Zigbee / Matter Motion Switch | Project Linus (Blue Series)

Hey guys – just popping in here to address some of the questions. I was MIA last week working on something for Apple (the feedback from this thread was referenced a ton). Hoping they’ll listen, but we’ll see!

Great idea – I’m going to work on the messaging this afternoon!

It’s one of those, “it should be able to handle it, but we’ve been getting mixed messaging about it, so let’s just be safe and not promise anything anymore”.

We’ve been able to update (via a wire harness) a Fan switch from Zigbee to Matter as both the Blue and White Series Fan Switch uses the MG24, so in theory it should work on the mmWave version, but until we’ve successfully done it, I don’t want to promise anything.

Good to know :slight_smile:

I think I’ve dug myself into a hole with this by saying we can just swap orders to Matter if you want it, so another Kickstarter wouldn’t be necessary. We can just eat the cost at this point to make things right. I just want to make people aware that Matter still has a way to go and if not Matter, then the hubs that support it do and while the switches have all these cool features that minimize user frustration (zones, interference areas, etc), those will not be possible on most Matter hubs.

I’m with you on putting a disclaimer on the site that says it’s not possible with Apple, Amazon, etc, but the amount of people who still order and then blast the poor user experience may be enough to convince others the product isn’t ready for prime-time, which is my concern.

Now, the mmWave sensor does have a built in, “learning-mode” which I think can be utilized and will work for Matter users on Apple, but I’d like to make sure it actually works as well. How this works is that you turn on your fan, or whatever other object you think will interfere with the sensor’s readings and then leave the room and the sensor automatically finds those interference devices and excludes it from its readings. In other words, it will set the interference areas for you.

Great question – without giving too much information (as I’m just honestly not sure what I can and cannot say and am erring on the side of caution), this person’s job is to help with integration with Matter devices (he also does Zigbee and Z-Wave) and works directly with the major hubs to at least make them aware of the issues. His words were, “Apple is a tough one to get in the door with unless they see a product that is new and exciting and benefits them – if they find those things, they will then be all over you to work with them”.

I was skeptical of the conversation going in, but given who he works for and who recommended him, I was actually impressed with the meeting and have faith that our concerns will at least reach Apple. Now, if they find our stuff new and exciting, I think they will, but I’m also biased lol.

Yeah, I agree. There’s definitely promise with Matter, and I firmly believe it will succeed. We’re just stuck in this interim area right now where we want to put out technologically advanced products, but with that leads to poor user experience until the hub manufacturers can catch up with their UI’s.

I do think the mmWave switch will be what persuades them as I had a conversation with my personal favorite IoT company and they could barely keep still when I told them we were working on an mmWave version of our switch. I made sure to reference this in the Apple presentation to see if it would budge them at all :wink:

You’re welcome, happy to help!

Haha, yeah… well the good news is it seems the sensor is in metric, so it will be us American’s that will have to use the conversion tables!

Great question – I think @bobzer knows more about it than me (I am honestly not sure) but maybe @EricM_Inovelli can help with this one.

Aside from any technical concerns, there’s also a capital concern. I’m not sure we have the resources (both time and money) to create an app right now. On the other hand, maybe we can’t afford not to. Something to consider.

Haha, ok, and I’ll add at the end, “our community members will also call you out for ignoring this disclaimer”.

I agree with you in terms of your Thread comments and mixing them up with Matter. Again, what I think the challenging thing for us is exactly what you hit on in your summary… it’s the people who see Matter and just buy it, only to realize that a lot of the advanced features shown cannot be done from within the UI of Apple, Amazon, Google, etc and that leads to frustration. And I know there’s people who use Home Assistant, SmartThings, and Hubitat where a lot of these settings can be accessed, but as this is primarily targeted at mass market, most mass market customers don’t really care about Thread vs WiFi – they see the word Matter and assume it works great with their Apple, Google, or Amazon device.

In all fairness, I understand their grievances as I remember the days when I first started and just bought devices based on the fact that it said it worked with my hub. With no programming background, I didn’t want to mess with anything crazy, just wanted a device that did the basics (which ours does too, but I think they get roped in by the flashy features).

What I’m trying to say is that it boils down to the marketing message which is my problem and I just need to sit down and figure out how to market to both the mass market which drives the sales for the White Series as well as educate them that the devices are limited on the mass market platforms.

Maybe I’m carrying some pessimism from when we first touted Z-Wave and I felt like there was zero support for newer customers who had no clue what Z-Wave does and therefore didn’t want to buy our products because they were much more expensive than the WiFi equivalent, but what I struggle with is that I don’t think the average customer cares about Thread (or any of the other protocols for that matter – pun intended). Rather, they just want to see a product that’s compatible with their hub.

