Where've you been?
I and others would be interested to hear your evaluation of Insteon. I understand you have converted 100% from X-10.
Been busy on other projects. Also had a classic worst-case disaster a while back - simultaneous hard-drive failures in my laptop and primary desktop computers, and since they were backing each other up... Lost a bunch of email; managed to recover most of the other critical stuff. Needless to say, my backup strategy is a little different now :)
Anyway, about Insteon:
My house was built in the early fifties. And the sixties. And the seventies. And the eighties. The house is now about 4000 square feet, about 90 feet end-to-end, and has an "interesting" mix of wiring styles. My HA computer is at one end of the house, a long distance from most of the stuff it controls. X-10 was always marginal, even with an array of phase couplers, boosters, filters, all the usual stuff. I've been completely Insteon for about a year now, and power-line control has been absolutely reliable. I was using my own HomeSeer plug-in until the disk problems; when I rebuilt the system I installed the "official" HomeSeer Insteon plugin.
My biggest reservation about Insteon right now is the physical reliability of the devices. Out of thirty or so SwitchLincs, I've got two SwitchLinc Relays with bad rocker contacts - one won't turn on; the other won't turn off. Both work fine under power line control. Even more alarming is that I've had three or four LampLincs die when bulbs they were controlling burned out.
All of my Insteon stuff is early production; I'm about to get all the failed devices replaced under warranty - I'll see if anything's improved, quality-wise.
- Dennis Brothers
From the Smarthome site: The Panasonic Network Camera provides high-quality color video that can be viewed from any Internet-connected PC in the world,
So the answer to your question would seem to be "No".
Seems to me that the 'natural place' would be for the IP camera and any proprietary SW to reside in its entirety in the camera at its IP address and not have any interaction with any other OS or gizmo until called upon, and then only to serve out images as directed -- all without any drivers on any other machine. (Seems to me that we've had at least part of this conversation before.)
And thus shouldn't be used. I can blow up CQ by hitting the hard drive with a hammer, too.
OKAY. So there is no PC-based home automation software that meets your desiderata. Seems that you will go without for theoretical reasons and for what others would consider unreasonable expectations.
So go without.
Others with a different set of expectations will do jist fine. I used Savoy's CyberHouse beginning in 1999 with Win98 for years with none of the problems that you seem to _want_ to induce.
What does this mean? I just bought a used rack-mounted 1.6 ghz , Intel-made mother board machine on ebay for $150 which will become the new HA server.
You've spent more than $350 just on X-10 signal enhancements some of which are now discarded. So price matters to you a lot on some matters, and less on others.
Wasn't it you who was saying not too long ago that low-powered mini-itx PC's that could be squirreled away most anywhere were the Future Of Home Automation? Were you also gonna have Sis do her homework on it? Since cpm, one CPU per user has been mantra. This is nothing new or demanding or unlikely. In my house, there are three wintel machine per person (1-desk, 1- laptop and 1- infrastructure. The laptop and desk overlap because in part of restrictions on use of work computers for private purposes.
Hippo talk. Nobody is suggesting one CPU for *all* applications. I/we are saying that it is best to dedicate a computer to HA. Consider re-reading your own previous posts about mini-itx. A pentium computer capable of running Homeseer, CyberHouse or CQS can be purchased from Dell that fits inside of a standard wall-switch junction box and runs off POE. So why the hyperbole about 'server farms'?
So you are stuck without, while others are doing jist fine with? It occurs to me to ask what experience you have with _actual_ _HA_ software running
24x7x365 on a PC. Are you fussing about doing in the future what others have done successfully for years? [snip]
In my real world (which gave up on tapes long ago) I find that I can have daily, weekly, periodic and long-term backups on different machines in my network, on drives physically located in the safety deposit box at the bank and backup files stored on a different server elsewhere in the world in cooperative fashion. By now, many folks have friends and family members in different locations connected by the net with 24*7 connections. This is not difficult or a jungle. Reliable and practical and amenable to unattended automation. Only "backup" software one needs is xcopy and the windows scheduler. The cost of removable hard drives used is trivial compared to the value of the data stored in my case. I jist do it -- instead of fussing about it. It's not 100% because nothing is.
