Insteon acquired and servers coming back up

[lilyoyo1]

3 hours ago, RPerrault said:

um - the past influences the future

no one knows the future - my guess is that the new owners did their due diligence - and it seems they are not new to ha products - might work - might not - hopefully they are successful but if not, they will look back without the regret of not trying

whatever happens, i have no input or control

the brains behind matter would not have chosen wifi if the rumored limitations were true - it might just be they know something we don't - and have overcome the rumored limitations

probably know more about the ecorealm than us

 

I think they chose wifi for it's pros (it's everywhere, no hub required, a variable they can control) not necessarily because it's the best or without issue

[ryarber]

I may be showing my ignorance here, but I don’t believe Thread is built off Wi-Fi in any way. Just another protocol that is taking up 2.4 GHz bandwidth. 

No doubt, efforts to adopt HomeKit were too expensive for the return on investment. Hopefully, Matter will be cheaper/easier to adopt. Apple made their standards too expensive to adopt for the smaller shops. 

[oskrypuch]

IANAE, but here are the physical layers for Matter ...

• Ethernet (LAN, a wired internet connection)
• Wi-Fi (WLAN)
• Thread (a wireless mesh network)
• Bluetooth Low Energy (BLE) for setup only

So, it doesn't define a new framework, it uses those that are now in existence. And Wifi is definitely in. It is an IP over xyz type construct.

It's earlier name was CHIP, connected home over IP.

* Orest

Edited June 22, 2022 by oskrypuch

[jec6613]

7 hours ago, upstatemike said:

4a. A sufficiently powerful processor to handle greatly expanded routing tables and DHCP reservation tables without bogging down.

5. A simple enough configuration UI to avoid being a barrier to non-techie consumers.

6. A low enough price point to avoid being a cost of entry barrier to casual consumers.

This is why I said it may not necessarily be cheap, and specified, "Unusual," requirements:

4a - is normal nowadays.  This is a memory and to a lesser extent storage bound problem, not CPU, and even a $50 router can handle the table load now.  DHCP table size is a complete nonfactor ... what is a factor is how large the table is specified to be by the firmware.

5 - is also normal nowadays.  App based installs make it trivial, though are only on better products.

6 - as I said, it may not be cheap.

[RPerrault]

i miss the point of saying wifi coverage would be a problem - my assumption is anyone putting in ha devices already have good wifi coverage - wifi can be used by a variety of devices - the coverage is probably already there

the same problem exists for proprietary packet delivery devices - you won't be sending the director''s cut of brokeback mountain via insteon - but we accept 'building out the mesh' and 'buy more devices' so that delivery of packets can be sent to ha devices - the world needs more rf - single use proprietary mesh .v. an open standard

i have never had any interference from a microwave or meat bag or anything else

and why is the 2.4 band 'already overcrowded'?  seems to be generally accepted here but i don't see that - residential or commercial - maybe a multifamily dwelling - just never seen it

these packets are tiny for ha devices

people think the packet delivery mechanism and the command set must all be encompassed in matter - it might be in what we call insteon - but those are distinct functions that it looks like matter will separate - guessing again, most already have wifi in place - but not an insteon 'mesh'

wifi supports multicasting - theoretically, one address could be used for every matter device - leave it to matter to determine if the command is for each device - there are potential problems with multicasting but matter could minimize those problems - i was wrong, but i assumed that is how insteon eliminated the 'popcorn' thing

if not multicasting, matter might have another way of overcoming the wifi 'limitations' - chances are, they thought things through before choosing wifi as one of the adopted packet delivery standards in matter

i'd love to see the processing power constraint - i doubt anyone knows the utilization of the components used to deliver wifi

if insteon and z-wave and zig big wee had not bundled packet delivery and its command set, the command set might have a better chance to survive and coexist 

and my favorite - there are tools available to monitor wifi and resolve problems - without the guessing about powerline noise and 'buy more to build out the mesh'

however - i assume using access points - not the faddish wifi mesh stuff

 

[stillwater]

6 minutes ago, RPerrault said:

i miss the point of saying wifi coverage would be a problem

There are a variety of a situations where the Insteon powerline transmission makes things easy that are harder with various RF only systems including wifi,   For example:

1) something in a faraday cage (metal cabinet) that has power inside

2) something at a chicken coop or gate or other installation that has power from the same transformer but not any other need for wifi

Finally with 2.4 GHZ wifi,  no matter how spiffy your own installation, you can still have issues in dense urban areas (apartments or townhouses) owing to channel congestion and poor channel decisions out of the control of a particular user.  