Without boring you to death with marketing, but here’s the uphill battle:

  • We’re never going to win on pricing – it’s impossible to compete with Chinese manufacturers that can sell a cheap matter light switch for $10, so our strategy has always been to win with firmware, win with customer support and win with you guys who help us develop amazing switches
  • That said, if a mass market customer sees a light switch that is $10 or one that is $50, they typically want to understand why there’s such a price difference. That’s where the flashy videos, cool features, etc are shown to say, “we’re more expensive, but you get a ton more features”
  • The problem then becomes, “well, in order to use these features, you have to either set them up manually at the switch, or use some weird workaround in the UI to get them to work” and then they become more of a headache than it’s worth

I can list out the features of Thread until I’m blue in the face, but at the end of the day, based on what I’ve seen with Z-Wave (which I know is slightly apples to oranges as Thread has the support of the major mass market hubs whereas Z-Wave does not aside from SmartThings) no one is paying a huge difference in price for a protocol when, in their mind, WiFi works just fine (sure, they can learn the hard way, as I did, that WiFi products come with their own issues, but who knows if they’re willing to take that chance).

Long winded way of saying, I know Thread makes sense to the power users of the world who care about which protocol their devices run on, but I promise you that the average person does not seem to care. This was a hard pill for me to swallow, but I had to think back to the days I first started and had no clue about Z-Wave, Zigbee, WiFi, etc. I just wanted a Wink compatible product.

Great question – we should know more when we start beta testing it with the larger team, but I can speak at a high level on my and the other Eric’s experience thus far.

For the first question, 100% of the time, the sensor still tracks the first person. I’ve never had it not detect me when I was sitting there and someone came and went and I’ve tried this at least 50 times.

For the second question, some use-cases for multiple people tracking would be:

  • To understand if your teen has her boyfriend over when you’re not expecting him to be… no further comment lol
  • To understand how many people are at your house or in a specific room at a given time (let’s say you want to make sure all your kids got home from school) – I know there are other ways to solve for this, but this would be another data point
  • If you want to get crazy, you could have separate automations based on where someone is in the room (ie: bed lamp on the left side is off if it’s just you and you’re on the right side of the bed, then turns on if your significant other joins).
  • An interesting use-case that was brought up was to use these if you’re monitoring your elderly parents – like why hasn’t dad moved in 20 hours and where is mom type deal

Yeah this is an excellent point – this hit me last week when I was thinking about this more. Glad I wasn’t crazy!

Yeah this is an interesting idea – let me see how much cost this would add. I think @adam5532’s point about how other, non-smart switches typically aren’t spaced properly makes me feel much better about it too.

Another good possibility. I’m not sure a needle-nose would be able to grip it though as it’s kind of small, but let me look into this as that’s a good idea.

I wish I had that pull lol.

Hmmm… let me look into this. We’ve had some weird issues with our site the past week. I planned on looking at all of them today now that I have some spare time.

Yeah absolutely – shoot me a PM if no one has worked with you yet. I’m not sure why no one has responded, but I’ll make sure someone helps you today if you haven’t already been taken care of.

4 Likes

…and then a year later, they buy you out for gobs of money and destroy your product from mis-management…with Dark Sky weather being a prime example.

Using non-China overseas project programmers may make the turnaround time palatable, as long as you get the source code and own the Apple/Google store entries where it is deployed. My former employer made that mistake…more than once…and had at least four “official” apps that they couldn’t delete or update…

1 Like

I’m unfamiliar with how this might work in practice, and by no means, do I think what I’m about to suggest be expected for launch, but after you described how the detection and interference zones worked and how a beta feature is people detection, it got me thinking if pet detection was possible? Really I’d just like to advocate for at least thinking about how that might work if any considerations might need to be made up-front to support something like that.

For example, I could think of a way to get some form of pet detection working by re-purposing an interference zone as a separate detection zone. I don’t even know if that’s possible, but again, if some up-front consideration needs to be made to make that possible, that might be an interesting way to support something like that where below a certain height it would be able to detect n-legged friends and above a certain height it would be detected as a person. Of course, there are some obvious issues with this solution but still, I think there’s something interesting there.

1 Like

Any version of this switch would benefit from an app as a way to tune the detection settings in real time. Having to measure and guess at co-ordinates and then watch the operation to see if it’s correct will be a pain in the ass.

1 Like

As someone trying to live solely in the HomeKit universe but finding a need to run Homebridge for a few items that didn’t have a good native solution, I’m EXTREMELY interested in a Matter/Thread version of this mmWave switch.

The current Matter/Thread White switch, with the updated firmware, has really given me a taste for what’s possible with a “switch as an appliance” and I’d LIKE MORE! But even there the switch parameters aren’t accessible inside the Home app. So the cost for an iOS app to set mmWave parms could actually be spread to the existing white switch. I’d contribute to a kickstarter campaign for that. I do initial setup for Lutron in their app but thereafter pretty much use Home app. The same for Hue and Eve. But having the manufacturer’s app available to address unique features for setup is something I view as a large value add for those respective manufacturers.

2 Likes

Why specify coordinates? Have the switch support a training mode where it tracks the person to outline a good or bad region. You can use the config button to start the train mode in 30 seconds and when movement “stops” for 15 seconds, it commits. The user can also command from HE, HA, etc. the only issue is if you move beyond the sensors field of regard, you will have to re-enter it within 15 seconds.

