Loxone and KNX.... what now?

[Julián Medrano Silvestre]

So... with the (already known) issue of KNX integration dissapearing from Miniserver Gen2 and the INCREDIBLE price of Loxone KNX Extension....

What are the alternatives for those of us that were using Loxone as "smart brain" for logic and KNX as dummy devices for inputs, outputs and "elegant" room controlers?

I'm now in deep depression....

[Rob_in]

Well, without KNX Loxone is dead to me.

I'll bet this is a business/marketing driven decision versus a technical one.

Given how cheaply one can buy a Rasberry Pi and add KNX connectivity (yes - it can be done cheaply and easily - Google it) I would find it incredibly difficult to justify paying for Loxone if starting out now. Especially given the route they are going down.

Yes, very sad indeed.

Robin

[Russ]

Yes, for european installations, better to go with KNX and drop Loxone.  KNX prices have come down, many manufacturers (370 with 7,000 products according to KNX site), and the basic paradigm (distributed intelligence with centralized control) is better anyway.  I think the Loxone software is still better than most of the KNX visualisation software out there, even if it still looks a bit clunky, but again, there are so many choices for KNX you're bound to find something you like. 

KNX gateways/interfaces for Dali, DMX, 1-wire, ModBux, etc.  all out there.  As long as you stay away from proprietary Loxone protocols (Air...) you can have the same functionality or replace your Loxone install with KNX if it comes to that.  If I was starting from scratch, KNX would be my choice, even with the annoyingly high price for the config software.

I did my wiring with KNX cabling and star topology exactly for this reason.

[Stefan L]

But is not the "server"  still more expensive or at least the same price, I think Thebens alternative to the miniserver is like €1,000? And then you still need the KNX programming etc etc. 

[Andras]

If you regard Loxone as the UI and configuration platform, I think it's still among the best performance and value out there. Vendor lock-in? Ofcourse. Luckily there's quite a few ways around it.

I have setup a raspberry pi running nodered as the integration platform: it talks to KNX (through a €100 IP gateway), it talks to Loxone, it talks to mqtt, it talks to alexa, ... runs flawlessly for a year now and it's rather easy to maintain.

On Tuesday, November 26, 2019 at 7:26:48 PM UTC+1, Julián Medrano Silvestre wrote:

[leo no]

If loxone drops server gen1, im done with them, Zennio Z41 pro cost 600 euros and lt has everything loxone offer, so entire knx will be the way, or maybe loxone offers both gen1 and gen2, like does with miniserver go.

--
You received this message because you are subscribed to the Google Groups "Loxone English" group.
To unsubscribe from this group and stop receiving emails from it, send an email to loxone-englis...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/loxone-english/fa56819f-d3fc-4f93-9f10-41ec25ef013f%40googlegroups.com.

[BartVB]

I've been planning the home automation system for my (to be built) house for quite some time now.

When I started evaluating options (2012) Loxone was one of the best contenders. But in the last couple of years they really dropped the ball if you ask me. They closed the community, very strong focus on their proprietary Tree and Air protocols, no improvements in KNX integration, etc, etc.

I'm using a Loxone miniserver in my current house but chances are getting very thin that it will move to the 'real' house. Need to start evaluating alternatives. Was already planning on using more KNX components (presence, buttons, CO2 sensors) so moving away from Loxone shouldn't be that hard.

I kind of liked the ease of use of their programming interface. But after playing with that for our current home I'm starting to become more and more aware of the limitations of this closed platform. Even really simple situations require very elaborate or creative workarounds. And there is no way to add custom functionality except through something like Loxberry. If I need a SBC anyway then why not move to something like NodeRed completely?

I'm very interested if you know of other platforms or products that I should look into when evaluating Loxone alternatives.