... Marc Marc_F_Hult
I recently lost an old laptop. It never recovered from a defrag. Fortunately, I could boot DOS and see the HDD so was able to salvage the files.
I have a network HDD that I try to use for backing up the things I really don't want to lose but the actual backing up is hit or miss. But the most critical stuff is distributed among 3 machines.
Could it be the Lamplincs went bad and took the bulbs out? I think that's more likely. Someone else reported 2 similar incidents on the Insteon forum and I suggested he make sure the latest failed module gets to SH engineering and they've taken steps to do that.
Your experience is sorta what I expected when I first looked at Insteon and reviewed it here. The basic technology looks good - Smarthome still looks like Smarthome. :(
I'm glad I only paid $100 for the SDK. ;)
Dennis' experiences are similar to mine. One of the V2 dimmers that SmartHome sent to replace an original version that could be coaxed into flickering had a bad SPST switch. This is ironic because the hardware itself looks to exceptionally well designed and built -- gold-flashed pcb traces, labeled and heat-shrunk leads, clean, residue free soldering, and seemingly high-quality components.
Our setup depends on locally linked 2-way and three-way switches. They work wunnerfully and are trivial to set up. Very occasionally, when I'm at the top of the stairs and simultaneously hit both the local switch for the light on the 2nd floor and the linked switch for the light on the first floor, the linked switch won't turn on and I have to hit it again. This doesn't happen if there is even a slight delay between hitting switches or if I _dim_ up or down. Also the INSTEON ON-OFF (relay) wall switch that controls the fluorescent T4's in the basement has a spit second delay when pressed even though it directly controls the load (not linked).
These are very minor quibbles compared to any and all X-10 tactile/feedback experiences I've had over the years. One could leave these INSTEON switches in place (eg when selling a home) without anyone knowing that they were automated. They look and work for all the world like Lutron Decora manual dimmers that cost just as much as INSTEON ICON dimmers. This by itself is a substantial virtue in my negative, previous experience with selling an X-10- equipped home. In other words, these work great as "manual switches that can be automated". I don't think that anyone could ever say that with a straight face about X10's X-10.
... Marc Marc_F_Hult
I'm not claiming immunity from bad drivers. I'm claiming that bad drivers are NOWHERE as common as you seem to think, and that within a couple months of the release of a driver for a highly used piece of hardware (the only kind you should be using) many thousands of people will have used it in many different configurations, and that's a pretty good sign that it will be safe for you to use. The best thing to do when you need high reliability is to 'buy behind the curve'. Buy hardware that has been out long enough to be heavily vetted.
We'll, they do have to worry about it, but they also get to charge for it. If the customer breaks the system, and they call the installer to fix it, and it turns out to be the customer's fault, the customer pays. OTOH, if you are doing it yourself, and you plug in the camera, and your system stops working, you can sit down and draw some diagrams and within a few hours of hard thought, you'll probably figure out that the camera is the problem and remove it until you figure out what the problem is.
No, I'm not calling anything junk. But, if it breaks your entire system when you plug it in, then it obviously has problems. Look, if these products broke systems all the time when they were plugged in, it would be pretty well known and a trivial amount of research on the web would tell you that. It's your responsibility to do the research if you choose ot to pay a professional to do it for you.
In the 'real world', our product runs on people's machines without problems despite the horrors you seem to be seeing all around you. So, though your concerns are theoretically true, as a practical matter, it works a lot better than you seem to think. We just don't have users having these problems.
If it breaks the whole thing, then it would be so disfunctional a product that you should send it back. Whether it is supported or not depends. If it can use a web client, then you can just embed a web browser widget in your CQC interface to view the camera. This is the most common way. We will be looking for some to support more directly in the future.
If you are, then you pay the price for that. What else would you expect? If you are going to use untested stuff, then what is your complaint here? Yeh, it might break something. This is not a suprise. But the bulk of people are NOT going to do that. They will use well known, well vetted hardware because they are interested in stability.
