jec6613

It honestly looks a ton like a NUC 11 Essential, right down to the price point. It looks like UDI only put in a small markup (or got a good deal on trays of 1k units): https://www.intel.com/content/www/us/en/products/sku/217669/intel-nuc-11-essential-kit-nuc11atkc4/specifications.html The N5105 has an extremely long support tail, since its target markets include embedded, IoT, and routers. Long term prospects are much better than Polisy, not the least of which is because it has enough CPU power and RAM to go forward with. For the curious, the Atlas Canyon CPUs are basically one of the quad E-Core bundles lifted out of Alder Lake (12th Gen), with an 11th Gen dual channel memory controller, and modern Intel GPU onboard. Unlike the AMD chip that runs Polisy where it was a modestly updated older design, this Intel chip is really cutting edge.

I interpreted this comment about upgrading the firmware on the Z-Wave devices themselves, which is definitely a pain for both Polisy and ISY994. Though, in my experience, at least with the old ISY Z-Wave 500 board, a firmware update would cause a decent size Z-Wave network to positively crawl (1 minute or so delays to commands) for an hour or two after startup while the ISY queried the whole Insteon and Z-Wave networks. Not so much on Polisy though.

I'd say the biggest things that the polisy gives you are much better performance, the ability to use USB PLMs and the like which expands the pool of available PLMs, and much better Z-wave support once the new daughter board (Zmatter? MatterZ? Whatever it's called) including firmware updates. You can use the Polyglot on Polisy in addition to your ISY, but there's no benefit to keeping the 994 around once you migrate to IoP/IoX. My 994 got shipped to my mother's place after her hub went offline, and before that it was sitting on a shelf for months.

I have a tool just for attaching those connectors, it's the same used when replacing a laptop WLAN or WWAN card.

My PLM ended up no longer responding after a near lightning strike, so I'm using the USB RF only dongle for now ... I'm hopeful the new PLMs come out soon because the performance is extremely poor compared to the PLM. FWIW, I only had one pre-2016 go dead due to capacitor issues, the other two PLM deaths were both due to lightning. They seem quite susceptible to strikes compared to most other devices, the only other device I had take a dive with a network switch that blew all of its capacitors at once and started leaking electrolyte from the case - the worst case on other devices was a hard reset and a restore.

If you're re-making the injection molds anyway due to wear on the old ones though, you might as well make the simpler relay one as well since it's easier to manufacture with fewer holes. Over the life of the molds I'm sure it saved money.

If you can find them still, Philips Warm Glow work very well. Most Amazon knock-off brands do pretty poorly though.

It definitely is a router firmware bug of some description. The Polisy adheres strictly to the DHCP and DNS protocol, which can cause issues with some buggy DHCP and DNS implementations. Is the Polisy pingable via IP and just not findable? In which case, it's an issue with the DNS not working properly. If not, it's an issue with DHCP not issuing an IP and Polisy being strict doesn't retry for a while.

@Michel Kohanimhas given us most of this info if you go back in his posts. eisy is a more powerful version of Polisy with some other updates, the ZMatter board will work in both Polisy and eisy with an adapter that will be available with the eisy, and you'll be able to transplant the board between the two when you choose to upgrade. My personal take on this is that I'll upgrade my Polisy for now, and upgrade to eisy when convenient to me. As for the lack of Matter devices, currently that is correct, but not for long. There are a number of manufacturers that have release devices that will work with Matter via a firmware update (including the entire Ikea TRADFRI lineup) but November seems to be the launch date for everybody.

Those are standard SMA antennas, you can easily get some 50 ohm sma extension cables.

I did as well. Boo, have to wait until November! I'm impatient and want my intermittent Z-Wave issues fixed!

Both CR and LF and supported in the GSM-7 character set; however, the paragraph mark that's inserted by just hitting Enter to feed to the next line is not. First thing I'd try is to ensure you're using the CR character by typing the message with Shift+Enter for each new line. Another possibility is that you're sending to an SMS gateway that strips formatting - I know Verizon in the US actually has two distinct gateways, one for SMS and one for MMS, that handle pure text messages differently.

I agree, it's not a good idea, just one that is technically possible. People also use male-to-male extension cords for Christmas lights ... just don't do it in the house next door to me. As long as you have constant power to the receptacle you can put it in the same box as the outlet, then there's no need for the 3 wire cable, just a few inches of pigtail. If your home uses some of the non-standard outlet shapes, it's a great trick to Insteon or Z-Wave enable them, and is also what many of our friends overseas do in order to get smart outlets.

A pair of diagonal cutters fixes that post haste - you can run each socket on a different input no problem.

