CHIP Protocol. Is Zigbee already obsolete?

Any thoughts on this?

https://www.connectedhomeip.com/

1 Like

Honestly, the reason why I like both Zigbee and Z-wave, is that the devices itself are not connected to the internet, just to their own network.

2 Likes

This ‘internet personality’ is a firehose of opinions but a dribble of facts.

17 Likes

Ghehehe… saw this one too. I like his way of presenting ‘improvement points’. “Ooooh the zigbee…” “Corperate greeeed.”. He might be a little cynical. Gotto look through it a bit with a wink.

But on topic, I have no idea if what Paul fears is going to happen with CH(O)IP, actually is true or based on facts. Too many unknowns IMHO.
The only thing I can come up with; if the company of CHIP does not present any real benefit of this new protocol, why should I drop my zigbee devices… Untill I read some convincing reason, I am sceptic.

But, money makes the world go round. There are thousands, milions of people using zigbee devices. The potential revenue getting them to buy new stuff would be huuuuge. So in that sense it would not surprize me at all. Too bad its not my idea though :slight_smile:

I did find him just a wee bit over the top. :upside_down_face:

I’ve been exclusively using Z-Wave so far so I’m not really familiar with the ZigBee stuff, but I’ve been looking at them for all my door and window sensors. I wasn’t sure how valid his actual points were, nobody else seems to really be talking about it I guess that should say something in itself.

To me, it looks like the CHIP protocol is more about having different devices communicate in a expected and standardized manner rather than defining how that communication will happen. I don’t see old products being dropped, especially HomeKit/Zigbee/Zwave devices which are local and part of a defined standard.

If they suceed with their goal, new products should end up being easier to integrate than today, if they fail it’ll end up being just another protocol. But there are some heavy names behind this new protocol and I hope purchasing and integrating diferrent smart devices becomes less of a chore.

1 Like

11 Likes

I haven’t read into this in detail but my guess is that the new CHIP standard is meant to be a way for the likes of Z-Wave and ZigBee hubs to communicate with each other, rather than the devices themselves. In other words, a way of not needing Home Assistant in the middle. Sure you can achieve this with Google Home / Alexa but that again requires a third party / entity in the middle.
I could be completely wrong, but I’d imagine this is what they are trying to do. Wifi simply doesn’t really lend itself well to super low power battery devices.

Just to use a famous quote,

Paul Hibbert is terrible, one-note joke that only stoners laugh at. He’s fucking clown shoes.

He is trying to make a fun of everything meanhwile understanding the business model behind it, and this way he makes users more and more confused.

First of all Wifi devices are terrible with power usages, no question. If you have a decent size house and you invest in Smart Devices at some point you will just overload your router/wifi network, not because of the traffic, but the limits how many devices a wifi AP can handle.
Currently on the market there are zillions of Wifi devices, many of them using different protocols to communicate with cloud servers and provide state updates or receive commands. Just think of it how many company has smart heating solutions or light bulbs, sockets with all different protocols.
No question, that most of the development time is wasted on developint integrations to these protocols for any central controller, being Google, Amazon, HA, OpenHab, SmartThings, Hubitat or even IFTTT.
Each protocol requires documentation, and development to integrate with controllers.
Meanwhile Zigbee and Zwave has a defined standard, what might be interpreted differently by different companies, but the basics are there. If you want to turn on and off a switch on a Zigbee network, you would send the same command to any device, not like with the current Wifi protocols, each has its own solution how it does things.
CHIP should solve this issue by defining a standardised framework for basic features on Wifi devices. It will doesn’t matter who manufactures the device and where, but if it supports the CHIP protocol every controller will know how to control it or get state updates without any additional work. It is favorable for all participant, because you don’t need to assigne a developer to each protocol, so development time can be used for adding features to your system instead of getting different protocols to work.

Zigbee is definitely not obsolete, it has the lowest power consumption from any standards, nobody has built yet self powered Wifi or Zwave switches. It is standardised, but gives room for individual development with proprietary usage options as well.
The Xiaomi example was a little bit dumb, as Xiaomi has a lot of devices which are not Zigbee certified, as they are not following the standard, but they are still can work on a Zigbee network. (Maybe the motion sensor is the only Zigbee Certified from that batch.)

But Paul Hibbert never mentions Thread, which is another Zigbee competing free protocol owned by Google, another CHIP group member. The CHIP development group has plenty companies which has experience with Zigbee standard which basicly provides the backbone for the Wifi standard development.
If you think of it, how many companies are not even publishing ways how their equipment communicates, and developers have to reverse engineer it, the whole CHIP idea is a big win for everyone.

2 Likes

I get what you are trying to say here but it bares some clarification. I have predominantly wireless sensors and have no issue at all. Mainly because I do not rely on one mom & pop all in one wif/router/modem. A simple unify setup with 3 AP lites covers a very large house and services well over 100 clients with ease.

