Seriously, how can anyone trust the security of your home to this constantly failing environment? To begin with I’m running a v1 hub, and have been for quite some time now, even evangelizing about the system to anyone who’d listen, and through it all I have been continually faced with failing routines, dropped devices, undocumented bugs… it goes on and on! For most of this time I’ve stuck to connected lights, more for the convenience and love of automation they allow as otherwise my kids/wife tend to leave lights on forever, but nothing critical to my home’s existence! However, I recently decided to go the CT100 route for my thermostat using rboy’s great 5-2 thermo app, the first mission critical device I’ve deployed in my house. It worked splendid for about a month and now it too has fallen victim to the plague of failed triggers. I thought maybe I should upgrade to the v2 hub with it’s local controls, opposed to the crazy-azz cloud round-trip required on the v1 for scheduled triggers but I’ve seen many posts in here about its short-comings as well.
So how can anyone in their right mind trust their security, door-locks, etc. to the ST platform? How can this late in the game these same problems continually pop up? At first I liked the challenge of maintaining the environment, looking to my peers within the community to fill the gaps left by official apps and device types, but now I just want something that works. I want it to be a transparent experience. Yeah, yeah, it’s open… it’s developer friendly… it’s all of these things, it’s also frequently a terrible mess as well. I want to go all-in, but sadly at this point it doesn’t seem to be a long-term solution for me. Anyone else at the point of feeling the same pain? What other systems are piquing your curiosity right now? Hell, I’m even considering trying the cheapo new Iris hub out for size!
There have been many discussions of reliability and SmartThings really, almost from the beginning.
Personally, I have never put anything On SmartThings except convenience use cases where I had an immediate plan B if it failed. I have a completely separate, and very reliable, security system that I’ve used for years and I’m very happy with. But I’m willing to pay a monthly fee, and many people are not.
As far as alternative home automation systems, again, much discussed in the forums.
If you can afford to pay $10,000 a room, there are some great systems out there, very reliable, local processing, tie everything together. That’s considerably outside of my budget however as it is for many people here.
If you’re looking at the low-end systems, where SmartThings is priced right now, you’re probably looking at a budget of $200-$500 room. And not every room in the house. In that price range, there is no one perfect system yet. Everything has pros and cons.
Staples connect made reliability the number one priority from the beginning, but they got that reliability by throwing out many features that other systems consider essential. There’s no there’s no IFT TT, there’s no voice control. No Geo fencing. A very limited choice of devices, and most of those most expensive in their groups. But if you just want something that turns the lights on on the schedule and opens the curtains and works every day, it’s a good choice.
There are also several excellent security systems under $2000 that are reliable and have a monthly fees of $20 a month or less. But they don’t typically tie into your other home automation, which is how they get their reliability. Some do have an IFTTT channel which gives you some additional notification options. But they don’t let you fooling around with schedules for Halloween decorations mess up your intruder alerts. Again, personally I’m fine with having a separate system for security monitoring, but for some people that’s a dealbreaker.
Again, search the forums for any other competitor that you’re interested in and you’ll find a lot of discussion.
My own expectations has been that by the summer of 2016 there will be several reliable plug-and-play systems that fit my own needs and budget. HomeKit/Insteon will clearly be one. Something based on nest will be another. My expectation is that SmartThings May be a third, but obviously they’re going to have to make a lot of improvements to get there. I’ll definitely look at them as a potential candidate in late spring.
For right now, you can put together a very nice, very inexpensive, very reliable system just for lights based on Phillips hue and Echo and then add other things later. But I know most people don’t want to wait.
I still don’t personally expect to put security on to that system but I have my own requirements for security.
Different people will have different priorities and come up with different answers.
You might find the following conversations of interest.
The first one is specific to security.
The second one is my own project report on what I did for my phase 1 and what happened.
I’ve never once had door locking issues and have 3 Schlage touchpad deadbolts. Several months ago I did experience issues with unlocking on presence arrival as part of a routine. I moved the operation to the device level which worked 100% of the time. I have since moved this back to the routine level with no issues for a couple months now.