I'm not sure where you are getting this from. Vista isn't going to be that big a deal for us. What exactly in Vista do you think is going to be so difficult to support that we will end up being broken on the rocks and go under? Please be specific. Are you a software engineer? If so, I assume you have some specific concerns as to what could be that horrible. Personally, I figure it will take a few days to make any required updates for Vista. I think we can handle that.
And there isn't even that huge a need to move to Vista anyway. As alreayd mentioned, mostly our system is running on servers in a closet and kiosk style touch screens. The OS is not exposed and therefore the OS version is not important. And our customers (being conservative about reliability for their automation system) will not be rushing to to Vista anyway. They will likely give it some break in time, as I mentioned above as the most obvious means to insure reliability. So we have plenty of time to move to Vista.
--------------------- Dean Roddey Chairman/CTO, Charmed Quark Systems, Ltd
Hard to imagine how a defective LampLinc could damage an ordinary 100W incandescent bulb. On the other hand, it's easy to see how a current surge as the bulb dies (typically accompanied by a blue flash) could damage a triac.
- Dennis Brothers
Why would there be a surge when the (resistive) circuit opens? The inrush current when a cold incandescent is turned on is significantly higher than after the filament heats up. If the filament is thin in a spot, the normal inrush current causes it to break.
Dave,
When a tungsten filament fails, a "tungsten arc" can occur through vaporized tungsten connecting the ends of the foreshortened filament. There is a positive feedback in this effect, so as more tungsten is consumed to support the arc, the current path through the remaining filament becomes shorter, and the current increases causes more vaporization. This can result in the bright flash that is sometimes seen when a filaments fails.
This is a well-known phenomenon. Some lamps have fuses built into the base to provide some measure of protection. One of the functions of the RC snubber circuit across a TRIAC is to protect the TRIAC gates from inductive spike during filament failure that can damage the TRIAC. This is also a problem with IGBT and MOSFETs used in reverse phase dimmers.
... Marc Marc_F_Hult
Marc,
So the net effect is indeed a current surge while the light bulb filament is breaking apart?
(x-posted to sci.engr.lighting,alt.engineering.electrical,sci.physics)
Thanks,
The "tungsten arc" is indeed well-known and most household light bulbs have the built-in fuse to prevent it from blowing mains fuses or tripping circuit breakers (or even destroying triacs). How many people have had a circuit breaker pop when a bulb blows?
I've recently had a spate of bulb failures that I attribute to ultra-cheap imported bulbs. They blew "not with a bang but a whimper" (actually, more like a "poof") and they did not harm the lamp modules (LM14A & LampLinc V2). Most (all but one) of my lamp modules are plugged into power strips with 15A breakers since I cannot get to the basement to replace fuses - none of the breakers have tripped.
Given that the folks behind Insteon have been designing triac dimmers for several years, why would they have not taken this into account?
You don't do much gaming, do you? You can get sometimes find a driver per week at some of the high end video card support sites. As more feedback and problem reports arrive, the fixes soon follow. The same is true for BIOS's in any new motherboard. You can see them issued once a month, if not once a week, after a new board is released. Some of these fixes are not minor cosmetic polish, either.
It may well be that in your corner of the world you don't see driver issues but I see them continuously. As Dennis B. pointed out, the arrival of Vista is going to seriously alter the driver landscape. Let's just chalk this disagreement up to our differing experience regarding drivers. Let's, for the moment, forget that the author of NOOKS reported drivers were at fault it 85% of all system crashes and that 5% of all Windows machines have to be rebooted daily.
Are you sure it's the camera and not the system? Perhaps there was a flaw in the OS's video driver that never came into play until Camera X came on line. I suppose that's where you and I differ the most. My experience in troubleshooting PC's has been that they are a LOT of trouble, and the more SW and HW that's installed, the more troublesome they become. I want to know the process by which you decided what's junk and what's not and what will break your system and what won't. I found that a slightly loose mouse plug on a ASUS PIII Coppermine motherboard stopped the entire system from booting. No cogent error messages. It just stopped after writing some boot up information to the screen and hung there.
