upstatemike

Did anybody get into this? Sounds like ISY was discussed quite a bit during the session.

Caseta: Switches cannot control other switches. They can be controlled from a Pico remote but most folks need control at the location of a load bearing switch so no go. Not recommmended. RA3: Not supported by ISY so not recommended. Z-Wave: Does not support smooth synchronized operation of groups of switches, bulbs, or anything. Not recommended. Zigbee: Requires Zigbee 3.0 for synchronized operation of groups. No high end Zigbee switches availble with 3.0 yet. Future option? Thread: Only used in Apple Homekit products at this time. Future uncertain. Matter: No firm release date, no detailed feature set. Future uncertain. Broadlink Fastcon BLE: No standard form factor switches yet. Future Option? HUE: No load bearing switches, only bulbs. Limited applications. LoRa: Dimmers not yet released. No local API yet. Future option? Insteon: Company in recovery with no new products available yet. Future uncertain. X-10: Prone to false triggering from powerline noise. Limited applications. UPB: Limited integration options. Future uncertain. Control4: Not available for DIY so not recommended.

I agree. I don't care about Matter apart from it is billed as the next standard and I want to know if it will meet my needs because currently Z-Wave, Zigbee, Caseta, Hue, Yolink, WiFi, and Bluetooth do not.

Most people don't need it initially when they first start using Home Automation. They appreciate the "starter kit" conveniently available at their local Home Depot, Target, or Wal-Mart. They are confident in their purchase decision because the product description promises a wide variety of expansion options so your system will grow with you. But soon the new user is comfortable with the technology and is ready to trade simplicity for some added features... and this is where the disappointment starts. The limitations of the "simple" interface becomes a problem when they want to do something more and find the logic won't support it. They want to expand their system only to find the big stores never get around to stocking those promised expansion items and there are only ever "starter kits" on the shelves. This is when buyers remorse sets in and they desperately come here looking for a way to salvage their investment and find a path to accomplish what they were promised when they originally committed to their particular protocol/brand/product. Even Apple Homekit users who have committed to the walled garden of the Appleverse eventually find it is not enough and seek salvation through Homebridge. It will be the same for those starting with Voice Assistant based systems. Eventually they will want more and will need something like eisy to move forward.

Let's not forget there is more to the story than just connectivity. From a connectivity perspective I could do more with Alexa than I do today but I am prevented by the lack of advanced logic options. You simply can't do everything that is needed with a simplistic trigger/condition/action model. You need OR logic in your triggers. You need nested logic. You need counters and flags and virtual devices. Making Voice Assistants into edge routers is not going to eliminate the need for integration controllers like polisy/eisy.

While I respect your preference I am just the opposite. I absolutely HATE lugging a phone around when I am at home. I cannot stand the time it takes to fish it out, fire it up, get to the correct screen, and actually do something with it. By contrast I have been using Alexa since it was introduced and while I was skeptical at first about learning commands and waiting for a cloud based response, I have to say the reliability, accuracy, and natural speech (no robotic voices or unnatural command structures) have won me over. I now speak to my 25+ Echos throughout the day and rely on them to control most things.

In some respects I care less about Matter than I do about Thread. Matter does not solve any problems that I am currently having but if Thread turns out to be an underlying protocol that fixes all the shortcomings in Z-Wave, Wi-Fi, and Zigbee while providing a larger universe of devices than is available from Lutron or LoRa then that could make life better.

So Z-Wave really is unfixable! Let's hope Thread/Matter is able to avoid these problems. Also wondering how Zigbee 3.0 will do by comparison.

Don't Z-Wave Group commands use multicast? I thought it was included in the protocol someplace. I also thought Home Assistant made an announcement that they were going to start supporting Z-Wave multicast last year? https://community.home-assistant.io/t/ok-zwave-js-has-multicast-now-how-do-i-use-it/321223

I am surprised to hear that Assigned Associations use routed messages. I would think that it would use a multicast to all of the associated devices to reduce traffic and assure a synchronized response. What is the logic of using routed messaging for associations?

All of these things could also be done with Assigned Associations. There is no real value in having 2 different types of associations in the first place. Just use the one type type that covers all scenarios.