While there may be solutions to these issues with some forethought, with Insteon it's just  a matter of plugging something in.  (Well assuming no interference which can be its own hell).   I am not advocating one vs. the other, just saying that there are reasons why wifi might not be the best in a particular situation.  

 

 

 

[lilyoyo1]

1 hour ago, RPerrault said:

i miss the point of saying wifi coverage would be a problem - my assumption is anyone putting in ha devices already have good wifi coverage - wifi can be used by a variety of devices - the coverage is probably already there

the same problem exists for proprietary packet delivery devices - you won't be sending the director''s cut of brokeback mountain via insteon - but we accept 'building out the mesh' and 'buy more devices' so that delivery of packets can be sent to ha devices - the world needs more rf - single use proprietary mesh .v. an open standard

i have never had any interference from a microwave or meat bag or anything else

and why is the 2.4 band 'already overcrowded'?  seems to be generally accepted here but i don't see that - residential or commercial - maybe a multifamily dwelling - just never seen it

these packets are tiny for ha devices

people think the packet delivery mechanism and the command set must all be encompassed in matter - it might be in what we call insteon - but those are distinct functions that it looks like matter will separate - guessing again, most already have wifi in place - but not an insteon 'mesh'

wifi supports multicasting - theoretically, one address could be used for every matter device - leave it to matter to determine if the command is for each device - there are potential problems with multicasting but matter could minimize those problems - i was wrong, but i assumed that is how insteon eliminated the 'popcorn' thing

if not multicasting, matter might have another way of overcoming the wifi 'limitations' - chances are, they thought things through before choosing wifi as one of the adopted packet delivery standards in matter

i'd love to see the processing power constraint - i doubt anyone knows the utilization of the components used to deliver wifi

if insteon and z-wave and zig big wee had not bundled packet delivery and its command set, the command set might have a better chance to survive and coexist 

and my favorite - there are tools available to monitor wifi and resolve problems - without the guessing about powerline noise and 'buy more to build out the mesh'

however - i assume using access points - not the faddish wifi mesh stuff

 

The thing about looking at your singular experience is that it assumes that millions of others are in your position. In my neighborhood, our houses are far enough away that no one interferes with each other. Go uptown and do a wifi scan, you'll run out of room on your wifi list (not literally).

High density cities such as NYC and LA can run into wifi issues and people often do. It also doesn't take into account that most do not have enough networking experience or knowledge of tools that can help troubleshoot. 

In addition most still use the cable company provided routers. If they add repeaters, it'll be the generic mesh systems that you knock...

We all still make alot of assumptions about a something that does not yet exist. It's easy for us to say so they need to do is XYZ. However, if it were, it would've already been done. 

[upstatemike]

2 hours ago, RPerrault said:

.

however - i assume using access points - not the faddish wifi mesh stuff

 

So I am trying to picture a typical 2.4GHz WiFi installation. There are 3 non-overlapping channels. Chances are somebody with a large Home Automation system might also have a Sonos system which uses one of those channels. So you have two channels and a bunch of Access Points. Each Access Point should be on a different channel than its nearest neighbors for clean roaming and minimal interferance but after AP1 and AP2 you find you can't put AP3 where you need coverage without it being close enough to interfere with either AP1, AP2, or Sonos and there is no 4th channel you can go to.

It doesn't seem like you have to be in NYC or LA to run into interferance issues.

[lilyoyo1]

5 hours ago, upstatemike said:

So I am trying to picture a typical 2.4GHz WiFi installation. There are 3 non-overlapping channels. Chances are somebody with a large Home Automation system might also have a Sonos system which uses one of those channels. So you have two channels and a bunch of Access Points. Each Access Point should be on a different channel than its nearest neighbors for clean roaming and minimal interferance but after AP1 and AP2 you find you can't put AP3 where you need coverage without it being close enough to interfere with either AP1, AP2, or Sonos and there is no 4th channel you can go to.

