Page 3 of 3

Posted: Mon Jan 16, 2006 9:14 pm
by ScottBot
The ramp rate is so fast it's hard to call it a 'ramp rate'.



My only concern now is that there is some serious latency when sending the commands. For example, if I want to turn on three lights, it takes about a second to turn on each light! Looks like X10. It's very possible that there is just something I'm doing wrong in the plugin. I've posted a question on the Insteon forums, but haven't heard back. It could also have to do with the fact that I don't have any of the RF repeaters (although I thought they were just to extend the range and bridge phases).



If you decide to go that route you need the USB PowerLinc V2. Smarthome as several different 'starter kits' that you may want to check out.

Posted: Mon Jan 16, 2006 10:49 pm
by Steve Horn
Is the latency problem you're encountering due to the ICON device(s) or is it with "genuine Insteon" too? If its just an ICON issue, couple that with the (non)ramp time characteristic maybe the extra $20 is worth it after all, at least if dimming is required. Did you try the same thing (switching multiple lamps in succession) with the PowerLinc controller software (Smarthome PC timer software), i.e. bypassing HB and the Insteon plug-in?

Posted: Tue Jan 17, 2006 8:53 am
by ScottBot
Steve Horn wrote:Is the latency problem you're encountering due to the ICON device(s) or is it with "genuine Insteon" too?
No. I've tried it with a standard Insteon plugin module too.
Did you try the same thing (switching multiple lamps in succession) with the PowerLinc controller software (Smarthome PC timer software), i.e. bypassing HB and the Insteon plug-in?
I've tried it with the development software application that sends the commands directly to the powerlinc, and it has the same delay.



My theory is that it's not transmitting via RF at all. Only powerline. The delays look just like X10. Although I've not read this anywhere (yet), I'm guessing that the Powerlinc V2 is strictly powerline and to get anything to respond to RF, you need to also install the powerline-RF modules. However, I'm not optimistic that this will speed things up, since it still has to cross the powerline to get to the RF transceivers.

Posted: Tue Jan 17, 2006 10:32 am
by pavlov70
ScottBot wrote: My only concern now is that there is some serious latency when sending the commands. For example, if I want to turn on three lights, it takes about a second to turn on each light! Looks like X10. It's very possible that there is just something I'm doing wrong in the plugin.


I haven't had as much time to play with your plugin as I'd have liked so I don't really remember how bad the latency is. Over at accessha.com there's a guy who's writing his own Insteon controller software. I've had a bit more time to play with his and light response seems to be pretty quick. Nowhere near as bad as X10.



Is your computer the PowerLinc hooked up to plugged into a filter as per the PowerLinc instructions? Perhaps noise is slowing things down.



Rob

Posted: Tue Jan 17, 2006 10:41 am
by pavlov70
My theory is that it's not transmitting via RF at all. Only powerline. The delays look just like X10. Although I've not read this anywhere (yet), I'm guessing that the Powerlinc V2 is strictly powerline and to get anything to respond to RF, you need to also install the powerline-RF modules. However, I'm not optimistic that this will speed things up, since it still has to cross the powerline to get to the RF transceivers.


AFAIK there's really no RF Insteon devices yet other than the SignaLincs. The Power/Lamp/SwitchLincs are all powerline only. The SignaLincs are for bridging phases and for future use like RF remotes, smoke detectors, etc.



Also, check out the AccessHA Insteon forumhere. It's one of the more active Insteon forums around (outside of the SH Insteon developer's board).

Posted: Tue Jan 17, 2006 10:46 am
by ScottBot
Thanks for the info. I was arriving at the conclusion that the switches and PowerLinc weren't RF (very misleading if you read the Insteon advertisements). That's a mark against it when comparing it to ZWave, unless they have significantly increased the powerline transmission speed and reliability.

Posted: Tue Jan 17, 2006 12:14 pm
by wetwired
Insteon isn´t RF? And here I´ve been feeling bad about investing in Zwave, Insteon was supposed to be so much better...

Posted: Tue Jan 17, 2006 1:26 pm
by pavlov70
wetwired wrote:Insteon isn´t RF? And here I´ve been feeling bad about investing in Zwave, Insteon was supposed to be so much better...


Insteon IS RF. AND PLC. Sort of. Kind of. Not really. I agree some of their advertising is a bit misleading. They make it sound like each device is xmitting RF and PLC when in fact only the SignaLincs are. Extreme powerline noise on a circuit can still break Insteon and no amount of SignaLincs will fix it since the switches aren't RF. I've never heard of anyone actually having this problem though. Learning that the current Insteon devices are really PLC-only doesn't change the fact that it's still a very quick and reliable protocol.



