The state of the nation (so to speak)

[JanJoh 1]

Ok, so Fibaro launched the HC Lite and it seem kind of angled as a version of the Zipabox-system. Great news, especially if they will release some generic I/O-modules for it.

There is also a new multisensor, also great news!

However, I am more curious about these questions:

Why is there still not a good official LUA/System documentation even for 3.x-release?

Why is so many bug more or less met with "Will be sorted in 4.x" with no further detail

Why is the user interface so inflexible? (Take a look as HSTouch for the homeseer-product range to see what I mean)

I commend Fibaro for Always wanting to move ahead to new and exciting things... but howabout sparing a thought of actually maintaining the Products they have already released? I am serisouly considering changing my controller...

[HA tester 1]

I know what you mean, I feel a bit the same.

For me it is not really that I'm dissapointed in Fibaro but very dissapointed in z-wave.

Zigbee (with Zipabox for example) is no solution neither because there are not that many modules with HA-standard, most of them are branded propriarity.

In some months I will start to test HA based on RF eventhough I'm also expecting problems.

I don't think I will find any affordable HA-system that will satisfy my needs soon.

Please login or register to see this image.

/emoticons/default_icon_sad.gif" alt=":-(" />

[Lauri 6]

what needs do you have that z-wave cant fullfill yet? For me, it seems that Zwave isvery powerful and you can do almost everything with HC2 and different z-wave modules from various manufacturers.

[HA tester 1]

I agree with you Lauri that HC2 is quite powerful if you understand LUA (or get help with that here on forum).

First of all I prefer an adaptive system (whithout the hassle of complex programming).

I want a system based on presence (combined with motionrate), I can think about a bedroomexample but I won't, but imagine: vacuming a livingroom is different than watching TV. In both cases I am in the room but I might want the temperature to be different in both cases.

Besides presence I want it to be based on temperature, lux and sound (and for outdoor also wind, rain and uv (I don't care about the weather from Google, Yahoo or whatever station)).

Is there any z-wave doorlock suitable for 3 deadbolts?

And ofcourse for a reasonable priceticket

Please login or register to see this image.

/emoticons/default_icon_razz.gif" alt=":-P" />

[JanJoh 1]

what needs do you have that z-wave cant fullfill yet? For me, it seems that Zwave isvery powerful and you can do almost everything with HC2 and different z-wave modules from various manufacturers.

For me: Z-wave can fulfill it. But the HC2 cannot.

There is no proper documentation. There is support only for very few devices. Everything is extremely sandboxed. You cannot define custom user interfaces that actually interacts with the system.

[ Added: 2014-03-03, 22:46 ]

I agree with you Lauri that HC2 is quite powerful if you understand LUA (or get help with that here on forum).

First of all I prefer an adaptive system (whithout the hassle of complex programming).

I want a system based on presence (combined with motionrate), I can think about a bedroomexample but I won't, but imagine: vacuming a livingroom is different than watching TV. In both cases I am in the room but I might want the temperature to be different in both cases.

Besides presence I want it to be based on temperature, lux and sound (and for outdoor also wind, rain and uv (I don't care about the weather from Google, Yahoo or whatever station)).

Is there any z-wave doorlock suitable for 3 deadbolts?

And ofcourse for a reasonable priceticket

Please login or register to see this image.

/emoticons/default_icon_razz.gif" alt=":-P" />

Well, I can code Lua quite well. (And also PHP, Python, Powershell, bash, C++, PERL, Assembler, various PLC dialects etc etc). So, that is not the issue.

The issue (again) is that Fibaro seem more interested in pushing forward with new things then to actually maintain what they have.

And yes, I have realized that Homeseer Pro is the solution that I should have gone with from the beginning.

Yes, I realize that I have a background that is not really the norm here (my first job after Highschool twenty years ago was designing hardware for mobile phones, and it has kind of moved in that direction). I have also designed the hardware and software for controlling my outside hot tub and other such things.

I chose the HC2 because i thought that the User interface would be the one my girlfriend was most happy with out of the available options. Turns out that she is the one that really does not like it (Why cant you have... Can you move this.... I want it do do this...)

