like gloves that show individual fingers, toe socks, chokers, swimsuits, and other form-fitting gear.
While this isn’t a major issue for blocky characters, Roblox—if it’s aiming to be a true metaverse—should start preparing for the eventual inclusion of complex, human-like characters, even if performance limitations make that infeasible right now.
Is there any direction or plan for this?
In my opinion, cages need to support subdivision, and it should be possible to clearly indicate which face of the hand belongs to the thumb, index finger, middle finger, etc.
Vertex colors are already used in the templates, but the fingers aren’t marked individually.
To make the hand cage more precise, each finger should be marked like other regions.
I sometimes get confused while editing the hand cage because it’s not clear where the thumb ends and the other fingers begin,
and there’s also no clear guidance on how large each region should be.