I’ve never once had a thermostat issue with my Ecobee3 using Yves code unless it was to update the code as the old code was no longer operational in some way.
I have had issues with sunrise / sunset which were well documented by many users on the forum. This was fixed a couple months ago and I’ve had no issues since.
I also had lighting issues which I later learned were caused by me and not by ST…ate crow there.
Now this is my experience and is not meant to insinuate that if yours differs it is something you are doing wrong.
Likewise, your experience doesn’t mean that others are experiencing the same; therefore, I am in my right mind
(ActionTiles.com co-founder Terry @ActionTiles; GitHub: @cosmicpuppy)
I don’t. What I’ve found with ST is to keep things simple. Adding more stuff and playing with SmartApps just calls for trouble, given the current state of the platform. I truly believe they are going to improve. Question is how long until then?
Yup, after the API timeout issue this morning, I said good bye to all my external API integrations. I cannot afford to keep troubleshooting my lights, if I also want to stay employed. Game on ST, I submit to your will. Let’s see how well you fare now. I have all but official apps and devices.
Keep in mind, that much of this is dependent upon multiple other technologies (it’s not necessarily a ST issue.) e.g. there could be issues/conflicts within the Zigbee network, or within the WiFi network, etc.) ST indeed “inherits” these issues. Ditto for web-based APIs and such to IFTTT and such (at any given time, we have no idea what might be up/down out in the land of the Internet…)
If you want a TRUE home-security system, then you need something that’s hardwired (with plenty of alternate/backup cabling & backup power, etc.) Anything else/less, will inherit issues. The more technologies/protocols/vendors/manufacturers/networks you try to introduce, the more potential hiccups you are adding to the solution.
ST tries to be the glue that ties all of this together (and does a pretty good job at it.) But, ST is indeed an “extra step/element,” too. So, it also introduces it’s own possible issues/hiccups/glitches.
One way I’ve been trying to overcome some of this, is via redundancy and double-checks. e.g. TWO triggers to cause an alarm – instead of just one. Ideally, each of these triggers are via a different protocol/network type. That way, you don’t have all of your eggs in the same/single basket (that ONE hiccup can take-out BOTH devices/triggers.) In some cases, I circumvent ST entirely with by “second alert.” So, if/when ST “alarms” based on an event, I check my “secondary alarm,” to “confirm” what has occurred – before I panic.
e.g. ST alerts that my front door opened while I was away at work. ISSUE!!!
So, I check my Netatmo Welcome camera, and I can see that the door never opened, and I indeed confirm it via the video replay. (false alarm.)
But, if the Netarmo Welcome DOES reflect an “motion event,” too?!!! Then, I have a double-confirmation that something’s up!
So, be careful about throwing stones at ST. Many of the "issues’ that we ALL experience, are simply “inherited” by ST. I’m NOT saying that ST isn’t without its own issues, and is flawless. I’m simply saying that ST is a DECENT piece of “glue” for most of these technologies to play-well together. ST is NOT an “enterprise-class” system for banking security, or maximum-security prisons, etc. They aren’t promising 99.999% uptime/availability.
We (the People) take a good thing, and simply try to do TOO MUCH with it. ST was designed as a tool/hub to help all of your home automation stuff inter-communicate. It wasn’t really designed to be a “home security” system.
With all due respect, my major was computer information systems and I worked as a network engineer at IBM. I had a lot of real world experience with Zigbee and zwave before ever buying SmartThings.
Like many people in the US, I have a zigbee-based security system which has been solid and reliable for six or seven years. No problems.
Indeed, one of the reasons why I chose SmartThings initially is because they were using two proven and reliable technologies in Zwave and Zigbee. I knew what was likely to go wrong. I knew the likely fixes. I set my expectations accordingly.
None of the problems that I have seen over the last year with my SmartThings installation have been due to network protocols with the single exception of the zigbee presence sensor which runs into some local interference from the neighbors boosted Wi-Fi.
