Why is the Elk M1 so popular?

[Dean Roddey]

In order to justify making such a box, it seems to me, it would have to target the professional market, or it would never have a big enough market to make it worth doing in any kind of business sense probably. And there's already the Elk and Omni kind of sitting in that space now, so that would make it even less inviting for someone who is in it to make money (and everyone has to be in it to make money if they want to be running a real business.)

On the best-of vs. integrated, I would have to agree with the above statements. A set of components does not a system make, necesarily. This is all too often proven in the A/V world, where a set of the best of breed A/V components often make a horrible system as a whole because they just don't get along. I think that clearly Crestron shows that the strategy that appeals to the professional installer is to provide more stuff out of the box. As a software based product, we'll never provide THAT much stuff out of the box, since Crestron makes all kinds of stuff that to me seem way out of their core competency. But, within our software automation world, we think that having as much of it in one box as possible is the best way to insure it all works together smoothly today, and continues to tomorrow.

Devices like the Elk and Omni clearly fairly best case scenarios in this respect. Because of what they are, they are very much designed to cleanly integrate into a larger automation system. So combining one of them with a software product on top seems to me to be a very reasonable compromise, or requirement if you are one of those folks who doesn't feel comfy handing off fundamental stuff to a software-based system.

I don't think though that you'll never get a hardware box that can compete with the broad level of functionalty of a PC based system. By the time you did that, you'd have created a PC basically. The power available in an XP or Linux based box, and the power and flexibility of the programming tools are enormous. It wobbles the mind to think how much it's grown in my lifetime, and will continue to grow. And you have an entire industry that does nothing but enhance this platform as rapidly as possible in as many directions as possible. A company like Crestron is dwarfed by the amount of R&D and manufacturing that goes on constantly in the PC world.

At some point we'll have micro-ITX boards that are as powerful as mini-ITX boards are now, with just a few big chips, and you'll start seeing very small boxes that can serve as automation controllers, inexpensive enough that you can put one in each room more cheaply than you can build a centralized architecture perhaps. With something like the UMPC devices probably down to $250 or some such thing by then, I think that's going to start making it difficult for proprietary hardware based automation companies to keep making their own and competing.

They'll still be able to play the reliability card for sure, and I'm sure it'll still be valid to an arguable degree. But it would a whole lot cheaper to make just a couple of technologies that you can fit into a PC based architecture that to build your own system from the ground up. I think that if someone came up with a super-reliable, proprietary wired/wireless network technology (Zigbee ain't nearly fast enough), that would take care of the biggest single worry of using a PC based automation system, which isn't the controller but the network. TCP/IP just wasn't really designed for the hands-off, automagical network that we really need in homes. Failure of name resolution mechanisms is the biggest single problem we have to deal with I think, in terms of failures in customer systems. I think that it would be quite reasonable to consider a parallel, and much smarter, network just for the key subsystems of the home.

Anyway, I'm way past rambling, but it's an interesting topic that obviously I ruminate a lot about, so it's nice to ruminate out loud sometimes....

 

[upstatemike]

4) With more people using VOIP (which often provides free voicemail) and more people using cell phones in lieu of landlines, the StarGate phone features are perceived as fluff.

Thinking of the Stargate IVB board as just a "voicemail" feature would be a complete misunderstanding of what it does.

It is an audio matrix that can connect line-in, phone-in, line-out, phone-out, and speaker out ports in any combination.

It can record real world audio and store it in user vooice slots. Audio can be voicemail, security audio, or whatever (slots are not fixed length and there are 128 of them.)

It is a full phone control interface offering DTMF input and output for control from any phone in the house, paging through a phone system, etc.

The built-in speach engine can speak the values of system and user variables (which the Stargate has lots of).

Rather than FLUFF I would say any system that does not support these features is substandard at best.

 

[Digger]

I know very little abuot the stargate but from what I read on here I dont know of any other system that comes close. It sounds amazing I agree but is there enough of a market for such a system to be profitable?