It’d be decent to have a mode like htat, but I doubt it would always work perfectly for everyone.

I’m curious, just generally, how many Inovelli users are “run their own server” Home Assistant/Hubitat users versus Amazon/Google/Apple users, and therefore how many would actually be impacted by the lack of easy configuration in the cloud-based smart home platforms.

As much as I want Inovelli to become a household name, because these products are flipping amazing, as a Home Assistant stan, I wonder how those using (IMO) the harder-to-use systems come across Inovelli products and decide to pursue adding them to their system.

3 Likes

We have been hearing a lot about the mmWave (with good reason), but I have not heard much on how the testing has been for the lux sensor. Has the side light bar been an issue in testing? That feature was the reason I found this switch, so hoping no news is good news!!

1 Like

Here’s my $0.02 Eric:

You have to try to serve a dual market: the DIYers who are running Home Assistant are happy to take the product you have now with X and Y and Z values and will tinker until it’s performing as they want and won’t complain; but the HomeKiter out-of-the-box-ers will be confused and won’t handle instructions about tapping the paddle twelve times to set the Y value of the interference zone… I think the play is to have the firmware out of the box have some sensible defaults so it behaves comparably to an occupancy sensor that you can get from Home Depot so that if you don’t mess with the configuration you are no worse off than that. Then market the mmWave capabilities as an expert feature that can be unlocked so that expectations are managed correctly to the mass market audience.

Building and maintaining an app is a serious commitment, so one idea is to instead build a mobile website where you can enter the X/Y/interference values for your room with some visualizations and then hold up the phone against the lux sensor in a programming mode and have it play out some flashes of on/off bright/dark to transmit the values from the phone/website to the device. That would be far cheaper to build/maintain and don’t have to worry about discovering the device on diverse local networks.

We’re rooting for you and the home assistant-ers like myself are looking forward to taking delivery of this mmWave matter+thread switch!

6 Likes

Then you need to start putting updates back on IndieGoGo for those of us who didn’t get the memo on the forums.

A lot of people outside this forum think the project has been a rugpull because of that.

3 Likes

This is amazing news!!! and how I expected mmwave sensors to work as well from what i have seen from brands like screek. Thanks for the update!

I think you are right here. Many of us power users have been burned by cloud-controlled devices and have sworn off wi-fi for many reasons, but Matter over thread changes things a bit. It has most of the benefits of the major home automation protocols (local control, speed) with all of the benefits of wi-fi (familiarity of use by users, no need for a dedicated hub, ability to have branded app control, cheaper.)

I think is why you see companies like Leviton have moved away from Zwave/Zigbee/Thread with their latest devices and are going all-in with Matter over Wi-Fi. If Matter does end up delivering on its promises, I don’t think Thread will be the winning protocol like many home enthusiast nerds think it will, I think the market will end gravitating towards Matter over Wi-Fi.

1 Like

Hey all, I couldnt find a thread for the red series version, but I was wondering:

  1. Will the features be the same as the blue?
  2. Is the timeline about the same?
  3. What is that timeline?
    Thanks!

Latest update (~1 month ago) here:

Thanks!

Neither physical layer needs to “win.” Wi-Fi, at least as it exists currently, would be a poor choice for most battery-powered Matter devices, as it is more power-hungry. That’s why you’ll normally see Thread as the protocol of choice for Matter motion sensors, etc. Since those devices work best with Thread routers on the network, I’m glad products like Inovelli’s exist to serve that function, even if Wi-Fi is a capable protocol for this particular device. :smiley:

But I certainly agree that most users could probably care less about protocols. And the Wi-Fi entry barrier is lower since you don’t need a Thread Border Router, so that’s definitely one advantage (even if lots of people have one already, whether they know it or not, and Matter hides most of the need to know this away from you — if you have one, at least).

1 Like

Most user will start to care when they will have problem. Most mainstream Access point will struggle to have more than 20 concurrent connection. not event talking about 100 or more. But we can have zigbee network with more than 100 devices without any issue. also there is the mesh capabilities which is very important and at to the reliability.
Thread allow to have redundance (multiple border router) which add more to the reliability, wifi again will not (even mesh wifi because you should not have a big other lap of your access point to avoid interference)
The only real benefit of wifi is for the beginner because it’s easy to setup but here I think you target much more advanced setup which can’t be just wifi. (the other benefit of wifi is the bandwidth but except for the specific usage of camera which should be via ethernet when possible, that amount of bandwidth is not needed)

Wow, nice new video from Linus just dropped! Linus Tech Tips Video

Great that they got sponsorship from Homey but really would have loved to see Home Assistant featured in this video! Great that he is highlighting the struggle many have with zigbee and zwave in the real world, just wish Jake would have touched on the potential of thread a bit more. Wonderful advertisement for Inovelli though, congrats team to getting the beta to Linus to be featured in the video

2 Likes

I get that Linus is a big reason why Inovelli is making these. But man, that guy is insufferable.

4 Likes