[2.5.1.135] Zooz ZEN56 rgbw - having "w" actually function on its own switch Bug? Design?

Finally getting around to replacing my RGBWs with the ZEN56 (yah, they got dusty on the project table and just now getting back to them.) The built-in driver appears to create a child device (just one for white) with Parameter 16: 1 – Channel Endpoints Hidden (default) and theoretically I gather does 4 total if the parameter is set to 0.

The Parent driver handles RGB and W using a mix of those channels, as well as controlling the built in animations. And seems to work just fine.

Even better, white is created as a child device allowing for an easy-to-control regular switch and actually appears to control the W channel of my strips as expected.

HOWEVER, I can’t seem to get the On/Off to actually operate unless the parent is on. But that leaves me theoretically having to set level down to 0 or 1? to use the W channel.

Do I have some odd install artifact going on here (all 3 devices are set this way from when I was first debugging and reporting some issues back in the beta) or is this by design (either Zooz or Hubitat intended??) If so, am I being stupid as to how to use this properly? Is it a bug in our driver, or a reason to reach out to Zooz and see about a firmware change?

Thanks. Maybe I’m missing the magic, but this is annoying… (@bcopeland this relates to the oddity of child on/off behaviors I reported as a quirk on the now-fixed parsing issue post in beta forum - I’m able to dig into this more if you need anything from me and it’s not as intended)

Update since yesterday: This behaves the same for a new install of a ZEN56 as well.