Having worked with many Alarm System manufacturers the money is not with the panels but with the accessories (especially wireless). For the monitoring companies the money is with the monitoring and adding the accesories as well (costs $150 for adding a wired motion that costs them $20).

I think that if the capabilities are segregated across several systems many of the features can be accomplished (marry an ELK to CQC, Mainlobby, Powerhome etc). While it may not be as robust and possibly not as stable and flexible it gets the job done for the most part.

If the stargate was going to be available and if I had the money I would buy it I admit. For those that have it I might be even a little jealous. But in the end I have my ELK (and no matter what Spanky says I never slept with it, but maybe the manual once). I have spent a significant amount of time testing the M1 and it is one of the best systems I have ever seen if not the best for what it is (a very rock solid security panel loaded with a decent amount of automation that is supported for the DIY). I think ELK is leading the way in this niche. I bet that more and more profesional installers will use this system. The fact that it has a UL Lisiting for Residential applications and a UL Listing for Commercial Burglar applications coming will make this a very desirable product for installers to sell.

In my opinion ELK got it right.

I had to post this now since I may be working for one of their competitors in a few weeks (if all works out well). But I will refuse to give up my ELK

 

[Mark S.]

Thinking of the Stargate IVB board as just a "voicemail" feature would be a complete misunderstanding of what it does.

It is an audio matrix that can connect line-in, phone-in, line-out, phone-out, and speaker out ports in any combination.

It can record real world audio and store it in user vooice slots. Audio can be voicemail, security audio, or whatever (slots are not fixed length and there are 128 of them.)

It is a full phone control interface offering DTMF input and output for control from any phone in the house, paging through a phone system, etc.

The built-in speach engine can speak the values of system and user variables (which the Stargate has lots of).

Rather than FLUFF I would say any system that does not support these features is substandard at best.

It sounds even more impressive when Upstate describes it - it was a great idea. Too bad JDS is giving up on the IVR board. I just hope mine never breaks.

So what combination of hardware and software could I use to replace my Stargate?

I don't know of any other hardware or software solution that can do what the SG IVR board can do - the closest thing is HomeSeer with HS Phone (along with some very complicated scripting). But I don't know what kind of audio switcher you would need to add to duplicate the SG phone/intercom/audio matrix. Ideas?

For those who advocate a software-based HA system - how would you duplicate the SG IVR board functionality with what's available on the HA market? I can't envision anything except an all-in-one hardware box with built-in firmware/software. But better yet, I'd love to envision such a hardware box that interfaces with an external software solution like HS or CQC.

As for the rest of SG's features:
Support for multi-function keypads
Lighting control
IR transmitter/reciever
thermostats support
security system support
digital inputs
analog inputs
relay outputs
serial ports
RS-485 bus
flags
timers
variables
an easy and foolproof programming interface
and more stuff that I'm probably forgetting.

I know that HAI, Homevision and maybe Elk come the closest to matching all the above. Add HomeSeer and I come even closer. But at the risk of sounding like a HA snob, after having it all for so long with SG, it's difficult to come to grips with the fact that I might have to take a step backwards. I have really come to rely on the I/O of SG - it is critical for my HA setup. Do any other hardware boxes offer multiple (at least 8 each) digital inputs, analog inputs and relays?

I grudgingly acknowledge some validity to the arguments that the market for a new uber-box would be limited. But I find it hard to believe that if RCS/JDS could do it about 15 years ago for $1000, nobody could do it again today, with todays advances in electronics and chips, for the same or even less. Just design it and have it made in China. I know - easier said than done.

Not willing to accept defeat,

Mark

 

[upstatemike]

I grudgingly acknowledge some validity to the arguments that the market for a new uber-box would be limited. But I find it hard to believe that if RCS/JDS could do it about 15 years ago for $1000, nobody could do it again today, with todays advances in electronics and chips, for the same or even less. Just design it and have it made in China. I know - easier said than done.

Not willing to accept defeat,

Mark

If a super mega ultra primo all-in-one box is not practical then a modular one might be the answer. A while back I suggested a roadmap of new modules that ELK could release to bring the current M1 up to the level of a Stargate and beyond and allow it to become the ultimate HA controller of all time. I thought the modular approach and retention of the existing product development investment would be the ideal option. I guess the folks at ELK didn't agree because none of my ideas were implemented.