If you have the power available at the sites you want to monitor you can not beat wifi for low latency and high reliability.

I found zwave to be hideously unreliable (at least with home assistant the way it was over a year ago).

And I am not polluting my wifi with zigbee, but otherwise have no issue with it. The best thing this alliance could do is piss that right off out of the wifi band. Then I might consider it.

They all use TCP/IP or UDP. Most of them have an API. How this API is implemented is the issue that could do with some standardisation.

ROFL :rofl: The web abounds with zigbee interoperational issues.

3 Likes

This is also applicable to Wifi :slight_smile: unless we are speaking of Tuya Tasmotiz-able :slight_smile: devices (that are becoming harder to find) or ESP32/ESP82xx DYI projects.

I was making a point about

The fact that some companies are locking down their wifi devices to cloud only requires only one response - don’t buy their crap. There are companies that are actively supporting open cloud free connection methods and we should be supporting them (Shelly for example).

3 Likes

You are right, but lets say, you want your cheap Tuya motion sensor to turn on your expensive LIFX bulbs.

You need two working integration for this, what has to be updated continuously as they change differently. Meanwhile if you would use the same standard CHIP the Tuya motion sensor would use the same protocol as the LIFX and could announce locally that it has detected motion and the bulb could react on it without a central controller, like how group associaciation works in Zigbee or Zwave. Sound cool isn’t it? The less effort what should have been used to develop integrations and connections between different brands would make it more effective for companies.
Of course CHIP is just an option for Wifi, it doesn’t mean everyone would implement it. But who would, then that should be compatible with each other.

You might want to consider how many Bluetooth devices do you have to pollute your Wifi. But usually it is the opposite way around, Wifi is polluting Zigbee, due to the low power, and they can live side by side. You can choose easily not overlapping channels. Anyhow, for high bandwidth wifi has moved already to 5Ghz, what is not affected by neither Zigbee or Bluetooth.
(Anyhow you are polluting your Wifi network with a lot of Wifi devices, and you have to use AirTime fairness to make some devices able to use the bandwidth.)

Did you know that, a bad USB3.0 port can knock down a Zigbee network with its interference? Far more harmful than Zigbee to Wifi.

I came from SmartThings, their Zigbee and Zwave implementation is quite solid, until they tried to be fully comply to the latest Zwave stack. It has turned out that many manufactures has interpreted things differently in the standard, including Fibaro and Qubino devices with multiple endpoints. The compliance attempt has rendered a lot of devices useless in a minute as they were not reporting states correctly.

The power consumption of Wifi devices are at least 10x bigger than Zigbee devices, and not mentioning the GreenPower Zigbee devices. If you have 10 bulbs, that 10x multiplier adds up and would have a sing in your yearly bill, if you consider the environment.
The self powered Zigbee switches are brilliant regarding power usage, no batteries, no cables. And they just work.

Can you qoute some? I am not saying that there isn’t but what exactly do you mean? There are different issues with different devices.
And Zigbee has multiple standards for different usecases. ZHA, ZLL, ZSE, etc… Some are compatible with each other, or can be made compatible.

No, I cant. Did you miss the part about having 3x APs?

That’s the whole wifi spectrum right there.

Which has no bearing on the matter when you consider that 100% of my 100 wifi IoT devices are on 2.4GHz.

Yes I have seen that, but really that’s about shitty USB EMI/EMC design and really has nothing to do with any of the IoT devices. Other than maybe suggesting that your low power devices aren’t all they’re cracked up to be in this day and age of pervasive cheap Chinese electronic devices. Another win for Wifi reliability.

I have 50 wifi lights. At 0.7w standby each they cost me less than $90 a year to use (compared to my $1800 annual power bill). We have close to 100% hydro power here so about as environmentally friendly as you can get. I’m happy to pay for the convenience and reliability.

Just doing a search on this fourn:
https://community.home-assistant.io/search?q=zigbee%20issue

1 Like

I can put in comparison with this…
https://community.home-assistant.io/search?q=wifi%20issue

:laughing:

Isn’t zigbee an zwave devices much more expensive to buy also?
I don’t have much interest in either of them but you need to add that too to the calculation.
Meaning even if this zigbee bulb uses less power it may not even out due to the initial purchase cost.
?

I only have one hue zigbee light but that’s a bit unfair :slight_smile:
When it works it’s great, but if it fails it’s horrible to fix.

I have plenty Zigbee, Zwave, Wifi devices. It is all the same if it fails. But I have never had any issue with any Hue bulbs, and I have a starter kit for 5-6 years already.

I’d put money on most of those being 1D-10T errors rather than issues with the devices though.

1 Like

Zwave certainly is. Zigbee can be very cost effective.

The same can be told about Zigbee as well, but Aqara is not Zigbee certified, or most of the products, and they are not following straight the standard. So most issues comes from that.
Osram, a Cree bulbs has terrible firmwares, what most people don’t know.

And by the way ESP chips can be dodgy as well…