It doesn't seem like you have to be in NYC or LA to run into interferance issues.

You don't have to be in either of those places but it was just an example that many in the US lives in areas where you can find hundreds of people in a small area.

I've had neighbors ask for help with their own issues (all were using cable provided routers). A few added a couple of ring cameras only to watch their network performance take a hit

[stillwater]

9 hours ago, upstatemike said:

So I am trying to picture a typical 2.4GHz WiFi installation. There are 3 non-overlapping channels. Chances are somebody with a large Home Automation system might also have a Sonos system which uses one of those channels. So you have two channels and a bunch of Access Points. Each Access Point should be on a different channel than its nearest neighbors for clean roaming and minimal interferance but after AP1 and AP2 you find you can't put AP3 where you need coverage without it being close enough to interfere with either AP1, AP2, or Sonos and there is no 4th channel you can go to.

It doesn't seem like you have to be in NYC or LA to run into interferance issues.

Good point.  This is one reason my Sonos system operates solely over ethernet cable.  But that isn't easy for every installation.

[RPerrault]

22 hours ago, lilyoyo1 said:

The thing about looking at your singular experience is that it assumes that millions of others are in your position. In my neighborhood, our houses are far enough away that no one interferes with each other. Go uptown and do a wifi scan, you'll run out of room on your wifi list (not literally).

High density cities such as NYC and LA can run into wifi issues and people often do. It also doesn't take into account that most do not have enough networking experience or knowledge of tools that can help troubleshoot. 

In addition most still use the cable company provided routers. If they add repeaters, it'll be the generic mesh systems that you knock...

We all still make alot of assumptions about a something that does not yet exist. It's easy for us to say so they need to do is XYZ. However, if it were, it would've already been done. 

somebody hold my beer

why would the room on my wifi 'list' matter?

tell me about my singular experience

i mentioned the possibility of complexities in high density housing

troubleshooting tools ARE available for wifi - how'd those tools work for you with insteon?

learn what mesh wifi is - there is a reason i prefer access points

its much better to spend the money building out a single purpose mesh line insteon instead of wifi that does a little more than turn a light off - ok

where are people that want 2 zwave devices located on either end of their house - so your advice is to ignore the wifi and build out an entirely new single function mesh - ok

if you don't care to participate in the assumption discussion - feel free not to comment

but you can't  

[RPerrault]

21 hours ago, upstatemike said:

So I am trying to picture a typical 2.4GHz WiFi installation. There are 3 non-overlapping channels. Chances are somebody with a large Home Automation system might also have a Sonos system which uses one of those channels. So you have two channels and a bunch of Access Points. Each Access Point should be on a different channel than its nearest neighbors for clean roaming and minimal interferance but after AP1 and AP2 you find you can't put AP3 where you need coverage without it being close enough to interfere with either AP1, AP2, or Sonos and there is no 4th channel you can go to.

It doesn't seem like you have to be in NYC or LA to run into interferance issues.

does sonos get assigned an entire channel that is exclusive to its use?  did not know that is allowed in wifi

i have 27 sonos devices - never had a problem

each access point uses 3 channels - not 3 channels per ssid - if the coverage overlaps, the access points will change channels - if you allow them to

streaming sound is not taxing - streaming high def video can be

 

[RPerrault]

12 hours ago, stillwater said:

Good point.  This is one reason my Sonos system operates solely over ethernet cable.  But that isn't easy for every installation.

sonos has a 'boost' device that takes your devices off wifi - i suppose for those without wifi

 

 

30 minutes ago, RPerrault said:

sonos has a 'boost' device that takes your devices off wifi - i suppose for those without wifi

 

but yeah - wired is preferable where possible

 

[lilyoyo1]

1 hour ago, RPerrault said:

somebody hold my beer

why would the room on my wifi 'list' matter?

tell me about my singular experience

i mentioned the possibility of complexities in high density housing

troubleshooting tools ARE available for wifi - how'd those tools work for you with insteon?

learn what mesh wifi is - there is a reason i prefer access points

its much better to spend the money building out a single purpose mesh line insteon instead of wifi that does a little more than turn a light off - ok

where are people that want 2 zwave devices located on either end of their house - so your advice is to ignore the wifi and build out an entirely new single function mesh - ok

if you don't care to participate in the assumption discussion - feel free not to comment

but you can't  

You probably should put your beer down so you can come up with coherent sentences and/or remember what you typed a few hours ago.

