Is it possible to learn Home Assistant and incorporate it for a good basic and reliable home automation system ? I guess I could clarify what I’m trying to establish by asking the following questions:-
Is it as reliable as commercial solutions by the likes of Control4 or Crestron ?
Once a Pi4 is setup as per instructions, is it stable ?
I read that an easy scripting or configuration assist has been implemented, is this easy to learn ?
Is HA applicable to both wired and wireless configurations ?
Has it got a nice client interface ? (Love the sample on the site with the floor-plan orientation).
First and foremost, HA is not on version 1.0 yet, so it will mean some things will need maintenance. Whether that answers your “reliability” question or not I’m not sure, are you willing to perform some minor modifications when a breaking change is added with a new version?
For the rest, the answer is pretty much yes to all your questions.
With HA, the sky’s the limit, there is no such thing as a proprietary eco-system, it just integrates with anything and (almost) everything.
For your customisations questions, again it depends on how much involved you want to be vs. how much “works out of the box” you want the balance to lean towards. Don’t forget that customisation and reliability don’t really go hand in hand without some expected work from your side…
Thank you Lolou44, I really appreciate your input. From the impression I get, I think HA will be the only system in the open source that may eventually compete with the big boys.
I guess what I’m asking is if it is reliable enough for an installer to incorporate it in client homes ?
I am far from being an expert in HA… But I started the journey alittle bit more 1 year ago (from scratch) and learned a lot on programming, scripting, integrating , and so on around HA… and now I have two raspberry pi 4B (2GB) + 1 for test and backup, running on two different locations speaking to each other via MQTT through internet (because I have only one interface gathering the status of the equipment on both locations), for various equipments: air conditionning, heaters, lights, pool, vacuum robot, solar panels, alarm systems, ip cam, covers/curtains, sending SMS and more… Never had an issue with the hardware, but sometimes with a new release of HA but as mentionned by @lolouk44, HA is not on version 1.0 yet… By the way this upgrade issue was generally fixed within one hour thanks to the HA community (you can even go back to the last succesful running version if you want to)… I do recommend HA.
Here are some examples of my user interface (only two panels out of 14 in total)… You can nearly do whatever you want…
I only dabble in home automation, and it takes probably on average an hour every day or two. Just keeping up with this forum is probably 30 minutes per day. Searching breaking changes every time there’s an update (which is frequently) takes some time, as does trying out new things I’ve learned from time to time. And, when there is a breaking change, it can be minutes or many hours before I’m running again.
So, if reliable means “set it and forget it,” no.
If you mean hardware reliability, that’s certainly possible. There’s been a lot of discussion about what to use for hardware here. Basically, you can build it to the level of reliability that you need. But it’s very likely your roll-your-own solution will cost more than something off the shelf. Especially if you add in the value of your own time.
Obviously I think very highly of HA and the great developers who are making it happen. But I’m also realistic about where it is in its life cycle right now.
Wow! - I am amazed how friendly and helpful this community is. Thank you everyone for your valuable input based on real experience.
Apart from wanting to automate my own house, I am an AV installer and some of my clients have begun to ask me if I offer a nice home automation solution. I did computer science in high school and have a good general understanding of programming and networking so it should not be too hard to setup HA and learn a scripting like language. I’m big on UI and UX because that is how a human interacts with the system regardless of how good it is under the hood. A simple and elegant UI always adds credible value to a system.
Taras, you have a good point and I would not mind tinkering with my own system endlessly because I enjoy it. However, it’s a bit of a different story when you are commissioned to do a job. Reputation is as important as the solution one applies if not more.
I think that if you have a client who wants “things automatyfied” (it a real word honest ! (would I lie to you ? ))
Then they will also want a glitzy front end.
Some guys have spent 18 months tweaking their front ends and they still aren’t finished.
This kinda defeats the object of home automation - stuff should just happen as you need it to and when you need it to. Most peoples lives are pretty consistant right up to the point when they change and then all hell lets loose. Your kid moves from one school to another so your wake up routine changes, your zones change, your notifications need to change etc etc etc.
This is stuff we don’t even consider, we just make the changes. What is your client going to do ? Pay $2,000 to have someone come in and change some settings ?
What about a bulb going ? is it a dumb dimmable or is it a hue/ikea/etc RGB/colour temp variable one that needs pairing before it will work and to be incorporated into a light set or scene ? What do you mean $500 to change a light bulb ?
Just things to consider
EDIT: Industry counts HIGH Voltage as anything above 600v
Extra Low Voltage is 55v or less
Everything else is Low Voltage
Very good point Mutt - Even I would be upset if I was charged hundreds of dollars every time I needed to make a change, that is not a way to make money and I refuse to take advantage of clients from that perspective. In my case, small changes would be free or I would charge very little money just to cover costs. However, if a client requested for a major change or expansion, then I would charge a very reasonable fee.
Having said that, I do agree with you. There should be flexibility but not at the expense of an over adventerous client who has the potential to bring-down the entire system and hold the installer responsible.
I agree. Home automation is a hobby for me (for ~15 years now) so ‘tinkering’ is part of the game but if I had to do this professionally, then I would not want to do nearly as much tinkering. Like you said, reputation is important and it would be undermined by having to repeatedly login (or worse, return to the customer’s site) to adjust and re-adjust malfunctioning code or devices.
Pros usually choose paid, commercial solutions because some offer priority technical support for professionals. In the case of Home Assistant, you would have as much priority as I do, which is to say none at all. Solutions to problems are resolved by volunteers according to their schedule and no one else. Solutions to complex problems may take days to resolve (and that’s not ideal when you have a customer waiting).
Those are words of wisdom Taras, it’s like you’ve read my mind. I really like HA even though I know little about it. The affordability aspect is super attractive. However, reputation comes first.
All major vendor tie you to expensive proprietary product with no support documentation.
After install, change is virtually impossible without huge cost
support if break has huge cost
product not future proof. Imagine 10 yr old tablet to control system
HA eliminate these issue. At best you can support and update at lost cost forever. At worst you spend less and have supportable product with same.service quality at major vendor.
If you have huge cash to throw away go with Creston and such. If you want working system use HA and PLAN system well.
HA works great if you PLAN and DESIGN system well.
Thank you tmjpugh, you have some very valid and important points.
Could you kindly expand on that very useful statement that you made: “PLAN and DESIGN system well.” How could I go about doing that?. Just to let you know I am meticulous when it comes to getting things right so I appreciate your input.
By the way, I love Crestron Home OS3, solid solution and an excellent UI and UX but I’m not willing to throw away hard earned cash.
tmjpugh, you are right actually, I would never compromise on the DESIGN and PLANNING aspects but would like to know what you mean when you say that. It will certainly help me to grasp the concept better once you explain.
Actually, that is the exact information that I have been looking for. Retro fit projects would obviously need excellent WiFi infrastructure but I would prefer wiring everything (High quality shielded CAT-6 properly terminated) in new builds. I often incorporate managed 1 gigabit switches wired to 11AC APs. I would rather spend a little more and get quality hardware: switches, dimmers, cameras etc that are fully supported by HA.
Is the Rasberry Pi4 running HAS.IO / HA as good as any controller by C4 or Crestron or is there an alternative device that is more resilient ?
could make a lot of writes to its database (for various reasons) so if you run it with SD card, it’ll fail sooner or later.
Dodgy PSU is another reason for glitches.
Nothing is perfect.
Btw, did you search this forum? I remember few very interesting discussions about using HA as a base for commercial home automation.
You have a good point there Ahmad. Is there any other alternative to Pi4 that has a better PSU and an inbuilt SSD synonymous to controllers made by the commercial providers ?
