Idle Speculation - What Does The Future of Loxone Hardware Look Like?

[Gavin Harper]

Hi All,

I am in the process of resuming the planning for a stalled installation at our guest house. I'd started down the road thinking about a KNX system with GIRA Homeserver, and thought that to begin with would focus on all of the basic things like switchgear that wouldn't go out of date, and then would look at the "front end" when the installation was complete. This was four years ago, life intervened, and I'm now looking at the project again.

Loxone seems like it has come on in leaps and bounds and is now an obvious choice for the front end...

However, there are still other things to do, and there isn't a mad rush, so whilst I'm planning around Loxone, I'm also holding off to the last minute in case there are any changes, updates, revisions or improvements....

And I thought this was an interesting catalyst for conversation... what would the future of Loxone hardware look like? Are they just going to stick with the Miniserver as is, or at some point, one would presume that they would expand the hardware capabilities, I don't know, faster processor, memory e.t.c.?

Has anyone ever run into any limitations around Loxone? At present, I'm considering whether I can run the whole installation on one Miniserver, or whether to have two, one as a "Front End" for the public features, and another for the "Back End", energy metering, heating control, alarms e.t.c.

I welcome anyone's thoughts... should I just get on with it and buy a Miniserver, or does anyone think there might be something around the corner?

[Duncan]

i cant answer about upcoming hardware, but even with a full memory configuration there isnt any noticable slowing of responsiveness unless you do something complex in real time - my sound to light configuration using an analog input driving dmx rgb leds can slow the whole system down to the point where there is a delay in button function

the current loxone config can run out of memory for a moderately large install if you are doing lighting, heating, blinds, security etc etc so depending on your planned install, 2 miniservers might make more sense from the outset

its impossible to totally future-proof, however using a structured wiring pattern with multiple interfaces in mind (i.e. star wiring and/or distributed wiring centres for lighting circuits, some switched sockets, heating controls etc with control/signal wiring arranged so you can you swap from loxone to any combination of knx/network/1-wire/dmx) will enable to you swap to alternative hardware in the future should it be necessary

i would suggest you plan to wire everything and reserve wireless for unplanned additions if have the option.

[Gavin Harper]

That's good to know. The sound to light sounds fascinating. Have you got a link to the project?

Yes, this will be everything, heating, with lots of 0-10v actuators being controlled and temperature sensors in all the rooms e.t.c. lights, window motors e.t.c.

I agree with your approach, I am going for wired by default as we have access to all areas at the moment.

In terms of running out of memory on large installations; is there any method to write data to an external server or data dump? I know people have written about the durability of memory cards with lots of write operations, and the need for good quality cards... I wonder whether it would be possible to dump data to a Synology server or something and free up the Miniserver resources? 

Thanks for your reply, some really useful stuff in there. Appreciate your time.

[Duncan]

the memory limit is not the physical card (although that may play a part), its the ram available to hold the number of objects in your loxone config - there seems to be a limit around 7500-7800 items = 100% full.

there are lots of tricks to bring this down such as clever programming, but its useful to have some reserve as there are some things that can take up a lot of logic blocks to create where its not an exisiting function.

one oddity is that if you can replace say 10 function blocks with a single formula block or state block it reduces the memory footprint in proportion to the reduction in function blocks.

[Rob_in]

On Thursday, 1 June 2017 12:32:00 UTC+2, Gavin Harper wrote:

And I thought this was an interesting catalyst for conversation... what would the future of Loxone hardware look like?

From musings on this forum one can glean two things:

- Loxone seem to be contracting end user support in favour of professional installers.

- Loxone is expanding it's own hardware line (ie. tree) and pushing this more and more.

It depends how draconian they want to get, but I think the availability and ubiquity of KNX devices makes them an easy choice for end users over things like the Loxone tree switches and detectors. If Loxone really want to sell more tree gear they either have to drop the price or remove the KNX interface to force everyone who wants a bus to use tree.

