Is there any advantages of using the built in Hubitat Homekit app Vs HomeBridge other than the obvious benefit of removing Homebridge which is just another potential failure point?
I've been using Tonesto7's app with Homebridge for a while with almost no issues (and the onesI did have were self inflicted) - response time has been great so I'm trying to decide if it's worth the time to remove and rebuild everything on the Homekit side.
is this a "if it's not broken, don't touch it" situation OR am I missing something?
Hubitat doesn't support everything / all types of devices.
I believe they have plans on being a legit integration - so they have to play buy Apple's rules to get to that point.
I'm using both Homebridge and Hubitat, it's working awesome.
It's really up to you. If everything is ok then leave it as is. In my case I had so many hubitat devices exposed to tonesto's homebridge plugin that my homebridge install wouldn't handle many more devices. I had to setup a 2nd VM with another homebridge install for more homebridge plugins to include devices not in hubitat. After Hubitat added the HomeKit integration I started using it and I was able to shutoff one of my homebridge VMs.
I think you've got the gist of it. For those who have spent the time building out Homebridge, it's going to do what it does very well for a long time. Homebridge runs as a NodeJS package and can run alongside a bunch of other NodeJS packages. As mentioned, you can and sometimes need, a 2nd or third instance. Easy to do. There are so many plug-ins for Homebridge, making it easy to integrate quickly.
I think the opposite. Maybe it's my environment, but I can have another instance of Homebridge up and running in less than half an hour I bet. Having a TimeMachine backup of the Mac Mini that is running my 2-3 instances of Homebridge would make it pretty quick and simple.
Thanks, Homebridge is also technically a non-supported integration from Apple's perspective but has been solid so has proven it's capabilities.
I can definitely see where having Homebridge running on a "real" machine would bring more confidence in it's long-term stability. I'm running Homebridge on a Pi and it has been rock solid but is not as robust as your mini. When I set it up, Pi's were cheap solutions but now they cost 4x more so if I needed to replace I would go in a different direction (ex: container on my NAS or VM).
I run both because Hubitat does not yet support Bond Bridge exposure to Homekit.
Homekit itself supports the Hue Bridge so I connect that directly to Homekit.
Everything else I have flows through Hubitat.
I pretty much just use Homekit for voice commands anyway and all automation happens in Hubitat.
Bond claims they are starting work on Homekit integration but the initial beta is only for shades and not fans and fireplaces which is what I use Bond for.
Some here. I'm slowly moving my standard devices to the native HE integration, but keeping things like my Garage Door & Locks through Homebridge.
That's the amazing thing about this community and to be honest - that HomeKit allows you to connect various hubs.
But as others have said, if it isn't broken, don't touch it. Now you know if something happens to your Homebridge you can always fall back to HE natively.
For me, Homebridge Hubitat integration has been ROCK solid. Even with 16.2 and new architecture. Hubitat Homekit was solid on 16.1, but 16.2 and new architecture made everything in Hubitat fall apart. HE staff has looked at the engineering logs of my hub to try and debug. They released an updated firmware a few hours ago, but that seems to have made it worse. A couple of others report no real issues with the beta HE Homekit. But my advice...if it's not broken, don't mess with it.