I'd never feel badly about going Zwave over Insteon. I was strongly considering both before going Insteon. Both are very good and each have their pros and cons. My biggest complaints with Insteon currently are lack of software to program the devices and the relativly high failure rate of earlier dimmers. I'm confident both will be resolved over time.

Posted: Thu Jan 19, 2006 9:08 pm
by Steve Horn
Does the Z-wave dimmer device have a reasonable ramp rate (soft on/off)? They sure are pricey compared to ICON. But I guess it's a "You get what you pay for" situation.

Zwave polling issues

Posted: Fri Feb 03, 2006 6:27 am
by jwadleigh
I have read a few post from others having issues polling z-wave devices like I am. Are their just a few of us experiencing this or is this a wide spread problem? I now have 20 or so Z-wave enabled devices that I can't poll so I can't control them properly. I originally gavitated to HB for 2 reasons. User created software remotes and z-wave control. If the polling issue is widespread that means I have quite an investment in switches I can't control. It's frustrating to see Meedio develop Plugins for Insteon if the Z-wave plugin won't poll consistantly. Does Meedio not realize this is a show stopper for those of use who have invested on z-wave



If anyone is polling z-wave devices consistantly I would love to here from you. Also I anyone know of another software control product that enables you to create your own remotes on PocketPC (not Homeseer) that information would be much appreciated.



Jay

Posted: Sun Mar 05, 2006 9:44 am
by JonFo
I switched from crappy X10 to Z-Wave a few months ago and am quite happy with the results. Polling vs notification is still my one gripe with Z-Wave, however, in my case (9 devices), polling works just fine. My interval is three minutes, so I’m not trying too hard to be just in time with changes. I do it mostly so the display on the PDA (which shows light status) tracks any manual changes at the switch (or via another Z-Wave remote).



So far, my experience has met or exceeded my expectations for light automation (finally after years of wasted effort and money with X10). Z-Wave really works.



Are you saying you can control devices but not able to poll their status?



What polling interval are you using?



Also, there could well be a race / collision condition in trying to simultaneously poll a large number of devices on a high frequency basis. Each device still needs time to respond. This is an aspect of Z-Wave I’m still fuzzy on.

I wonder if there is a way to sequence the polls with a ‘guard band’ or if HouseBot already does that. Scott, could you please clarify that item for us?

Posted: Sun Mar 05, 2006 4:32 pm
by ScottBot
I don't use Z-Wave polling at all, personally, so I can't really comment on how well it works for me.



However, you should be able to poll multiple devices at one time, and the Hardware Interface will manage the status poll responses without collision. That's one of the real advantages in HouseBot to using multiple Devices that all connect to a single Hardware Interface (The hardware Interface can police multiple requests and deal with timing issues and the Device can remain dumb).



BUT, the majority of the "Z-Wave" specific code that the plugin uses to communicate with the USB controller was supplied by ZenSys as part of the SDK. Since I didn't write it myself, I can't say for certain that there are not issues in their code for handling these things. In looking quickly, it seems that there is indeed some synchronization in the code, but not being intimate with all of the message flow, I can't say for sure whether it is adequate. It's also possible that there have been updates to the SDK code that have not been applied to the plugin. I forward the Meedio guys notifications about Z-Wave updates, but I never receive any confirmation that they have been addressed in the HouseBot plugin.

z-wave polling issues

Posted: Mon Mar 06, 2006 7:34 pm
by interruptvector
Jay,



I've had similar polling issues as you describe. Although I do not yet have as many z-wave switches as you do (I only have about 6 installed at the moment), I find that after housebot has been started, it stops all polling after a certain period of time (usually around 1/2 hour or so.. sometimes more..). I'm convinced it is an issue with the z-wave USB interface of some sort -- either that or the zensys SDK used to put the housebot driver together. I've read many places that using a serial interface eliminates all these problems -- but there is no serial driver for housebot so it is not something I've been able to attempt. (Just read some of the problems described in posts you can find on the Homeseer forum).



The way I get around this issue it by running a script that closes and reloads housebot automatically every half hour or so. I'm guessing that when housebot reinitializes the z-wave USB device at start-up it gets things back on track and working until it gets clogged up and stops polling again.



Can you poll any of your devices when housebot first loads? If so, maybe my workaround will work for you. It's not ideal, but it is better than no polling at all!



Wes