I'm not going to bother researching the numerous tales of people who have done nothing more sinister than download the MS patch of the month and who have then had scanners, printers, cameras, sounds cards, applications and more just stop working. They didn't choose or load "bad" equipment. They were caught in the "you have to upgrade bind where the upgrade can both giveth and taketh away.
Anyone working with more than a home installation of Windows knows how easy it is for a Windows system to grind to a halt for any number of reasons. We both know, or at least I hope we do - me from bitter experience and you from writing drivers both to spec and to actual hardware in hand - you can research something to bloody death but the rubber hits the road when you install it. With the outrageous variety of HW and SW available on the PC market, it's entirely likely a serious error in a driver will go totally undetected until it runs into just the right hardware and software combination. Truthfully, can you say that your research results have always corresponded to the real world behavior of a piece of HW?
That's what I wanted to know. Thanks. Does it seem so unusual that I would see an IP camera from a respected player in the business like Panasonic at a price point that I find agreeable and want to fold it into my HA setup?
While turning on lights to simulate occupants while away leads people into X-10 and light automation, the desire to install a nanny cam or to see the home while on travel on the cellphone is another very popular "entry point" for people into home automation. I would expect a general use HA SW product to be able to accommodate them both generically and specifically.
Untested by whom? I have a little problem here trying to visualize the process by which Joe Average is able to test and evaluate a driver, or to even figure out who to trust to test the equipment. This is what I call a "magic dingus" in that it calls for process that's not visible. In your case, granted, you have skilled installers and maintainers making these decisions about drivers and what is "good gear" and what is "bad gear." To the average end user, if it says PC compatible, they believe it should run. I doubt if one percent of the population is able to scan a driver with HexEdit, look to see if the author filled out an informative property table, look for misspellings or in any other way evaluate the quality of a driver supplied with something like an IP camera or a printer.
You live in a far, far different world than most of the people I know and work with that use PC's. The truly experienced know that every piece of HW or SW they add is akin to playing Russian Roulette in that you never know when the whole thing will just stop.
As for "bulk people" I would have to counter your observation and say that the "bulk" of the people I know just install things. They don't backup beforehand, they don't use an install log, they don't examine the processes list before and after to see what's changed and they neither back up the registry nor do a registry concordance to see what a new program has done to their system. Those are among the many things it takes to approach 24/7/365 reliability. Very, very few average users do them.
A few days? I think you're quite the optimist, Dean. We'll have to wait and see. I've worked on a number of software projects for various clients. When you have to support another OS, particularly a new one, it can double your testing load, it can increase the size and complexity of your distribution media, it can require multiple sets of help files tailored to each OS. Housekeeping and version control can just eat you alive if you run into significant issues porting.
Has CQ ever had to make the trip to a brand new OS? It's an interesting journey. My organization created several widely used in-house applications over the years for thousands of users. The programs ranged from collecting survey data to decision modeling software. When we had to bridge from DOS to WIN3 to Win98 to NT to W2K we had to spend a fortune. We had to set up a very complex test lab so that we would have each different OS available to bench test with a variety of configurations and network topologies. New documentation was required, much more complex and "environmentally aware" installation programs were needed, issues with video drivers cropped up and were quite a plague until MS and ATI worked them out after months of mutual fingerpointing, etc. and various bug reports and feedback from the field kept us working double shifts.
Did I miss it (my delete key was toggled for a while)? Did you explain how your business plan would cover a prolonged illness of you, the key man? That little issue alone has been enough to sink projects far larger than yours. It's rumored that Wordstar died because the key man left. It's pretty obvious at ADI that when Dan Boone left, development on ADI and Ocelot gear stopped. If you get overwhelmed by a Vista port being harder than you expect, who takes up the slack and covers for you? If I were to evaluate your attitude now as compared to a year ago, you seem far more tense than you did before. That probably means the workload is creeping up on you more than you realize. (-: Chill dude, chill.
-- Bobby G.
Please fix your sig line delimiter--- ;-)
Anyway, I have never experienced what you describe (a circuit breaker trip) on failure of a bulb. My common sense tells me the arc (air gap resistor) would be a great, sudden, resistance that would indeed draw a large current spike. Part of me thinks the opposite is true - for which I have no basis!