The first letter you type was I.

Second line started with "my assumption"

3rd paragraph once again started with "I have never had"

4th paragraph- "I have never seen that"- though you did leave a slight opening for MDUs. 

Next to last paragraph- "and my favorite"

Last paragraph- "I assume"

So yes, with the amount of I's and me's being dropped in your explanation, please forgive me for thinking you were referring to your own bubble. 

I guess you only read my post just to argue back which is fine. Maybe you should learn to read for comprehension. Had you done so, you would have seen that I wasn't disagreeing with you but instead pointing out that the problem with different wifi setups and tools is assuming most people have the knowledge about them in addition to the skill set to use them properly. Guess what- Most do not. What you prefer and another uses are 2 different things. A person with knowledge, desire, and skills can use anything that's available to accomplish any given task. Someone else will probably take a simpler approach whether it's a good solution or not. The world is full of the latter. 

Just because you've read something and only have dealt with it in your personal life does not make you an expert. I've been doing homes for years and still learn new stuff unlike you who think they know it all. Ive never had a use for tools to troubleshoot insteon. I took the time to learn how it works through actual use and not just in my home. I experimented in family and friends houses. Putting stuff in/out and using different combinations.  Ditto with C4, zwave, and now Lutron. 

Up until i stopped installing it, I made sure i installed it in such a way to give me a positive outcome. I still do with other systems as well. 

I have no issue in having discussions about assumptions. However, I don't lose sight about what it in front of us now instead of believing some magical pipe dream of what i wish something could be. My home network is strong enough where i could run everything on wifi but i don't. Anything that can be hardwired is still hardwired. People not wanting to use a bunch of wifi devices in their home (or others) is their perogative.

If you like wifi devices and have no issues so be it. Others may not want to take the time and expense of filling their home with them only to create new issues that they didn't have before with other stuff. At least with insteon, zwave, or any other mesh type system, issues with those do not affect anything else in the house that others may need (for the most part).

[ase]

9 hours ago, RPerrault said:

somebody hold my beer

why would the room on my wifi 'list' matter?

tell me about my singular experience

i mentioned the possibility of complexities in high density housing

troubleshooting tools ARE available for wifi - how'd those tools work for you with insteon?

learn what mesh wifi is - there is a reason i prefer access points

its much better to spend the money building out a single purpose mesh line insteon instead of wifi that does a little more than turn a light off - ok

where are people that want 2 zwave devices located on either end of their house - so your advice is to ignore the wifi and build out an entirely new single function mesh - ok

if you don't care to participate in the assumption discussion - feel free not to comment

but you can't  

Well here are a few things to think about.

The reason the number of aps showing on your list matters, is each one is an ap taking a chunk out of the available spectrum.

The wifi alliance has basically said that the USA wifi spectrum is substantially congested. While there is momentum to open 6ghz up to wifi, that's going to be a bit yet.

Next issue is most consumer grade hardware is maxed at 253 devices. Even if you change the IP class the router most likely has a routing table that can only handle the 253 devices.

Next issue is most switches and routers have roughly 1/4 to 1/2 switching capacity as there max port throughput. This means that the device cannot maxout any interface. This also means small data being continuously exchanged weights down faster larger data streams. So having a light switch beaconing updates could cause performance issues.

Both 4G and 5G had sub protocols and dedicated memory in devices for IOT for a reason. WIFI doesn't have such setups, although it is being discussed for WIFI 7, we aren't there yet. 

Next issue is that TCP/IP is a routed protocol. That being the case each piece of equipment must have some level routing table. That means the only method of broadcast is via the subnet broadcast address. This isn't exactly reliable across WIFI. The alternative is UDP, king of speed and failed packet delivery. 

As for WIFI "tools" there is only one worth talking about. Spectrum Analyzer and one thay will give you the whole picture costs about 1800 bucks. On the Insteon side any PLM can be paired with a computer and see where the potential issues are. As for cost, zip, the tools are free open source.