As for user interfaces.. do a Google Image Search for HSTouch and you will see my point. Combine that with stuff like:

Please login or register to see this link.

and i personally would be having some serious fun...

[Dalle1985 16]

Well the whole issue is that home automation is extremely nerd'ish. It is bought by nerds and it is developed by nerds. A piece of advice: if you want a home interface that your girlfriend will like, you'll have to build it yourself (check out my link below).

HSTouch has the exact same problem as HC2 it is far too busy doing everything in one screen.

My setup does most of what you want HA tester. When you turn on the television and the light level is below a certain threshold, it will set all lights to "Hygge" with low lights and a warm color. If we are in the kitchen for more than 60 seconds, it will change the color tone to a more adequate work-setting.

Same goes if you plug in the vacuum cleaner (it has it's own fibaro plug permanently mounted). If the lux-sensors meassure a light reading below 75 lux, it will automatically turn on the light in any room where presence is detected.

In the night, if you have to go to the toilet and turn on a bedlamp, it will only turn it to 15% and furthermore, it turns on one of three lamps in the room you have to pass through at a very low setting, and ofcourse a lamp in the bathroom at low power. This way you can see what you do but won't have to be blinded at 2 in the morning.

And ofcourse all rooms have presence detection to automatically turn on lights when you enter and off after you leave - all based on mode selection.

Any component which sends signals over wifi and has an api can interact quite well with fibaro. Right now, I have a netatmo weather station, several hue bulbs, my television and a few other components communicating flawlessly with my setup.

As for temperature, I think you are expecting way too much fom your heating system. Any heating system will have a certain inertia, meaning it takes time to change temperature. I've been following the temperature control thread on here and can't help but chuckle a little. Trying to control room temperature based on wind, light influx and outside temperature will be like trying to drive your car, only based only on what is happening a kilometer down the road. You need to accept the fact that the system has a certain inertia and the fact that you can only base your temperature on what is happening right now - not what you expect is happening. The PI-controller in the thermostat will compensate for any additional energy requirement if the sun stops shinning or the wind suddenly picks up. If you try to do its job for it, it will either end in a cold room or a huge energy bill, so make sure that you base your expectations on what your heating system can actually do, instead of trying to force it to do something it can't.

If you have airconditioning inertia is going to be small, if you have radiators, you will have medium inertia and finally, if you have floor heating you'll have huge inertia and in that case, not even night lowering will be adviceable as the energy saved during the night will be lost during the day trying to heat back up your heat mass in the floor.

My setup is based on radiators and when i initiate the night dimmer (which turns off all lights except the bed side lamps which dim out over 15 minutes) it will go to a night time setting throughout the house. At 5 in the morning, it increases the temperature again so that the temp is nice when we get up.

If no presence has been detected in a room for 12 hours, it will turn down the temp by 0,5 degrees. After an additional 12 hours it will reduce 1 degree and after further 24 hours, it will turn the temp down by another 2,5 degrees. As soon as presence is detected, it will jump back to setpoint.

I of course also have a vacation mode and different weekend modes.

What i'm trying to say, is that i think that combined with a bit of lateral thinkingm the hc2, z-wave and a bit of http communication, you can do almost anything that is needed in a normal home. And quite a bit that isn't needed

Please login or register to see this image.

/emoticons/default_icon_biggrin.gif" alt=":-D" />

But yes, documentation and updates should be better...

[JanJoh 1]

HSTouch has the exact same problem as HC2 it is far too busy doing everything in one screen.

But, you can set up multiple screens quite easily, and that is something i would appreciate highly

Any component which sends signals over wifi and has an api can interact quite well with fibaro. Right now, I have a netatmo weather station, several hue bulbs, my television and a few other components communicating flawlessly with my setup.

Well, until we've seen what the API in 4.x brings us, i would not call it integrating as you can still not pass arbitrary values to core functions.

[HA tester 1]

For me: Z-wave can fulfill it. But the HC2 cannot.

example from my situation:

I walk into a room and I can see the led of the motionsensor blink, one step further I can hear the relay click and then the light turns on.

Is this delay due to z-wave? Fibaro? Or the motionsensor? (in my opinion it is z-wave because same problem exists with other brand gateway and other brand motionsensor)

There is no proper documentation. There is support only for very few devices

I have the same feeling with other gateways.