I was not aware that common bulbs had a fuse either, I thought the filament IS the fuse.
My interest in this has nothing to do with Insteon devices, it has more to do with what I can do to protect the circuit for devices I install, if the theory is true. It may explain some anomalies I've experienced with security devices.
And that's getting a lot worse with the "applefication" of "Father-Knows-Best" patches. I cringe everytime I see the "Click here to install the latest patches" message like I saw this week on my XP machine. Had this morning's 1-1/2 hr power failure (courtesy of Duke Energy who seems to think that frequent glitches will help the regulators approve their requested rate hike) more closely followed the patch, I might have just burned the XP machine. (I did need some way to cook breakfast and make coffee.)
No, I don't. And we are talking about automation, not gaming, so I'm not sure what the relevance is.
The differences he pointed out was NON-AVAILABILITY of drivers for old hardware, nothing to do with quality. I have no idea what the author of NOOKs knows or how valid any of his information is. I don't particularly doubt it, but it's not particularly relevant to the discussion at hand that I can see, because 85% of the problems that our customers have would be a small absolute number.
Irrelevant, for the most part. The OS is there. It's what you are using. If the camera doesn't work with it, then it doesn't work with it. Your options are work with the relevant parties to figure out why or move on.
I'm sorry, but I don't see these problems. I have 4 systems in my home right now, recently replaced one that I used every day for hard core development and it never game me any problems except occasionally it would fail to shut down after upgrades. The others have been working just fine. I've had automaton controllers that run without problems for a year or more (and would have continued had I not replaced them.)
This is because I either buy a pre-fab configuration (Dell's mostly) or I research what other people have been using and use hardware that I know other people are using without problems.
The average person will probably not do this, any more than the average person will fix their own car.
I was refering to the bulk of people implementing an automation system, not of the public at large, since I thought that was the issue at hand.
I've done more cross platform software development than most people on the planet, I'm sure. I've already explained how our product is architected to minimize platform impact. You can believe it or not.
No CQC the product, but the underlying object infrastructure has. I'm well aware of the issues, and have long since put the mechanisms in place to deal with it. The underlying object platform used to run on XP and Linux simultaneously back in the day, and because of the very mechanisms I've pointed out it required no changes in the code, just the use of the virtual kernel layer being implemented on each platform.
We've made it clear that if, in the very unlikely chance that would happen before we are in a position to deal with it, that we would open source the product. So no one would be screwed, except me if I died.
I'm not tense at all actually, at least not about the software issues, which are primarily in your imagination. We are in a better position than ever and interest in our product is ramping up at an ever increasing rate, because it's a quality product and doesn't suffer from any of the problems you seem obsessed with. I just find that your arguments seem primarily oriented towards spreading FUD, without any actual factual basis, and frustratingly incomphensible.
--------------------- Dean Roddey Chairman/CTO, Charmed Quark Systems, Ltd
The tungsten-arc is for real. The built-in fuse is also. If you hear a "pop" and see a bright flash, you're buying cheap bulbs. If you hear a "poof" and notice a bit of a flicker before darkness descends, your bulbs are fused (even if cheap).
The inrush current for an incandescent is about 10 times the current once the filament has warmed to its normal temperature. For a 100W bulb this means an inrush at turn-on of +8A. The "fuse", if any, built in to the wire leading from base to the filament should withstand the inrush but blow before a 15A breaker is tripped by a tungsten-arc.
Come on... This is all like discussing how many angels can dance on the head of a pin...
In reality, don't most folks buy light bulbs at Home Depot, Lowes, Target, Walmart, (or at their local food store if they ran short)!
I doubt that 99.9999% of the folks would buy "fused bulbs" vs "unfused bulbs" but will buy what's > G. Morgan wrote:
The fundamental question for me is whether or not light bulbs can cause equipment damage. And no, I have no life.
Asked by someone kept by a cat. ;-)
It really depends on the equipment. Anything that uses incandescent lamps "should" deal with this without failing.
Cabling-Design.com Forums website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.