I'm not really sure what's wrong. I've re-implemented the vertex conversion twice, the second time duplicating all the vertices for each mesh, assigning the appropriate joint IDs for each vertex, and still get this odd transformation problem. And where are the shoulders?? #PSO #ReverseEngineering
Giving up on this for now. I'm not sure what's wrong and I don't know what kind of assertion test I could write to ensure the NJCM parse is correct. It would be nice if Noesis v4 was open source. #PSO
Also honestly at this point I'm ready to accept that glTF is just a bad format for interchange and should only be used for runtime delivery. And even then, I'm becoming less convinced of its usefulness.