Kind regards,

Bart

On Tuesday, November 26, 2019 at 7:26:48 PM UTC+1, Julián Medrano Silvestre wrote:

[Rob_in]

On Wednesday, 27 November 2019 12:40:40 UTC+1, leo no wrote:

If loxone drops server gen1, im done with them, Zennio Z41 pro cost 600 euros and lt has everything loxone offer, so entire knx will be the way, or maybe loxone offers both gen1 and gen2, like does with miniserver go.

I can find that Zennio Z41 Pro for under 500EUR and it actually and it looks pretty interesting. I have had some dealings with Zennio as have a KNX interface to a Daikin heat pump and find their support pretty good.

That said, would certainly have to look into it more before knowing if it could replace a Miniserver... which I'm not inclined to do right now as don't need that (yet!).

On Wednesday, 27 November 2019 14:37:58 UTC+1, BartVB wrote:

When I started evaluating options (2012) Loxone was one of the best contenders. But in the last couple of years they really dropped the ball if you ask me. They closed the community, very strong focus on their proprietary Tree and Air protocols, no improvements in KNX integration, etc, etc.

Exactly. Loxone will die because of this. People want to buy products that follow standards where if the manufacturer disappears there are others following that same standard where you can get replacements. Loxone Tree appears nice, but IMHO you'd be mad to implement it because if Loxone stop supporting it or go bust you're stuffed.

With KNX if a light switch or a relay stops working and the manufacturer of said switch or relay no longer exists you can get a KNX replacement from another vendor easily enough.

I'm using a Loxone miniserver in my current house but chances are getting very thin that it will move to the 'real' house. Need to start evaluating alternatives. Was already planning on using more KNX components (presence, buttons, CO2 sensors) so moving away from Loxone shouldn't be that hard.

Yes and no. When integrating with Loxone (or similar) I don't mind programming things like KNX light switches one at a time as it's easy enough, but if you went full KNX installation that would become annoying. Ie. I think that for a fully KNX installation one could no longer get by with the demo EIB license and that would suck because a proper EIB license is more than the 'outrageous' price of the MS gen2 + KNX extension we're complaining about :(

 

I kind of liked the ease of use of their programming interface. But after playing with that for our current home I'm starting to become more and more aware of the limitations of this closed platform. Even really simple situations require very elaborate or creative workarounds. And there is no way to add custom functionality except through something like Loxberry. If I need a SBC anyway then why not move to something like NodeRed completely?

TBH, I hate graphical 'programming' interfaces. For this reason I'm no fan of Loxone Config and NodeRed does not impress me at all either.

If an SBC seems overkill to 'add' to Loxone functionality then there are alternatives (depending on requirement of course). For example, I built some 1-Wire air quality sensors out of an Arduino Mega connected to a 'raw' MQ135 sensor (3-4EUR in parts shipped from Aliexpress). Works like a charm and I love that stuff so low level as an Arduino really can't go wrong once it's running. If you need various 'complicated' things to happen that Loxone can't do you could build them in low level hardware like that. Although if you need a few of these things then there would quickly become a point you might as well use an SBC - doh!

Sorry, but I digress...

If I was forced to replace our Miniserver today I would get a Rasperry Pi with a KNX interface (and to add Modbus, DMX and 1-Wire to a Pi is trivial too) and either code it myself with nodejs or use an open source package like OpenHab or Home Assistant.

Bear in mind that I'm a nerd and professional developer though, probably not your standard user ;)

Cheers,

Rob

[Stefan L]

This is of course bad news but since all my light control is DMX I doubt this can easily be integrated to KNX. KNX to DMX is possible but not sure it is easy... 

DMX gives the option to use stage lighting controllers in parallel via a DMX merger for unrivaled light effects. (not to mention the economy of DMX) 

Most of you here are much more tech but it would appear to me that  Loxone still provide the most flexible solution although they move in the wrong direction and the entry level with KNX just became more expensive. So far I do not have any KNX equipment but I am just getting started. 

On Tuesday, November 26, 2019 at 7:26:48 PM UTC+1, Julián Medrano Silvestre wrote:

[Andras]

On Thursday, November 28, 2019 at 8:10:56 AM UTC+1, Rob_in wrote:

Bear in mind that I'm a nerd and professional developer though, probably not your standard user ;)

And I think this is a critical point here. What is the appeal of the Loxone platform (transforming more into a closed ecosystem right now)?