You cannot define custom user interfaces that actually interacts with the system.

I don't care about GUI. I want to be able to add device or module, put settings to my needs and it must be working. I don't need shiny apps for tablet(s) or phones, it is called automation if I need to touch or swipe on a screen I might as well use the wallswitch

Well, I can code Lua quite well. (And also PHP, Python, Powershell, bash, C++, PERL, Assembler, various PLC dialects etc etc). So, that is not the issue.

I don't have adaquate programming experience, but why should I?

I see all kinds of fantastic scripts here so why are there no more options with the graphical blocks and they should be litlle bit more user (noob) friendly.

I already sold one gateway, two more to sell and meanwhile I will look for something completely else.

[ Added: 2014-03-04, 22:00 ]

Well the whole issue is that home automation is extremely nerd'ish. It is bought by nerds and it is developed by nerds. A piece of advice: if you want a home interface that your girlfriend will like, you'll have to build it yourself (check out my link below).

I'm not nerd (enough). I bought the system because I want to have things automated my life is already busy and stressfull enough (you can call me lazy, I don't care) and I want to save some energy (just for my wallet not the environment)

My setup does most of what you want HA tester. When you turn on the television and the light level is below a certain threshold, it will set all lights to "Hygge" with low lights and a warm color. If we are in the kitchen for more than 60 seconds, it will change the color tone to a more adequate work-setting.

Same goes if you plug in the vacuum cleaner (it has it's own fibaro plug permanently mounted). If the lux-sensors meassure a light reading below 75 lux, it will automatically turn on the light in any room where presence is detected.

In the night, if you have to go to the toilet and turn on a bedlamp, it will only turn it to 15% and furthermore, it turns on one of three lamps in the room you have to pass through at a very low setting, and ofcourse a lamp in the bathroom at low power. This way you can see what you do but won't have to be blinded at 2 in the morning.

And ofcourse all rooms have presence detection to automatically turn on lights when you enter and off after you leave - all based on mode selection.

I don't no in what topic I mentioned it before but I can't use dimmer values below 50%

I'm curious about your presence detection.

As for temperature, I think you are expecting way too much fom your heating system. Any heating system will have a certain inertia, meaning it takes time to change temperature. I've been following the temperature control thread on here and can't help but chuckle a little. Trying to control room temperature based on wind, light influx and outside temperature will be like trying to drive your car, only based only on what is happening a kilometer down the road. You need to accept the fact that the system has a certain inertia and the fact that you can only base your temperature on what is happening right now - not what you expect is happening. The PI-controller in the thermostat will compensate for any additional energy requirement if the sun stops shinning or the wind suddenly picks up. If you try to do its job for it, it will either end in a cold room or a huge energy bill, so make sure that you base your expectations on what your heating system can actually do, instead of trying to force it to do something it can't.

If you have airconditioning inertia is going to be small, if you have radiators, you will have medium inertia and finally, if you have floor heating you'll have huge inertia and in that case, not even night lowering will be adviceable as the energy saved during the night will be lost during the day trying to heat back up your heat mass in the floor.

My setup is based on radiators and when i initiate the night dimmer (which turns off all lights except the bed side lamps which dim out over 15 minutes) it will go to a night time setting throughout the house. At 5 in the morning, it increases the temperature again so that the temp is nice when we get up.

If no presence has been detected in a room for 12 hours, it will turn down the temp by 0,5 degrees. After an additional 12 hours it will reduce 1 degree and after further 24 hours, it will turn the temp down by another 2,5 degrees. As soon as presence is detected, it will jump back to setpoint.

I of course also have a vacation mode and different weekend modes.

I'm not expecting to much from my heatingsystem and I do understand inertia and hysteresis.

I meant with my example: If roomtemperature is 20 degrees and I am cleaning the room I probably want the heater set to 17 degrees (I will get warm because of physical action) but when I sitting on a couch watching tv I might feel getting cold and raise the temperature. since I only have the possibility of detecting motion and not presence (including amount of movement) and can't do things I want.