Can a WIFI IOT/Home Automation network be built that works well? Yes of course. However, it almost requires a dedicated wifi network of its own. Furthermore way too many of the factors are out of the owners control, you can buy the best of everything and have it all fail because of something outside of your control.

The creators of the many Home Automation protocols are not morons, they understood the issues with just trying to strap a wifi chip onto a device.

Insteon was/is considered superior to most because it built on the bedrock of X10. Each device did a little bit of the heavy lifting, it was a true and complete mesh. Devices didn't require a controller or a central hub.

Edited June 24, 2022 by ase

[larryllix]

WiFi 6 routers and usage has been in the user market for about 5-6 years now. It enhances the 2.4GHz and 5GHz band usage with advanced arbitration schemes. WiFi 6 s not defined as a frequency band addition but rather a protocol improvement for new and existing bands. My router will connect to many devices at just under 1 Gbps on the existing 2.4GHz channels using the 802.11ax (WiFi6) protocol. My VR headset connects at 1.2Gbps on any band using WiFi6. Faster devices, using WiFi 6 are already being sold,for a few years now.

 

WiFi 6E is being developed and available in higher end routers, using higher frequency bands. Some of those frequencies have been included on many routers already sold for the last 3-4 years (or more).

https://en.wikipedia.org/wiki/Wi-Fi_6

 

Edited June 24, 2022 by larryllix

[ase]

29 minutes ago, larryllix said:

WiFi 6 routers and usage has been in the user market for about 5-6 years now. It enhances the 2.4GHz and 5GHz band usage with advanced arbitration schemes. WiFi 6 s not defined as a frequency band addition but rather a protocol improvement for new and existing bands. My router will connect to many devices at just under 1 Gbps on the existing 2.4GHz channels using the 802.11ax (WiFi6) protocol. My VR headset connects at 1.2Gbps on any band using WiFi6. Faster devices, using WiFi 6 are already being sold,for a few years now.

 

WiFi 6E is being developed and available in higher end routers, using higher frequency bands. Some of those frequencies have been included on many routers already sold for the last 3-4 years (or more).

https://en.wikipedia.org/wiki/Wi-Fi_6

 

The fcc has not given the green light on 6e yet and 6ghz in the US has lots of strings attached. 

https://www.fcc.gov/document/fcc-opens-6-ghz-band-wi-fi-and-other-unlicensed-uses

 

Most consumer grade equipment hasn't yet added 6ghz in the US. Even if they added it it will reach less distance than 5ghz and 2.4ghz. And none of that does any good without upgrades in memory and CPU to speed up switch capacity. 

None of it makes wifi for iot any more reliable. For that we need a sub-protocol and dedicated memory for iot routes. Which is why it is in draft for wifi7.

[larryllix]

25 minutes ago, ase said:

The fcc has not given the green light on 6e yet and 6ghz in the US has lots of strings attached. 

https://www.fcc.gov/document/fcc-opens-6-ghz-band-wi-fi-and-other-unlicensed-uses

 

Most consumer grade equipment hasn't yet added 6ghz in the US. Even if they added it it will reach less distance than 5ghz and 2.4ghz. And none of that does any good without upgrades in memory and CPU to speed up switch capacity. 

None of it makes wifi for iot any more reliable. For that we need a sub-protocol and dedicated memory for iot routes. Which is why it is in draft for wifi7.

Out of date information.

https://www.fcc.gov/document/fcc-opens-6-ghz-band-wi-fi-and-other-unlicensed-uses-0

Most router companies released 6GHz routers a few years ago. I have been using two ASUS 6GHz band routers since about 2019? Perhaps the FCC was delayed from the Canadian DoC rules there? That never previously happened. Usually Canada followed the US rules.

My 6GHz band reaches much further than the 5GHz band. However I believe my ASUS routers have destroyed the 5GHz band signals for some reason, making that band almost useless for more than the immediate room. The 6GHz band reaches any corner of my 200 x 200' property much better for some reason.

