Too soon? when is the C8 coming out?

I'm being offered 'touchlinking' zigbee 3.0 devices from Alibaba - and it does intrigue me. I also look with disdain at the 100Mb ethernet port the c7 came out with. In my home LAN I've got 4 100mb devices - all of them are 15 yrs old and I dislike them as they are weak and perform slowly.
While there is a ridiculous argument that 100mb is fine (it's not) I feel the c7 isn't valuable enough to invest in so I'm wondering what the Hubitat hardware platform future holds. Small applause for zwave upgrade of the c7 but it's not enough for me to move up. oh, and while I'm at it - how bout POE support as well? My hub is literally sitting on top of a 180w budget POE smart switch thats got 5 more open ports.

Ok, I'll bite. Why is 100 Mbit FD not sufficient for this application? It's not a media streamer.

Also: These are a thing that exists.


Not a use-case for this application, but it'd be nice to be able to run a speed test from my hub and propagate that data :slight_smile:

1 Like

Hubitat could come with 10BaseT and be fine. My Roku Ultras aren't even gigabit. This argument doesn't make sense to me.


This seems like a use case on the edge of relevance when it comes to Hubitat. Certainly something that I would oppose if it meant that the cost of the hardware were to rise because of it. If you really need it, I would say that it should be trivial for a groovy developer to write a driver to store a value, expose it to MakerAPI and then write a python app to populate it via webhook.

1 Like

There's only one thing that really need to upgrade to a C8... not that it Matters. However, if it did Matter, I would buy it as soon as it was released. :wink:


I’m curious about this too. What does Hubitat do with its connection to a LAN that requires a gigabit Ethernet port?

Re: when the next hardware revision of the hub is coming out (if there is one), they have not publicly commented on that in the past, and I think it’s very unlikely they will now.


Good question. Even with LAN integrations for lighting (like Lutron Caseta), my motion lighting response time is on the order of 130-150 milliseconds from motion detected to the Lutron bridge reporting back to Hubitat that the lights are on.

So I guess the actual time to when the lights turn on is a little LESS than that.

1 Like

100mb is just fine. Power usage is more consuming with 1G and there is no need for it. Here is my view where I would take the HE product line with C8.

Higher Performance Box - More CPU \ memory and storage capability (Removable Flash)
Build SNMP engine Receiver
Build Ping\ICMP Management (DONE)
Build a analytics engine collector and storage with web reporting (flash)
Build export to external data sources ODBC, XML, CRV
Build better SMS, EMAIL Notification engines w\ Report Formatting (MUST)
     - Hubitat can charge a service fee for SMS and EMAIL.
Look to Ethernet PoE Device Automation IoT Standards
Partner \ Integrate Camera Security Systems -w / Motion / ID / Etc. aka like Verkada
1 Like

Wait... like ADOBE flash? :flushed:

Edit: oooh, I bet flash storage... better-ish.

I knew this was going to happen - and it tickles me to no end that I'd be forced to defend this. we can get in the trench and talk about:
Buffering on 10G switches (a huge problem)
comm protocol wrapper efficiency.
The fact that Hubitat recently began adding the ability to store files on the device
timeout to internet
smart switch optimization tables
Bill Gates saying 16k was enough

Seriously? we need to discuss this? And the Roku thing made me go look - just ONE comment of many:
Community Streaming Expert
‎01-02-2019 08:06 AM
Re: Why does no Roku streaming device include Gigabit Ethernet?

"twiceover" wrote:
Because even local 4k streaming won't saturate a 100mb connection.  Why spend the extra money for something none of your users are going to need.

Actually, I can saturate my connection with my local UHD rips. I see a steady 70-80 Mbps (or more) on my router graphs, with peaks exceeding 120 Mbps. It causes buffering often enough to make it difficult to watch. But since the majority of Roku customers aren't using them for local content, just the online sources, 100 Mbps is more than enough. However, I agree that they should offer a flagship product with gigabit Ethernet and lossless audio bitstreaming.... "

Why did the rPi4 move to giga? hmm. could it be future proofing?

Anyways. go ahead and argue amongst yourselves - Thats why I posted in the Lounge. (and I didn't see any flame about POE or ZB3...)

I suppose in 2-3 years IF Matter/Thread/Chip/whatever are a viable option, I bet that Hubitat will explore those options. For all we know, it could be another dead-end protocol. Google seems to kill off a lot of stuff that they were very heavily pushing. And that doesn't take into account the fact that there won't be device availability for this much aggrandized Matter stuff for at least a couple years due to chip shortages.

I think there are going to be a lot of disappointed people when this turns out to just be yet another boring variant of Zigbee that doesn't do anything more than Zigbee or Zwave currently can.

1 Like

Yeah, it was more a tongue in cheek comment from me. The other threads on the subject are probably better for this line of convo.


I realize you are just "poking the bear" so to speak. But I'm curious, what benefit to the Hubitat hub performance would be gained by a faster network connection?

Or you just one of those folks who gets "stars in their eyes" by specifications?



Okay. that I'm guilty of. and I don't like amber lights on my switches. Iwant them all green !! (or blue for the 10G).
I once asked my manager about the difference between hardware and software engineers (I always felt it was a grey between them - like ISO on networks) and he said:
"hardware guys build faster engines.
software guys make it more efficient."

Then he told me a joke - "How many software engineers does it take to change a lightbulb?" A: "None. it's a hardware problem." (I just realized I'm on a software dev board - what was I thinking!)

1 Like

All of these things pale in comparison to the latency introduced by the Z-protocols. Can almost guarantee that users wouldn't see a difference in the performance of their automations if HE had a Gig port.

rPi in general is a different device for different uses. In fact, it doesn't really have a defined use whereas HE does... :apple: to :orange_circle:

I upgraded a couple of my 3B+ to 4 mainly due to that upgrade.

Yeah, I would agree here that a Gbe connection on a media streamer is table stakes at this point.


Maybe this is useful to my argument:

I just don’t see what any of that has to do with Hubitat’s LAN connection. Maybe I’m misunderstanding the hub’s IP network bandwidth needs but I don’t think so.

Sure we can save files to the hub now, but are you planning to use your hub as some kind of file server with multiple users simultaneously writing and reading files to/from the hub? It’s not a NAS device.

And I’m also not sure how Roku is related. That’s a device made for streaming video from the internet that naturally consumes much more bandwidth than what the hub does. A raspberry pi can do all kinds of things that might require more bandwidth too.

Ferraris are really nice cars, but you don’t need one to drive to the closest grocery store or pick the kids up from school.


Wrong. I need one for those things.


I'm afraid I still don't get your argument. Any automation system is going to be as slow as its slowest component. In this case, the slowest components are z-wave (9.6 to 100 kbit/sec) and zigbee (250 kbit/sec). They are slower than a 100 mbit LAN connection by several orders of magnitude!

1 Like

Download the Hubitat app