For my outdoor part I want real weatherinfo from something like netatmo and not the forecast from somekind of station 25 km from my home wich you can pick from the web

What i'm trying to say, is that i think that combined with a bit of lateral thinkingm the hc2, z-wave and a bit of http communication, you can do almost anything that is needed in a normal home. And quite a bit that isn't needed

Please login or register to see this image.

/emoticons/default_icon_biggrin.gif" alt=":-D" />

Maybe you are right but I'm no nerd so it is useless?

Fibaro claims: your room, your imagination

Without proper knowledge I'm stuck in imagination only

Your project looks promising, but I don't have and don't want a running webserver.

[Dalle1985 16]

Well without a webserver, you will be stuck with what fibaro can deliver... And that is (pardon my french) crap. I have just invested in a raspberry pi for my webserver. It is running at 3W power supply and doubles as an airplay station for my music. It has cost me the chiefly sum of 45€ including case and a 32GB memory card to facilitate a large mysql database. Anyway, the short of it is that i have an almost zero footprint webserver which takes care of all user interaction through some dirt cheap android tablets.

Setting up the pi is very simple. Once my project is complete, I'll compile everything together both the software i'm running and the needed hardware and create a homepage with some how-tos.

As for the nerd part, going into this, I had no. Lua knowledge, no PHP knowledge, no javascript knowledge, no SQL knowledge and only a head full of ideas, and what I started 6 weeks ago. So though it might seem daunting going in, this stuff really isn't that difficult... But a little programming skills expand the possibilities into infinity.

As to presence detection, it is based on different inputs, but mainly motion. The motion detectors are sensitive enough to register even the slightest movement. But it is overridden by other components, i.e. If a television is on it will assume presence in the room unless it detects no motion for 3 hours - then it will turn off the television.

Anyway, i believe that the reason for the poor documentation and the slow support in some cases is because fibaro is trying to achieve exactly what you long for; simple application for the end user. However, making it simple also voids some of the really clever - but nerd skill requiring - stuff you can do with the system. So for most users who aren't nerds (remember, being a nerd is a good thing - we rule the world) a professional would be their best bet for a really nice install, but that is going to be very expensive and will intimidate most buyers. So what fibaro should do, is provide the advanced tools we (the nerds) need to create the proper integration and then use us to develop all the cool solutions which can then be installed into the home center as plugins - for instance a custom web interface or an advanced lighting controller. The products that prosper today are the products with strong community support and therefore the manufacturers should cater to the communities need, then let the sales department do the end user development and use the entire power of the interweb for development.

But i do still believe that putting a bit of effort into learning the ropes will increase your possibilities tenfold. My system is already - after 6 weeks - easy enough that my girlfriend can understand it, but still way more advanced than what any single proprietary system can deliver. Even with the poor documentation!

[ Added: 2014-03-04, 23:10 ]

HSTouch has the exact same problem as HC2 it is far too busy doing everything in one screen.

But, you can set up multiple screens quite easily, and that is something i would appreciate highly

Any component which sends signals over wifi and has an api can interact quite well with fibaro. Right now, I have a netatmo weather station, several hue bulbs, my television and a few other components communicating flawlessly with my setup.

Well, until we've seen what the API in 4.x brings us, i would not call it integrating as you can still not pass arbitrary values to core functions.

Multiple screens is exactly the problem. You need to have one screen to control your key functions - thats it. Otherwise, if you have to go through three pages to turn off a lamp after you've turned down the temperature. It will just be easier to go to the wall switch and that pretty much defeats the purpose of having the screen in the first place.

Regarding integration, then I don't care about being able to communicate directly to core functions and i'm pretty sure that the missus don't either, as long as she can control the components in an easy way. The interaction behind the scene really isn't important imho as long as it does what you want.

[JanJoh 1]

Multiple screens is exactly the problem. You need to have one screen to control your key functions - thats it. Otherwise, if you have to go through three pages to turn off a lamp after you've turned down the temperature. It will just be easier to go to the wall switch and that pretty much defeats the purpose of having the screen in the first place.

Regarding integration, then I don't care about being able to communicate directly to core functions and i'm pretty sure that the missus don't either, as long as she can control the components in an easy way. The interaction behind the scene really isn't important imho as long as it does what you want.

