In our project we sometimes end up having 150-200 skeletons of varying complexity in the scene, and while the performance is acceptable for now (on PC), the single-threaded mesh updates do take a significant chunk of frame time and we'd like to address that at some point.
I have some ideas on how to hack SkeletonRenderer and MeshGenerator for parallel updates, but I thought I'd check first since you have the rewrite in your backlog - is there any rough timeframe for when this could be officially implemented, to help our planning?
