Roughly one year ago, Silicon Labs released a publicly available Z-Wave SDK and a Raspberry Pi 3 Image to make it easier to work with the wireless protocol targetting home automation.
But Z-Wave specifications were still closed, which meant Z-Wave chips could only be purchased from Silicon Labs, a bit like LoRa chip can only be purchased from Semtech. The advantage of being closed is that you’re the only supplier, but this will limit market adoption, and customers may be wary of relying on a single partner for their long term plans.
That must be why The Z-Wave Alliance and Silicon Labs have now decided to open Z-Wave specifications to Silicon and Stack suppliers. That paves the way to third-party software platforms and Z-Wave radios from some of the 700+ companies which are members of the Z-Wave Alliance.
The Z-Wave specification release is scheduled for H2 2020, and will include the ITU.G9959 PHY/MAC radio specification, the application layer, the network layer, and the host-device communication protocol.
The Z-Wave Alliance will maintain its current certification program and expand certifications to both hardware and stack certification and product manufacturers with application layer certification.
Development on the opened Z-Wave specification will be handled by a new working group in Q3 2020, and details on the silicon and stack platform certification program also will be announced at the same time.
Thanks to Andreas for the tip.
Jean-Luc started CNX Software in 2010 as a part-time endeavor, before quitting his job as a software engineering manager, and starting to write daily news, and reviews full time later in 2011.
Support CNX Software! Donate via cryptocurrencies, become a Patron on Patreon, or purchase goods on Amazon or Aliexpress
Still feels like too little, too late. Who’s going to invest in making chips for Z-Wave?
Anyone who designs and makes Zigbee chips can probably make both Zigbee chips and Z-Wave chips in the future
Z-Wave has a lot of peculiarities that ZigBee doesn’t have, which makes it a lot less appealing to make. ZigBee is 802.15.4 and as such can do several other protocols, whereas Z-Wave has no such advantage.
Zigbee uses the standard ISM global bands. 868, 2.4 Ghz etc. while Z-Wave can use less crowded frequencies. This is an advantage, but this also increases the price of the chips.
Interesting. Maybe we will see some really cheap chips from Chinese manufactures. Z-wave didn’t have any success yet. Maybe this changes it a bit. We will see. Would like to see open Lora. This is currently the only thing I don’t like about Lora and avoid using Lora when I can.
Sorry what? Z-Wave is the main standard in the US for home automation by far.
Is it? What company uses it? I only see 802.11.4 (Zigbee), Wi-Fi, 433 (ASK, FSK), Bluetooth and Lora everywhere.
Z-Wave is also the premium standard in Europe for high end home automation. Other than Philips Hue, Ikea Trådfri and Osram stuff which are all dinky toys compared to the existing Z-wave products on the market from Fibaro, Qubino, Aeotec (Aeon Labs), Zipato, Nexa, and such
True. But as you said: Premium. The premium market is not that huge. I, as a engineer, even don’t participate in this premium market. The huge market potential is more at the low end. Currently only Xiaomi is trying to get into this market.
They opened it up cause they saw zigbee get a huge vote of confidence from Apple Google and Amazon.
Don’t forget IKEA, Philips, Osram, And Xiaomi
True. Especially Xiaomi. IKEA, Philips and Osram have to follow standards in order to compete with Xiaomi. They have a huge advantage with their low priced products. Xiaomi is the next Apple. Closed system. At least they are trying hard enough.
Has anyone tested Espressif’s “long range” WiFi mode ? From what it seems, it should provide 250 kbps over a kilometer or so, maybe that could be a alternative to such standards which require dedicated hardware.
interesting, wait and see their price, so far I hope LORA can be totally open to speed up its adoption before NB-IoT takes over everything with its crappy power-hungry and expensive approach.
> interesting, wait and see their price
Well, there’s no extra price, it’s available *right now* in any ESP32 if you use a recent enough SDK.
> NB-IoT takes over everything with its crappy power-hungry
That’s precisely the problem in IT industry nowadays: with increased battery capacities and availability of low-power plugs everywhere (USB), it seems power doesn’t count and allows lazy and incompetent people to do whatever to sell their crap. Some people dream about higher capacity batteries to increase their phone’s autonomy but they don’t understand it works backwards, we actually *waste* as much battery’s energy as permitted between two charging cycles. 20 years ago pocket computers like PSION could run for one month on a pair of batteries with their reflective screen. Nowadays your smartphone can achieve the same task for 4 hours before requiring to be recharged. The difference is Java, bright screens, animations everywhere and lazy coding. I still hope that IoT will force this to change a bit, but I have doubts, considering that in order to replace my old home thermostat which runs for two years on batteries, the best I can do is one week with ESP8266 for now :-/
Don’t you guys get tired of this “everyone that doesn’t manually input machine code via switches is lazy and incompetent” trope? You would have thought all the super badass greybeards would have enough experience to know better.
For having seen absolutely horrible things in field literally taking several seconds to produce trivial pages on 8 cores machines *only* because of laziness and incompetence combined together, no I’m not tired of saying this. And the worst part of it is that once people go in the wrong direction, instead of declaring failure and trashing everything to restart cleanly, they insist the wrong direction and stack even more crap on top of it to hide most of it (like caches and “pre-heating” machines or whatever just to hide a part of the misery). I’m not advocating for anyone doing machine code, but at least it would be nice to think 3 seconds before deciding to embed something more or less close to your needs but which does and undoes the same transformations up to 500 times with no added value because you just picked something wrong 🙁 It’s unimaginable the number of times data gets transformed/transcoded between layers adding and removing value across all the stack just to finish at the top with something very close to what it was at the beginning but via a very long path.
In the embedded world there is so much more going on than just not giving a shit. For starters a ton of it is outsourced to cheaper places and those guys generally aren’t wasting a week unrolling a loop to save 1 machine clock because they’d get fired if they don’t have something that works in the tiny amount of time their boss promised.
Don’t see much of a future for Z-wave, too late for getting traction as an open standard
1. non IP based
2. Different frequencies for different countries
3. Limited mesh capabilities compared to thread
Maybe they will ask the IETF for an IPv6 standard. Otherwise they are not able to compete.