Right now there are two technical issues I see with the M1 architecture that prevents it from moving forward in any dramatic way:

1- The ROM space for new features is very limited and constrains the system from supporting a lot of new technologies internally. I know it is not cost effective to put a lot of empty ROM space on a system that may not use it right away, but would addition ROM sockets have been too outrageous? If the sockets were there then ROM space could have been expanded for years into the future allowing support for whatever comes along.

2- The bandwidth of the databus is too low. The RS-485 based databus is ideal for attaching expansion boards and keypads but you can't hang too many serial ports off it if those ports are each handling high volumes of data. The system needs a second databus dedicated to HA data that is much faster and isolated from the slower but more critical I/O-security bus. Maybe a 100baseT ethernet type bus with spare pairs used for power and audio?

 

[Steve]

If a super mega ultra primo all-in-one box is not practical then a modular one might be the answer. A while back I suggested a roadmap of new modules that ELK could release to bring the current M1 up to the level of a Stargate and beyond and allow it to become the ultimate HA controller of all time. I thought the modular approach and retention of the existing product development investment would be the ideal option. I guess the folks at ELK didn't agree because none of my ideas were implemented.

Well, good things don't happen overnight, but I have to admit I like the idea and prospect of a system like this and it is not too far off from the M1 concept now. Instead of a central controller with expanders/devices that live on a serial bus, a controller with dedicated 'snap in' modules on a high speed standard or even proprietary bus . So a user could buy the plain controller and get security plus real basic HA control. Want lighting? Plug in the lighting module (like the serial expander), Want IR? Plug in the IR module, Want Voice/TTS? Plug in the Voice module, etc...

This way you can have the inexpensive base unit and add modules as wanted and they would all talk the same language and play nicely together. It's much like the existing M1 but having more modules that fit together better than the 485 bus and easier than kludging together things like magic modules.

This may not be the direction Elk wants to take but I admit I like the concept and if anyone can implement it, Elk can.

 

[BraveSirRobbin]

Instead of a central controller with expanders/devices that live on a serial bus, a controller with dedicated 'snap in' modules on a high speed standard or even proprietary bus . So a user could buy the plain controller and get security plus real basic HA control. Want lighting? Plug in the lighting module (like the serial expander), Want IR? Plug in the IR module, Want Voice/TTS? Plug in the Voice module, etc...

This way you can have the inexpensive base unit and add modules as wanted and they would all talk the same language and play nicely together. It's much like the existing M1 but having more modules that fit together better than the 485 bus and easier than kludging together things like magic modules.

Take a look at THIS concept. It's a start...

 

[toymaster458]

This sounds alot like what HomeLogic does with their PowerBricks. But again we are looking at a Dealer/Installer only product that does not have the customization ability that we want. This is also a Software/Hardware solution made to look like a Hardware Based Solution.

 

[upstatemike]

Take a look at THIS concept. It's a start...

mControl doesn't look like it controls all that much... lighting, HVAC and an interface to MCE (BFD!)

 

[upstatemike]

This sounds alot like what HomeLogic does with their PowerBricks. But again we are looking at a Dealer/Installer only product that does not have the customization ability that we want. This is also a Software/Hardware solution made to look like a Hardware Based Solution.

Exactly. When push comes to shove there aren't that many "bricks" to choose from: Climate, Irrigation, Lighting and Serial. Period. The End.

 

[DavidL]

Yup, no single solution. Sorry Dean!
Just ones that use quality integration with agreed upon connections.
Not unlike what integrators have been doing for years.
Just now with software AND hardware.

 

[Steve]

Yup, no single solution. Sorry Dean!
Just ones that use quality integration with agreed upon connections.
Not unlike what integrators have been doing for years.
Just now with software AND hardware.

If you're talking about Homelogic then Dean is actually right! It is not separate disparate best of breed products glued together but rather products under one roof built from the ground up to work together.