I'm not sure I understand the point of direct association. The chances of two devices that you want to associate being in direct range of each othe seems remote except in the smallest, obstacle free and interferance free environments. I would expect Assigned Association to be the standard. I can't think why any Z-Wave device would be designed to only support one group so this also should not be an issue. I understand the S2 security requirement but if the enrollment process was designed to only allow devices to be added at the same level as existing devices there should not be much chance to get into trouble there either. It still looks like the protocol can do what is needed if implemented correctly. Maybe there needs to be more safeguards to prevent manufacturers from not implementing the full feature set and to restrict users from implementing settings in a way that could disable those features.

I have been critical of Z-Wave because of the popcorn effect and the lack of useful keypad options but in reviewing details of the protocol I am confused on why these things are so bad. The Z-Wave protocol includes Assigned Association which should let you link multiple switches or keypad buttons even if they are not in direct range of each other. Once linked I would expect they would all operate in unison the same way linked Insteon switches do. Z-Wave also has Groups. I would expect that all switches assigned to the same group should respond in unison without any popcorn effect because they are all responding to a single command instead of individual commands sent to each switch. With these two features included in the protocol can anyone explain why Z-Wave switches can't be made to behave exactly the same as Insteon switches?

I tried one of the Wiz Wi-Fi bulbs and did not like it. There is no color saturation and everything is just pastel shades. Not acceptable by any standard.

There are 3 "Deal Breaker" issues with Z-Wave that show no signs of getting addressed: 1- The popcorn effect where groups of devices cannot operate in unison. 2- The lack of keypads where the buttons can be used as switches the way Insteon buttons can. Zooz makes a nice scene controller but all it does is load a value into the central scene and allow some rudimentary associations. If you want to truly keep the buttons and associated LEDs in sync with other switches in a group you are going to be relying on a lot of programming in your controller and even then the results are sub-optimal. Z-Wave simply does not have a command class for keypad buttons that do anything other than trigger scenes. 3- Like Insteon, Z-Wave is a single source technology. You are at the mercy of Silicon Labs for all Z-Wave hardware and they do not appear to be any better than Insteon when it comes to responding to what end users want; focusing instead on catering to the companies who pay the license fees. This allows companies to implement the protocol in a way that promotes brand lock-in if you want to use advanced features and defeats the idea of interoperability between brands except at a very basic level. To be fair though it looks like Matter is heading down this same path with most of the delay in introduction being down to negotiating with participating companies who want to preserve ways to "differentiate" themselves in the market place. This does not mean unique styling or competitive pricing but rather some leeway in implementing the protocol to lock folks into a brand in order to get proprietary features. Zigbee 3.0 looks like it may solve the popcorn effect problem. At least the demos I have seen of the Inovelli Blue Line switch suggest it will. I am also curious to see what the new Yolink LoRa dimmer can do. Z-Wave however seems to have stopped improving and may be approaching end of life as a competitive protocol.

I purchased my Polisy with no wireless boards. To upgrade it do I need the Matter board plus the Wi-Fi board? I use ethernet so is there anything that Wi-Fi will add for me? Some Matter edge router protocol translation magic something?

Maybe they didn't get enough hub subscriptions to stay viable until they had products to sell.

Just an update to my own post. I completed the transition back to Insteon and the result is better than I expected. Everything is configured the way I want (not even possible with Z-Wave) and it just works. Everything is link based so it will keep on working even if my ISY goes offline for some reason. This is what I expect from a lighting system. Hopefully Matter will do as well in this regard.

I didn't mean Matter will fix Z-Wave (I am convinced now that it is unfixable). I am just saying Matter will not have the same problems that Z-Wave has. Matter switches and keypad buttons will all work as controllers regardless of manufacturer. You will be able to backup a controller from one manufacturer and restore it to a controller from another manufacturer. Direct association will work as expected between all devices and all buttons on keypad devices regardless of what mix of manufacturers you are using. There will be no popcorn effect when controlling groups of devices. Latency will be imperceptible under all conditions. Etc. etc.

Looking forward to Matter to solve this kind of nonsense.

It kinda still doesn't exist. Maybe they will finalize the specifications pretty soon; maybe.

Even if the Nokia hubs were available the initial quantity produced must have been small so it doesn't sound like it would be a sustainable solution in any case.

So there are instuctions for fixing bad PLMs but is there a DIY way to upgrade them? Besides caps I am thinking overclocking, added RAM, tweak the power supply to broadcast a stronger signal, etc.

That works for me!

Hopefully the PLM Pro will be an Insteon device similiar to what the XTB-II was for X-10 http://jvde.us/xtb/xtb-iir.pdf