You are thinking device-centered UI aparently, I am thinking task-centered UI.

Does one need a single main control screen? No, not really considering you would have several control points.

But, what I do expect is to be able to design a UI from a task centered point of view.

So, when my system becomes aware that I am watching a movie, i expect my main screen of a remote to change accordingly and display the devices/scenes/etc that is pertinent to that activity.

[ivan.ruban 3]

The issue (again) is that Fibaro seem more interested in pushing forward with new things then to actually maintain what they have.

That is the biggest the problem with Fibaro. They are focused sometimes on delivering new things without caring too much about current problems. As a result a lot of bugs in bugzilla opened for years.

I chose the HC2 because i thought that the User interface would be the one my girlfriend was most happy with out of the available options. Turns out that she is the one that really does not like it (Why cant you have... Can you move this.... I want it do do this...)

Take a look at

Please login or register to see this link.

They are focused on UI - you can customize it quite simply. I tested them on RPI with z-wave module (raZberry). It worked almost fine. But I didn't get to work scripting part (LUA-like). And I also missed other functionality like f.ex. heating panel. That's why I moved to HC2 finally (after testing also Zipato and Vera by the way)

Well without a webserver, you will be stuck with what fibaro can deliver...

Yes, you have to be a web developer to create custom GUI in HC2. But it is anyway possible with HC2's API! Read this:

Please login or register to see this link.

Besides presence I want it to be based on temperature, lux and sound

RGBW controller and anaolgue sensors will help with that.

What I really would like and what is missing. Real time speech recognition - so I can say "turn on light" without pressing any button! If LILI is now a premium feature not available in Lite version, please improve it and make a real time.

My setup does most of what you want HA tester. When you turn on the television and the light level is below a certain threshold, it will set all lights to "Hygge" with low lights and a warm color. If we are in the kitchen for more than 60 seconds...etc

Dalle! Thank you so much for good scene examples. It is always difficult to find good scenes!

So, when my system becomes aware that I am watching a movie, i expect my main screen of a remote to change accordingly and display the devices/scenes/etc that is pertinent to that activity.

+1 That is definitely missing and I want it!

[Dalle1985 16]

Well without a webserver, you will be stuck with what fibaro can deliver...

Yes, you have to be a web developer to create custom GUI in HC2. But it is anyway possible with HC2's API! Read this:

Please login or register to see this link.

I don't agree! You just need to be willing to put a bit of effort into it. As I wrote further up, I've only been programming this sort of stuff for 6 weeks - since I've received my HC2. The fact is that people are often daunted by the prospect of having to learn to programme to do this sort of stuff... But try it and it turns out to be quite simple. With the help of one single tutorial, it took me less than 10 minutes to set up my RasPi with a working webserver, mysql and even as an airplay receiver for my non airplay stereo - with zero experience in doing so. I'm no programmer - far from it... I'm a marine engineer used to working on ships and now i work in the oil industry...

And yes, I've spent quite some time on my web interface, but I only spent about 6 hours getting to know php. But that was actually the point I was trying to make before - this shouldn't be necessary... Because I completely agree, Fibaro should focus on their community and setting up the tools for the community to provide all the cool features. Then they can start to focus on bringing us cool components afterwards. But they do ofcourse need a customer base to ensure proper cashflow and that does take a little development. But more focus on bringing us the best possible tools (i.e. Full documentation) could benefit not just us, but also the less advanced users and thereby creating a way stronger product for all.

So, when my system becomes aware that I am watching a movie, i expect my main screen of a remote to change accordingly and display the devices/scenes/etc that is pertinent to that activity.

Well, coming back to the part about lateral thinking, I'm almost dead certain that XBMC is capable of announcing the fact that you are watching a movie. All you then need to do is make that a scene trigger and then as soon as you start playing a movie on XBMC, it will trigger a scene turning down the light, closing the blinds and whatever else you want it to do. Making a web interface which integrates both the television and fibaro into one page should be pretty straight forward. I already have a scene which forces a certain light setting based on the input source on my television (when the television is set to HDMI4 - AppleTV it will enable a movie mode). Then it's just a few scripts and a determinator variable to determine what elements to show on your remote... That should be really easy for someone knowing as many languages as you do to achieve.

