I've installed Lutron Caseta solutions in our previous homes and I would agree for light control they are rock sold. In or current home I wanted to have less independent solutions and try to simpify things. I went with Z Wave over Zigbee because Z Wave claims certification where Zigbee is just an open standard which gave me pause for compatibility. I do have some Zigbee devices and from what I can tell they've been reliable overall but I don't want to give ZigBee too much credit here because the count is very low.
My takeaway from all this and hidesite being 20/20, I should have used the the Latest Caseta system which have switches and Dimmers that look very standard and work very well from the end users perceptive. The total cost of the hardware and the human hours of installation would have been a more reliable / robust solution over Z wave switches build by Jasco (branded by GE, Enbrighten). I also have several Scene switches from Zooz which have worked well overall. The Zooz scene switches do tend to each battery faster then advertised so it's wise to keep a lot of spare batteries in a draw.
My one other observation around the Hubitat itself. I want us all to keep in mind it's meant to be a very cost effective engineer centric solution. While very robust in feature it's CPU can quickly get overwhelmed and start dropping tasks when it experiences a spike in things it needs to do. For the most part one you get all the hickups fixed and find a decent steady state, it's been reasonably reliable. I'm hoping future iterations get stronger CPUs to handle the workload spikes.
A difference solution I've been testing more along the lines of learning over actual migration is the Home Assistant Yellow design which requires the purchase of a Raspberry PI CPU. The total cost of ownership of such a solution has been around $500 which is higher then most want to risk and that price is do to the Raspberry PI shortage which increases the price up significantly. My takeaway form there design is ONLY the Rasberry Pi CPU appears to have a lot more spare Spike/Inrush room to handle issues that may arise. I'm not advocating for any migrations, just observing performance. Which begs the question could the Hubitat team create a luxury high end version for say $300 with greater CPU. Might be worth looking into but not sure if they have the bandwidth for such an undertaking.
I used to be the same way, once I started using the Ring motion sensors, which required S2 and saw no noticeable impact, I stopped worrying about it. I did finally get rid of all but one S0, and that's a lock. I'll probably replace the module eventually.
But yes, Zigbee has always been a flaky pain for me.
I have tried Home Assistant and I have had reliability issues. It is a fun interface to play with or for your personal home but it is definitely not a commercial solution; Hubitat is.
I disagree with your suggestion of a DIY solution. Commercial solutions, such as Control4 and Crestron, are more reliable and performant than DIY solutions like Hubitat. Hubitat is designed for budget-minded consumers who are willing to deal with compatibility issues and a complex user interface. Hubitat is not a bad product, but it is not comparable to commercial solutions such as Control4 or Crestron in terms of quality.
Hubitat's biggest advantage over Home Assistant is its tight-knit community and team focused on a single hardware platform. This allows the relatively small team to fix bugs faster than open hardware solutions like Home Assistant.
However, Hubitat's biggest weakness is its performance under high load. During normal operation, the system works as expected, but when there is a spike in demand from users or its modem components managing the Zigbee/Z-Wave networks, the system can quickly become overloaded, resulting in delays or failed executions.
Offering more powerful versions of the Hubitat hub at a premium price could be a profitable move for the company, both in terms of sales and adoption. My comparison to Home Assistant was intended to highlight the available CPU performance differences and how that platform takes advantage of all that extra power.
There are other issues with Home Assistant that I will not go into here, and I am not advocating for one platform over another. I am simply suggesting a way to make Hubitat more reliable for a wider audience.
Overloads usually come a roguedevice or from a bad app. I have a shite ton of rules and near 200 devices on my hub, it purrs along fine and gets a lot of demand. This is the experience of most of us. It tends not to seem it though as when you come here all you see are complaints or issues but that is from a small percentage of people who actually use hubitat. You don't hear from people who's stuff is working as expected. I will agree with you things like crestron and control to a point. They are very controlled in what is allowed on their systems and what is has been tested and re tested and tested again. With systems like hubitat, HA, ST, it's more of a "We have devices that test as expected and that's why it's on the compatibility list but we will let on anything that follows spec in general" which can throw unexpected things into the mix. Not to mention being open to community drivers and apps. Don't get me wrong, those that provide community supported drivers and apps are farking brilliant. I couldn't program my way out of a wet paper bag but sometimes a badly written app or driver can cause a lot of issues too... For $150 bux we're gonna put up with a lot more than someone who paid 5k for a system
Background: I have all of Hubitat, Control4, and Home Assistant. Plus Alexa and Apple Home.
More reliable I'd mostly agree. Control4 is by every definition a legacy system with all the good and bad that come with it. Plus its extensive dealer support network. I too wish the Hubitat hubs were more resilient.
More performant I'd disagree. Control4's competitive advantage is very good AV integration and control + good enough UI controls + good enough integrations + good enough programming. Where capabilities overlap - non-AV & security integrations and programming - IMO Hubitat is ahead of C4.
For $150 you can get the previous generation C4 low level controller (EA-1) on eBay. It will be supported for a long time. It lacks some of the AV capabilities of the higher models. Paired with a C4 remote ($100-200 on eBay) it would be a very good replacement for Harmony users. One easy to include integration that can be run in parallel with Hubitat is Lutron lighting. So it's pretty easy to have AV + lighting control on an AV remote control.
For me the only lighting control I do is when the channel changes (or rather app) on my tv, after 6, the lights dim to tv watching modes. Should it be back on the home screen for more than 5 mins the lights raise. That's about it...(the channel change on my google tv activates a scene on my lutron hub via hubitat). Once I got the chromecast kinks worked out, that worked fine.
As far as hubitat vs c4 I was thinking of things in terms of more stability because they're so rigorously closed.
I don't do very much automating in my C4 system, that's in Hubitat. The legacy C4 remote has a menu system where you can control any devices on the system. It has a few hard buttons I use to control lighting scenes when watching TV as there are no "standard" light levels in our spaces. But I can also lock the back door via menu.
It's really not that closed. Anyone can write drivers for the system, not unlike Hubitat. Dealers won't deploy unreliable drivers as it drives up support calls. C4 does exercise control over Zigbee device certification. Apparently they're finally going to move to Zigbee 3.0, we'll see if that opens up to generic Zigbee components. On the Z-Wave side their current controller line can use them, but must use C4 provided drivers, which would not be considered feature complete in Hubitat world. That means 3rd party drivers end up being written for IP controllable devices. They also tend to be for a fee. Like $90 MSRP for TP-Link.
I pull out those solutions because when they go bad it is very hard to fix them. Lutron Caseta PRO with Hubitat is very reliable. Caseta being the most reliable and Hubitat backing it up. As a professional I will actually recommend pulling out a Control 4 system and putting in the Lutron Caseta system. To keep my Hubitat reliable by only installing the APPs that I must have. I also make all timed events on the Lutron Hub unless it is a non Lutron light (I sometimes add with a few Zigbee switches). I avoid Z-wave if possible.