The 802.11ax protocol (WiFi6) does increase reliability of the data. Not exactly sure how the new arbitration works but it sounds similar the the technique cell phone systems use. with time sliced frequency sharing and shorter packet pre/postambles.  Trouble with all that is my 40 some odd WiFi HA devices only understand the old protocols so 2.4GHz is mostly dedicated to that. However, brag about the 802.11ax protocol states it handles non-participants better and I believe it after using it with 80-90 devices.

 

[ase]

17 hours ago, larryllix said:

Out of date information.

https://www.fcc.gov/document/fcc-opens-6-ghz-band-wi-fi-and-other-unlicensed-uses-0

Most router companies released 6GHz routers a few years ago. I have been using two ASUS 6GHz band routers since about 2019? Perhaps the FCC was delayed from the Canadian DoC rules there? That never previously happened. Usually Canada followed the US rules.

My 6GHz band reaches much further than the 5GHz band. However I believe my ASUS routers have destroyed the 5GHz band signals for some reason, making that band almost useless for more than the immediate room. The 6GHz band reaches any corner of my 200 x 200' property much better for some reason.

The 802.11ax protocol (WiFi6) does increase reliability of the data. Not exactly sure how the new arbitration works but it sounds similar the the technique cell phone systems use. with time sliced frequency sharing and shorter packet pre/postambles.  Trouble with all that is my 40 some odd WiFi HA devices only understand the old protocols so 2.4GHz is mostly dedicated to that. However, brag about the 802.11ax protocol states it handles non-participants better and I believe it after using it with 80-90 devices.

 

You're getting things mixed up. Rule one higher frequency always has lower propagation. The higher the frequency the more difficulty getting it through walls. Look at satellites, we can send very high frequencies from earth to space and back. Put the dish indoors and see what happens. 

The fcc hasn't signed off on 6e yet, it is still in comments. 6 ghz is not fully available on any consumer grade hardware. TP-link is the manufacturer that has deployed the most 6ghz and they max at 80 per channel. 6e should extend that to aggregated 1200 mhz. Either way it has to exist in co-exist mode(partial channels). We still have other things in that band.

This creates the complex issue of maximum time a device can hold the channel. 

I think you are confused as to the difference between draft and full approval. The fcc often approves use during the draft period. 

You started seeing 5g in 2019 even though the draft wasn't closed until 2021. 

The next issue is packet size. Both wifi and tcp itself has standard packet sizes. This makes them very efficient for larger data but very inefficient for smaller data.

Think of it like you needed a button battery and you ordered it from Amazon. Instead of an envelope, they send it in a 30x30x30 box and they pack it full of air packs so that the box doesn't collapse. This is kinda what is happening when you send small bits of data along tcp. There are several levels involved in sending that data that just isn't needed for control of a simple device.

Meanwhile this effects the capacity of the device at a switching level.

Example you can max out a 5 port gigabit consumer grade switch with just a couple of security cameras. 1Mbps continued will max out the switching capacity of that device. 

Now use a professional device and you couldn't replicate it. 

Most people just use whatever their ISP gives them. A low capacity consumer grade wifi ap gateway.

Go talk to the claims department of your local Walmart, routers are amongst the top disposition items. They get sent back nearly as fast as they come in. For all kinds of reasons. 

Wifi and consumer grade networking just isn't the best method for IOT/home automation. 

Like I said, if installed properly, routed properly with proper professional equipment in an uncrowded environment, sure it will work great.

Unfortunately you can't guarantee those conditions.

[RPerrault]

13 hours ago, ase said:

You're getting things mixed up. Rule one higher frequency always has lower propagation. The higher the frequency the more difficulty getting it through walls. Look at satellites, we can send very high frequencies from earth to space and back. Put the dish indoors and see what happens. 

The fcc hasn't signed off on 6e yet, it is still in comments. 6 ghz is not fully available on any consumer grade hardware. TP-link is the manufacturer that has deployed the most 6ghz and they max at 80 per channel. 6e should extend that to aggregated 1200 mhz. Either way it has to exist in co-exist mode(partial channels). We still have other things in that band.

This creates the complex issue of maximum time a device can hold the channel. 

I think you are confused as to the difference between draft and full approval. The fcc often approves use during the draft period. 

You started seeing 5g in 2019 even though the draft wasn't closed until 2021. 