When I built my smarthome, it was my design choice to keep it simple, have dedicated hardware buttons (or simple automations) for 95% of usecases. For the rest, there's the app that my wife and family understand and like quite much. This appeal in my eyes towards Loxone hasn't disappeared.

Me having a dev background as well, I also said "I can do better!" but quickly realised that I could do better only if I had no dayjob, no kids and a few months of free time :) 

Am I disgruntled by the fact that things got even more closed in the Loxone ecosystem? Yes.

Am I disgruntled that Loxone favours installers (eg: push of Tree bus) over enthusiasts? Yes. 

To me we are still not at a point where these negatives outweigh the positives, but the situation makes me to be even more on the lookout for devices that can be easily integrated in case things turn even worse. If the integration is taken away or made unusable for some reason, that's my point where Loxone is dead to me.

[Peter van Es]

The one reason I chose Loxone (even though I am a s/w developer and hardware guy myself) is because a home-built system detracts value from your house when you try to sell it. Ask yourself: if you were to have a car accident, would your wife still be happy with the system you tinkered together, if something needed changing, or something breaks?

At least with Loxone or KNX and a properly documented installation, your wife, or the purchaser of your house would be able to maintain, repair and change the system, through an independent third party. However, the direction Loxone is taking is not very wise. Staying with standards is the best way to go... or alternatively become a standard.

Peter

[Rydens]

Agree - a system that is maintainable without me is important - but so are open standards and good functionality. Loxone seems to be moving away from home users - which is a shame. 

I installed Loxone when doing house renovation - I went on the course and so far am pleased with it. 

However to extend it to control of things like Somfy blinds / curtains and TP Link plugs  I have added Home Assistant (HASSIO)  - this is free software and runs brilliantly on a raspberry pi. 

There are also interfaces to many other devices in use, and I can send messages to and from Loxone so I get the best of all worlds. It also provides Alexa / Google Home integration.

So now through voice I can trigger Home Assistant which in turn changes the Loxone lights. 

Another device I have started using is the Shelly 2.5 relay - really smart device just like the Loxone Nano but over wifi. These are very flexible with smart software and Alexa / Google Home integration. 

Cheers David

[David Wallis]

I have to agree - I've just spent a fair chunk of money on a load of knx interfaces (2k +) to get me 100's of digital inputs and roller shutter controllers, relay outputs, analog inputs, psu's, rain sensor, brightness, etc etc.. 

When I initially saw this I didnt realise they had the knx extension so envisaged they would then drop support for it.. I lost some sleep over this thinking about what to do.

I also have a bit of dev skills (c#) but can 'get by' with node js etc - however the reason I bought the miniserver is that I didnt want to have to do all the dev - I don't have the time - nor do I have a test environment - and they crash - its just a fact of life - a PLC can.. (edit to say yes they can but the watchdog will get this up and running quicker I'd argue)

I'd just rather someone else did the dev and test - I dont have time to write unit, regression, integration tests for home and then work out how to do releases and keep state without the wife noticing. I just don't have the time - hence why I'm not that active in the previous opensource home automation project I was committing too.

For me the stuff I was coding has / can be replaced easier with node-red and this can contribute to integration with the mini server for non critical stuff..  and this is the route I'll carry on with until such things change..

That said I know we discuss wiring a lot - I will not ever wire for tree - But I will now be also wiring my switches for KNX as well as DI's. 

[armando puente]

That’s correct, today loxone prices are to high with the exclusion of knx conection and the force to pay for and extra device, such as knx extension, i think they will reconsider, if you take the miniserver gen1 price, around 550 euros, and the tree extension, 100 euros it not make sense that today price, gen2 with tree built in, and 600 euros for knx extension, that will almost 1300 euros in total for the same tecnology , so its a nonsense, its like you want to kill your own business.