I find this scary as imagine later models could remove the KNX interface. Their logic will likely be that they don't need KNX as Loxone are now able to supply all the functionality you need with tree.

Good or bad? Single supplier is good for close integration, but bad for choice and catastrophic if they drop support or simply disappear.

TBH, I prefer an open standard like KNX that is very mature and supported by multiple manufacturers. The risk of not being able to replace dead KNX switches or actuators (not that I have any of those - using KMTronic stuff) in the future is pretty low. As with most people here, I like the Miniserver as a compliment to the KNX standard so think it would be crazy for Loxone to drop that, but stranger things have happened in the pursuit of more sales!

Robin

[seb303]

From a business point of view, it's understandable the direction that Loxone is moving it.  On the overall scale of home automation hardware, their kit is reasonably priced, and I very much doubt it would be sustainable for them to be directly supporting 1000s of DIYers who've just bought a Miniserver or so.

That said, I can't see them making any software changes that are deliberately going to break backwards compatibility for existing users.  Sure, they might choose not to put a lot of development resources into things like additional KNX features, but if it works currently I would be confident that it will continue to do so.

[tkn]

Let me entirely disagree with you there.

The reason home automation will always be niche is nobody wants their home to have a brand name on it. Competing standards, lock-in, etc. Especially in something fast moving.

I don't want an Apple home. Nor a Google home. Or, honestly, a Loxone home. I want home automation that works. And company greed is what is preventing me from having it.

I am deliberately avoiding buying anything Loxone except for the bare minimum. My wiring will probably be able to fall back onto KNX. And if they remove the external interfaces, I will probably find a bunch of other dissatisfied customers and start working on using a regular commercial PLC rather than face lock-in.

The reasons why are numerous, but include:

1. Risk of company failure or buy-out

2. Rapid pace of technology and concerns about obsolescence

3. No company can be the best at everything. They don't have a complete product line and never will. And their products will never match a specialist in that area. There is zero chance they can have the best LED lights, the best HVAC controls, the best shade controls; not to mention the lack of awning/canopy control and other such advances.

Loxone is interesting because it is PLC based and is interfaceable with everything. As they lock it down, they will lose that value proposition and become just another me-too player and nothing very special at that. Control 4 and Savant and Crestron and the rest have that market and there is almost no need for another entrant.

[seb303]

As far as I know, Loxone are not 'locking down' anything.  Sure they are pushing solutions that will be commercially viable for them, but doesn't stop anyone using their basic system and software in other ways, interfacing with whatever.  The nature of a 'connected' home means there will be different components to link together.  From my experience, the Loxone system seems open enough to manage this fairly easily in most cases.

Customers who are "deliberately avoiding buying anything Loxone except for the bare minimum" are obviously not going to create a sustainable business for them alone, so one can't blame Loxone for focusing on attracting other types of customers too :)

[tkn]

Well. They could make their products not locked in. They are the ones that chose to go with some BS proprietary wireless protocol instead of Zigbee or Z-wave.

Loxone Air is the perfect example of vendor lock in. I won't do it. I have no problem buying their solutions where they make sense. But to lock myself into a proprietary solution like that seems really short sighted.

And then there are the massive up charges on their equipment. I can handle a 100% markup on the amps. That seems petty normal. But over 300% for the music server which might be violating GPL? Hmm. Maybe not. They've also discontinue some of their multichannel input devices for ones with fewer inputs. Maybe to drive more sales? No idea.

Anyway they are doing a lot of things right as well. They will definitely be getting a large chunk of money from me. But there are definitely worrying trends going on as well.

[seb303]

I take your point about their wireless products being proprietary.  But these are an option on top of everything else.  Personally I would avoid wireless wherever possible, regardless of the protocol.  If you want to use standard wireless kit, there are Z-Wave to Ethernet bridges for example.

The base Loxone system remains highly connectible in standard ways (Ethernet, KNX, DMX, RS232, Modbus, etc.)