 

[Stinger]

It's good to see this topic getting a bit more back on track of my original purpose of posting it.

This was not meant to be an Elk bashing thread, and I suppose I should have made the title a bit less abraisive, but it attracted the attention that I hoped it would from many of the key players in the industry (thank you Cocoontech for having such a place to do that). I was just fustrated after hearing so many great things about the Elk and getting excited about switching to it, and then finding that it isn't an ideal replacement for my setup.

Brave, I think you're missing the point completely when you suggest multiple systems, I want integration with core functions. I would be an Elk adopter in a flash if it had anywhere near the capability of my aging system! I would imagine that eventually the Elk will become all of what I'm looking for and more, but I am perplexed as to what the delay is. I don't want the core system to handle everything, I think the homeseer and mainlobby combination will always be great for complex tasks and increased flexibility. However, I expect logic, variables and timers in any new hardware solutions.

I believe it's obvious that there is a demand for such a system out there. I will fully agree that stargate owners are clearly in the minority and I'm not sure that the market will bear the cost of R&R and production of such a system. I believe a controller that is sub-$1000 would be able to match the capability of the Stargate and provide much, much, much more and still leave plenty of room for expandability for new technology. What other field can someone say a piece of technology has stayed in the market for over 10 years without major changes and still competes with it's newer competition?!?!?

Where we are right now is the frontier of Home Automation.. New homes are getting built with minor home automation for the normal consumer. In ten years, every house will have some form of automation. What we need is a controller that makes it simple to program functions for the house. It needs to be stable, flexible and EASY to bring on new consumers. I mean who would have imagined that the population as a whole would have adopted MP3s???? I was listening and ripping in 1996 long before they hit mainstream, and here we are where even grandma can listen to an Ipod.

I don't hear any of the vendors talking about a product they are working on to get there. I would love to beta test or participate in a research group, but I'm not clear if one even exists. Hopefully we'll all be saved at some point when a large company decides that HA is indeed a viable market.

DIYers are the pulse of this market, and while many of us take it beyond the point of a typical consumer, it would be wise to note trends for the DIYer early.

I am very impressed by the representation of Elk here on this board, and I appreciate the open dialogue. Mainlobby and QQQ representation has been great as well. I'm very unimpressed by Tink's message, but it's nice to see him at least chime in. I'm not sure what happened over at Homeseer since I found it many years ago, but I believe they lost sight of the reason they were successful - Their community made them successful. All of the folks that put endless hours into scripts and shared them on the message boards. Many of that is long gone, however in their defense, in order to have a stable product, the way they sell the plug-ins now can assure that the software remains stable.

I am not trying to tout the Stargate as a solution for anyone here. Never would I recommend an ancient piece of technology, but at the same time, it's been good to me and obviously stood the test of time quite well. Bring on the next generation.

 

[toymaster458]

If Dean wanted he could get a nice looking box like HomeLogic or MControl uses, Change his software to work on XP Embedded and then it would look more like a Hardware Automation Controller then a Software based Controller on a computer. This is the same thing Homeseer is doing with the Pro-100.

 

[Dean Roddey]

I believe a controller that is sub-$1000 would be able to match the capability of the Stargate and provide much, much, much more and still leave plenty of room for expandability for new technology.

In the fairly low volume business of HA, the manufacturer would probably need to get 200% markup to make it worth it. So, even if they only sold direct, that would mean that a $1000 box would have to cost less than $500 to make, and a sub-$1000 box would have to cost less than $500. If they are going to sell it through anyone else such as CIs, then it would probably have to cost more like $250 or less to makem in order for both of them to get their markup. That could be a pretty tough row to hoe if you want it to be really high function.

If it were going to be a high volume product, then I'm sure you could afford to put lmost the whole thing on a few big custom chips and get the price down pretty low. But you'd have to commit up front to a 'big deal' and do the whole 'shoot star amazing new company with lots of bucks' hype thing to basically blast your way into the market and get the level of sales that would take. That would be a huge investment and risk, that could go down the toilet in a big way.