And for your calm, when the miniserver failed down, pick a z41 pro and forget about Loxone, was  a good device until today, its time to change.

Enviado desde mi iPad

El nov. 30, 2019, a la(s) 11:37 a. m., David Wallis <da...@wallis2000.co.uk> escribió:



I have to agree - I've just spent a fair chunk of money on a load of knx interfaces (2k +) to get me 100's of digital inputs and roller shutter controllers, relay outputs, analog inputs, psu's, rain sensor, brightness, etc etc.. 

When I initially saw this I didnt realise they had the knx extension so envisaged they would then drop support for it.. I lost some sleep over this thinking about what to do.

I also have a bit of dev skills (c#) but can 'get by' with node js etc - however the reason I bought the miniserver is that I didnt want to have to do all the dev - I don't have the time - nor do I have a test environment - and they crash - its just a fact of life - a PLC can but I'd rather someone else did the dev and test - I dont have time to write unit, regression, integration tests for home and then work out how to do releases and keep state without the wife noticing. I just dont have the time - hence why I'm not that active in the previous opensource home automation project I was committing too.

For me the stuff I was coding has / can be replaced easier with node-red and this can contribute to integration with the mini server for non critical stuff..  and this is the route I'll carry on with until such things change..

That said I know we discuss wiring a lot - I will not ever wire for tree - But I will now be also wiring my switches for KNX as well as DI's. 

On Tuesday, November 26, 2019 at 6:26:48 PM UTC, Julián Medrano Silvestre wrote:

So... with the (already known) issue of KNX integration dissapearing from Miniserver Gen2 and the INCREDIBLE price of Loxone KNX Extension....

What are the alternatives for those of us that were using Loxone as "smart brain" for logic and KNX as dummy devices for inputs, outputs and "elegant" room controlers?

I'm now in deep depression....

--

You received this message because you are subscribed to the Google Groups "Loxone English" group.
To unsubscribe from this group and stop receiving emails from it, send an email to loxone-englis...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/loxone-english/e2cbc39e-0524-48e2-85b0-1e0757f13305%40googlegroups.com.

[Vahur Kivaste]

Hi

If gen1 will continue to be sold, would it make sense in some installations to use gen1+gen2 miniservers together instead of an extension? One would get knx, tree, analog in/out, 16 relays etc.

Vahur

[ivusosk ivusosk]

Yes it is possible. That was i told when i speak with loxone over FB. That Why they didnt add SSL for gen1 and they told me that gen1 is 10year old and Tthe reason is performance. For them is was not issue that I nought their system only 2years ago. Not mentioned users who buy Gen1 this year... it’s slap for us...:/

--
You received this message because you are subscribed to the Google Groups "Loxone English" group.
To unsubscribe from this group and stop receiving emails from it, send an email to loxone-englis...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/loxone-english/b384aed8-68cf-4c7c-807c-feffbeebb8f3%40googlegroups.com.

[Rob_in]

On Thursday, 28 November 2019 08:10:56 UTC+1, Rob_in wrote:

TBH, I hate graphical 'programming' interfaces. For this reason I'm no fan of Loxone Config and NodeRed does not impress me at all either.

Thought I should mention that since November I have stumbled across ioBroker. I like this very much as one can program (configure) it using a graphical 'blockly' UI very similar to Loxone or code stuff in raw Javascript (it's based on NodeJS) if you like - choice at last! It is organised using Adapters and Objects which are very similar to Loxone's periphery. I'm using ioBroker as a Zwave gateway and it's great (switched from Home Assistant which I really didn't like).

Raspberry Pi running ioBroker plus even commercial KNX IP interface would be easily under 300EUR (and under 150 if you make your own KNX interface) and can do everything Loxone can plus more (a Raspberry Pi 4 would destroy the Miniserver gen 2 in any benchmark you care to mention).

If you don't like ioBroker, run NodeRed or HomeAssistant or any number of other solutions.

Robin