You can enclose it to make it outdoor rated. Another option would be one of the Z-Wave or Insteon micro dimmer modules - though this runs into the, "You're dimming a receptacle and that's not at all to code," problem, so remember to rip it out if you ever sell the house.

Given that your customers likely have more devices on average, they may see quite a bit of benefit to helping you because we can and will purchase many of the Nokia devices if ISY/Polisy/Eisy support them. I'm hopeful but definitely not counting on it.

I mean, $200 to not have an entire Insteon network end up unable to talk to a controller seems perfectly reasonable, but that's why I'm not a business owner.

In keeping up with the latest news, both Intel and AMD have now both forecast about a 20% downturn in near future chip production due to so many ongoing order cancellations. So you can probably spread that across most of the chip production industry. Not a good sign for the tech economy but it does mean that some orders are probably moving to the front of the queues. Of course that is probably only the tip of the iceberg as many buyers were placing duplicate and triplicate orders with multiple vendors hoping Intel's fabs have one customer: Intel. So order cancellations aren't the problem for the, they're not a foundry (yet, they are entering that business). For Intel this is about demand softening. AMD produces no chips, so is also 100% about demand softening. Other orders are moving to the front of queues, but nobody puts in orders across different manufacturers for the same product as they're not interchangeable. One cannot simple swap from a 35nm Samsung to 35nm TSMC node, the chip needs to be redesigned with the target fab and node in mind. This is why when the GlobalFoundries fab in upstate NY was closing, the USAF put in an emergency $1Bn order for chips that could only be made there as the last fab on that particular process to keep a number of chips in stock through the end of life of the equipment they were installed in.

You can just get the Aeotec door sensor and put an arbitrary long sensor on it. The new Aeotec leak sensor and door sensor are the same device, one just has a water sensor soldered on where the other had screw terminals.

The door sensor was originally the TriggerLinc and they advertised it more heavily - you can use it to handle any dry contact switches including a simple push button, which I've used in a few locations either with a remote reed switch itself could be subject to weather (and therefore bad to have the door sensor itself on it) or to use a pair of reed switches to detect either of a pair of doors using a single door sensor - though in this case either door closing would show the sensor closed, it's still effective in certain situations - in my case, double door reach-in closets where the light only turns on with both open, and will turn off with either closed. Unless I'm doing a direct link to a light switch though, nowadays I use the Aeotec Door Sensor 7 Pro, which is smaller and performs just as well.

You know that with a simple remote probe that the door sensor becomes a leak detector, right? ?

This also has less obvious but more important uses than that - if you have an Elk alarm system or similar critical device that you've linked to the Polisy, the built-in switch means that you do not rely on the rest of your network being up at all for that integration to work properly. Depending on how robust your network is, that may be pretty important!

Are you on WiFi? The AC is very sensitive to even a momentary connection drop, I had to disable all of my NIC's power saving to keep it from disconnecting periodically - but since then, well, I launched it 2 days ago and it's still open and connected.

As the end of June is upon us, any news on this? I don't actually expect to have it in my hands for a few months, but I'm ready to whip out the credit card and upgrade my original Geek Batch Polisy.

WiFi (802.11n era) needs to give the clear to send signal at minimum intervals for all STAs. Once you're on 802.11ac with MU-MIMO, or 802.11ax, the problem mostly disappears. That's why in 802.11ac Wave 2 was such a boon for dense configuration. But, most IoT are still 802.11n 2.4 GHz only. As for the backhaul, on good mesh devices, it's a dedicated radio so there's nothing eating up bandwidth. As an example, the Netgear RBRE960 sports: 2.4 GHz 802.11ax for clients - 4x4 (1200 Mbps PHY) 5 GHz 802.11ax for clients - 4x4 (2400 Mbps PHY) 6 GHz 802.11ax for clients - 4x4 (2400 Mbps PHY) 5 GHz 802.11ax backhaul only - 8x8 (4800 Mbps PHY) Now, obviously, in a normal mesh configuration you'd be using a 600 Mbps 20 MHz channel for 2.4 GHz so you could get an aggregate 1800 Mbps across three WAPs, but, yeah. Good wireless backhaul can outperform Gigabit Ethernet. Yes it can, although in IPv4 nobody's done it, it's been used in IPv6 for some experimental uses and backhaul among networks. This depends on the vendor, but the high quality ones use similar auto channel selection to a traditional Cisco controller - which means normally yes, different channels. They also use the power back strategy on a per radio basis, and on newer WiFi 6 devices they dynamically adjust power based on which STA are currently connected and reduce power to radios on other WAPs.