And yes, their hardware is expensive for what it is, but you have to consider this alongside their whole ecosystem of software, support, etc.  DIY hackers aside, the Loxone hardware is still a relatively small part of the overall install cost of an automated home.  And of course as a DIY hacker, one has many options such as the Elexol IO72 or Logitech Media Server to substitute for the more expensive Loxone components.

Personally I find it reassuring to see that Loxone are finding a business model that is going to be more profitable than selling the odd Miniserver to DIYers.  The worst situation would be that they didn't make any money and the business shut altogether.  I see no likelihood that they will drop any of the standard protocols that they already offer, so I think it's a reasonably future proof solution.  As Duncan pointed out earlier in this thread, it's impossible to future proof a home automation system completely, but making sensible choices when wiring it goes a long way.

[Duncan]

when you look back at the home automation origins, you have crestron, lutron, control4 etc - nearly all have been based on completely proprietry hardware (including extremely expensive switches) and you are locked into an approved installer at extremely high prices - absolutely no chance of diy at all.

the original home automation scene started with lights, heating, ventilation and audio, particularly for a larger or open plan building where lighting moods are more appropiate than single circuits - a role which loxones fills as well as most, but gives you the option of non-proprietry input and output devices if you want to avoid loxones own, but they are still available if you want a single supplier ecosystem. in addition, for those who want to, you can evolve your automation with no extra cost other than hardware.

i think loxone is gradually moving to that model - single supplier, approved installers etc, but we represent unusually demanding customers compared to most where
i guess most people want a service, have no idea what home automation really is or what functions they really want, and have no interest in fiddling once its up and running
the more loxone sell their hardware via this model, the longer the company stays in business, indirectly supporting us hackers and its diverse range of open interfaces.

google, amazon etc recent 'smart' products are far less integrated home automation, rather an attempt to tie individual 'smart' products together instead of using a number of different apps or interfaces. personally i wouldnt call a mishmash of a few speakers, a next thermostat, a couple of smoke detector and a few switchable sockets home automation. personally i want my automation to run largely unseen in the background, with wall switches for lights and remotes for AV, not reliant on any internet/cloud service and be operable by my children and parents/friends/visitors - there may be a place for voice-activated features but i dont feel confident on it being the main way of operating things any more than relying on wireless comms.

[tkn]

Fundamentally I think of it as top-down and bottom-up type systems.

Realistically people don't want to fuss over their systems, and, often, don't know what they are capable of. So, if they have money, they go Control 4, et al; and if they don't, they pick up some Hue bulbs and maybe a Nest.

Google and Amazon recognize that nobody wants that mish-mash of interfaces, but are so wedded to their ideas of lock-in and compatibility, that it is hard for them to unify the interface. Apple offers their own solution that is mediocre as well for other reasons (lock-in again, hardware requirements, low attention span to interface, etc).

So Loxone has this unique niche of being powerful enough to do everything, but a bit tweaky. So I have no problem with them rolling out solutions that are plug and play to fill in the technical gaps. But not at the expense of the ability to interface with everything. Realistically if they are committed to this niche (which I could see becoming much more dominant over time).

What I'd like to see?

1. No proprietary standards. They claim Loxone Air is somehow better than Z-wave and Zigbee. Perhaps it is technically, perhaps not. But practically it is an inferior solution because no other products use it.  It is the kind of lock-in nobody wants. I agree that wireless solutions are suboptimal, but sometimes they are the only solutions and there are convenience aspects.

2. Closing the forums was IMO an absolutely huge mistake - it puts people's experience of your product out of your control entirely.

3. Finally some sort of third party database that is unofficial, but where the company lists solutions that have worked and the products used. I have spent far too long trying to figure out solutions for things that have been solved but I need to dig in and figure it out on my own. The forums could have been the perfect avenue for this - just have a subforum for unofficial third party solutions.

For example, my insurance company most likely will require a monitored security solution. So the Loxone security options are not going to work for me without some way to interface with a service. I emailed them to ask about options here and got a very generic "some of our customers have found solutions." Yeah, thanks for nothing.

-Tarun