I'll actually claim that I could do such a webscript in less than a week with a remote screen for a tablet, which will control lighting, room temperature (and any other zwave component) along with my television and a mediacenter based on Xbmc (including setting up an XBMC server as i don't have that now). The screen will change when the XBMC source is selected on the television and then again when you start the movie. I don't see any issues in doing that. But that is besides the point. To be brutally honest, we are back to the nerd argument then. How many ordinary users are going to expect that level of control from their home automation? And that is the core issue here. Is this product supposed to cater to the highly specialized customers with big requirements? Or is it supposed to be a mass market product supposed to target the normal user who just want some basic functionality? My guess is the later. And if that is the case, then I actually think it is quite noteworthy that Fibaro is offering the tools that they are!

[JanJoh 1]

Well, coming back to the part about lateral thinking, I'm almost dead certain that XBMC is capable of announcing the fact that you are watching a movie.

It certainly is.

All you then need to do is make that a scene trigger and then as soon as you start playing a movie on XBMC, it will trigger a scene turning down the light, closing the blinds and whatever else you want it to do.

Ah, but that is where I Think you go wrong.

Can i get the XBMC to trigger actions in HC2? Heck yes, that's easy.

But, until I am able to actually INTEGRATE the XBMC in the user interface of the controller, then that is really a moot Point.

I already pull all sorts of data into the HC2 and display them in all sorts of labels. But, that is a bit of the problem right there: Labels. That is pretty much the only way you have to pull data from external systems into the UI for HC2. Now, compare that to a HSTouch interface and tell me that the Fibaro way is not sadly lacking?

Ive already built a cpl of self sustaining temp monitors on the Electrical Imp board. But getting HC2 to recognize them as a temperature source for a room? Heck no.

How many ordinary users are going to expect that level of control from their home automation?

As with most of my arguments, I talk about what _I_ want. And I re-iterate that it looks like i have chosen the wrong system given that it seem that the priorities of Fibaro is not to provide the most capable system

Sure, the API in relese 4 may give me some of what I want... but excuse me for being sceptical, given how the state fo documentation looks for release 1 and 3... I am not overly optimistical in regards to actually being able to use it.

[1152 33]

Sure, the API in relese 4 may give me some of what I want... but excuse me for being sceptical, given how the state fo documentation looks for release 1 and 3... I am not overly optimistical in regards to actually being able to use it.

Point well made. Fibaro listen up! I'm confident I bought a very capable system, I just don't know how to make the best of it...

[Guest Kuuno]

what are the biggest things You want to achieve?

[1152 33]

what are the biggest things You want to achieve?

The thing is - it is not a list of things as such - (allthough I could of course make a list at present). The thing is: I like a better understanding of what I have bought, and how I could capitalize on it. Documentation of how things work would have been nice. I tend to find my self in a guessing game, solving some times more by trial and error than anything else. (At present I have been battling with alarms. Smoke detectors, flood detectors, temp warnings, motion, door/window sensors and the like - Alarm panel settings, seperate scenes, and device settings). Yes I know wait, but the thing is I hope the system will last for years and years, but I think I will have to be realistic and to some extent I think it is fair to say I am loosing some of the lifetime both for me and my equipment...

[JanJoh 1]

what are the biggest things You want to achieve?

In the order of importance:

1. Fibaro to actually start to FIX the bugs that has been around for ages and not just give the "oh, next release"-answer.

2. A _well documented and supported API_. That actually allows me to INTEGRATE with the system.

3. A configurable user interface for Android/iPhone

4. Global codebase (enough of having to post the same code all over the place. It makes development and debugging a flipping nightmare)

[robmac 173]

+1

[StarkJohan 6]

+1

[Brick 13]

What I really would like and what is missing. Real time speech recognition - so I can say "turn on light" without pressing any button! If LILI is now a premium feature not available in Lite version, please improve it and make a real time.

+1

That will definitively set a turning point in home automation. The company which manages to develop such a feature in a reliable and efficient way will hit the market, imho.

Any solution based on pre-activation of the device may be cool and fancy but it is not really useful, except maybe for handicapped applications.