When is the Unity ecosystem expected to be "stable" again?

Our team recently upgraded from 2019.1 to 2019.3.

We have adopted the URP and XR Plugin System which are both very, very welcome developments, but so far our experience is unbearable. I could file bug reports for days on end, but we often don't know if the bug is in core Unity, URP, or one of the XR plugins, which don't all play nicely together (Windows XR, Magic Leap, Oculus, Mock HMD)

I absolutely understand the scale of task and motivation behind Unity's move to a package-based system, but is there an estimate of when we can expect to see the dust settle?

 

Out of interest, why did you make the change when some of the packages are still UNreleased (i.e. preview - and you have to manually click through the warnings that tell you this is a bad idea), and some of the most important ones don't exist at all yet (*cough*Valve*cough*)?

All the important projects I have are still running legacy and will remain so for at least 6 months after XR is finally launched, since Unity has a long history of showstopper bugs in every release for the first 3-6 months .

I've got a couple of toy projects I'm using to learn/debug XR, but very much these are throwaways, and I went into it with eyes open, knowing there'd be a lot of broken stuff. So far it's been a good excuse for me to read and learn Unity's source code , which will probably all change in future updates, but is still good background (especially intresting to figure out how they solved some of the architectural issues).

 

I left Valve off my list for a reason
URP, Windows XR, Magic Leap, and Oculus are out of preview, though.

We work closely with Magic Leap and have been locked into a relatively old SDK version for almost a year due to other packages requiring Unity 2019.1. We finally had to make the jump.