But not running scheduled events when scheduled has nothing to do with the protocol. Randomly unlocking my Zwave door lock is not a zwave fault. Randomly triggering a virtual switch obviously has nothing to do with the protocol-- The switch has no RF protocol. It’s virtual.
I run multiple RF systems in my home now, including echo, the Phillips hue bridge, the sprinkler system, a medical monitoring system, the zigbee security system I mentioned, etc.
Only SmartThings has this level of instability and requires this level of maintenance.
I don’t doubt that there are people who have it running perfectly with no problems. And there’ll always be some problems, like my neighbors Wi-Fi, which are just local interference. But I also have no doubt that most of the reliability problems with SmartThings are not due to the network protocols.
I do… BUT… I wasn’t trusting security to anything before smart things, so it’s really just an extra benefit. I didn’t have any sort of Alarm before, so even a fallible system now is better than what I had before.
As someone who works with managing IT systems, this is a very important piece to remember.
I’ve had people ask me before: “How do glitches happen? Aren’t computers just designed to do what you instruct them to do? If so, why do they sometimes not do it or do something else?”
And the number answer for me is that Merlin wrote above: Complexity. The more complexity you add into a system, the more chance there is for something to go wrong. The more different systems you have talking to each other… the more instructions you have that might end up conflicting with each other… all more users who are using the same back end…
Think of a football team (American football for you on the other side of the pond). The play calls for the receiver to run a deep route along the edge of the field. The play happens and the ball ends up hitting the receiver in the back of the helmet. What a stupid receiver, right? Why wasn’t he looking for the ball?
Well, the Quarter Back dropped back, but the TE missed his block, which means the OT tried to pick up that rusher as well as the guy he was normally supposed to block. Well, his normal guy takes advantage of that and slips by the OT and gets to the QB much faster than anticipated which means the QB has to throw the ball sooner than he wanted too which means the receiver never turned around because the ball wasn’t supposed to be thrown to that spot. So the receiver wasn’t the problem, it was the TE that cause the problem.*
This doesn’t absolve ST of any responsibility of course. First, some of the issues are their own fault. Second, if they are advertising that they can work with other systems, there’s an expectation that they will be able to do that consistently. So yeah, ST is at somewhat to blame, we do need to be somewhat realistic with expectations as you say Merlin.
(*yes… I know that technically the receiver should still have been looking in case of a broken play and so he was somewhat at fault, but it was only an analogy.)
I’ve started my HA hobby on one hand because I wanted some insight in my home for security purposes and on the other hand to have control over my lights. I had ADT for years in my old home and when I first learned about Wink, I thought I discovered the gold. It seemed so doable to use one device to serve both purposes. Well, things weren’t going well, so I started looking at alternatives. And that’s how I got hooked on ST. The CES announcement back in January, coupled with extremely responsive support and an overwhelmingly knowledgeable community made the future of ST look so promising. I now realize that IoT remains the best hobby, but there is not one company that can live up to the expectation of delivering a rock solid DIY security system and to manage the lights at the same time. And most certainly the quest for a more viable solution has been heightened by the recent platform blunders. We are just not there yet!
I bOught 2 of the original hubs offered via Kickstarter. I’ve been treating them as extended beta since then. Too many false alarms have prevented me from trusting them. Just recently I changed a battery in an original door sensor and now the system won’t recognize the sensor. I want to love the system and expand my use, but I just don’t trust it.
I’m considering resetting one hub and reinstalling things and routines. Has anyone else had any real success at improving reliability by taking this approach?
Candidly, I bet we will see a reliable wireless home automation system when Apple takes a serious approach to this technology.
I was hoping the update, SmartThings emailed me about, would help address the issue of inconsistent performance. From day-to-day, I never know what lights will go on or off and when they may do so. It’s like lighting Russian roulette. Of course, after I contacted them, I was informed the update had nothing to do with my hub.
SmartThings’ performance seems to be getting worse, not better. This is causing me to hesitate to add any new devices and to stay far away from controlling anything whose function might be considered critical. And, should I invest in the new hub? Why would I when SmartThings has demonstrated they’re either not interested in offering a reliable product or incapable of doing so. How can Samsung let this go on?