The next issue is packet size. Both wifi and tcp itself has standard packet sizes. This makes them very efficient for larger data but very inefficient for smaller data.

Think of it like you needed a button battery and you ordered it from Amazon. Instead of an envelope, they send it in a 30x30x30 box and they pack it full of air packs so that the box doesn't collapse. This is kinda what is happening when you send small bits of data along tcp. There are several levels involved in sending that data that just isn't needed for control of a simple device.

Meanwhile this effects the capacity of the device at a switching level.

Example you can max out a 5 port gigabit consumer grade switch with just a couple of security cameras. 1Mbps continued will max out the switching capacity of that device. 

Now use a professional device and you couldn't replicate it. 

Most people just use whatever their ISP gives them. A low capacity consumer grade wifi ap gateway.

Go talk to the claims department of your local Walmart, routers are amongst the top disposition items. They get sent back nearly as fast as they come in. For all kinds of reasons. 

Wifi and consumer grade networking just isn't the best method for IOT/home automation. 

Like I said, if installed properly, routed properly with proper professional equipment in an uncrowded environment, sure it will work great.

Unfortunately you can't guarantee those conditions.

thanks for taking the time to post - someone that understands wifi and networking and took the time to participate

i had no clue how large the transport frame was for wifi - great point 

i also recall reading about the antenna sizes needed for the different frequencies - which could present a challenge

i have always been a hard wired bigot - i think its almost abusive when consumers are told they can stream hi-def or 4k video - i guess it works - i'd only do it as an ad hoc thing - not for 4 tvs 

this is my reasoning when talking about network equipment limitations - someone has to tell the consumer - so much uses wifi and we are constantly adding to it - people and wire cutting and streaming the video content as an alternative to cable and satellite - if you want to use the router provided by your isp, you can't do it all - the money you invest in a router, network switches and access points (and controller) is better spent than in a single use new frequency - and if you don't have an equipment area, you soon will - not sure ha would be the straw that breaks the camel's back but you know this stuff better and can foresee the problems - which was the point of my discussions

On 6/24/2022 at 5:16 AM, ase said:

Next issue is that TCP/IP is a routed protocol. That being the case each piece of equipment must have some level routing table. That means the only method of broadcast is via the subnet broadcast address. This isn't exactly reliable across WIFI.

ok getting lost here - what devices would need routing capability?

i think uds gets a bad rap - i tell people 'it offends my sense of completeness' but for the right application it works remarkably well - voip and such - another question - does streaming video use udp?  if so, it would explain the pixilating 

On 6/24/2022 at 5:16 AM, ase said:

On the Insteon side any PLM can be paired with a computer and see where the potential issues are.

didn't know that

the only wifi troubleshooting i have much experience with was with cisco's level 2 - when their 8821 wifi phones were new - had about 150 or more and got to fight the birthing pains of testing new firmware - as best i recall, i placed one in a diagnostic mode and located the phone close to an ap - not sure what all they used but wireshark was part of the process - but i was cisco end-to-end - all new equipment wired and wifi - added the be6kt hat was gifted from corporate 

On 6/24/2022 at 5:16 AM, ase said:

The creators of the many Home Automation protocols are not morons, they understood the issues with just trying to strap a wifi chip onto a device.

i know they are not morons - which is why i was asking why they did not choose wifi as their packet transport - again - thanks for taking the time to educate me

time will tell - but the people behind matter are not morons either - and some pretty smart people with lots of money and market influence have bought in - (obviously no guarantee of success)

[ase]

7 hours ago, RPerrault said:

thanks for taking the time to post - someone that understands wifi and networking and took the time to participate

i had no clue how large the transport frame was for wifi - great point 

i also recall reading about the antenna sizes needed for the different frequencies - which could present a challenge

i have always been a hard wired bigot - i think its almost abusive when consumers are told they can stream hi-def or 4k video - i guess it works - i'd only do it as an ad hoc thing - not for 4 tvs 

this is my reasoning when talking about network equipment limitations - someone has to tell the consumer - so much uses wifi and we are constantly adding to it - people and wire cutting and streaming the video content as an alternative to cable and satellite - if you want to use the router provided by your isp, you can't do it all - the money you invest in a router, network switches and access points (and controller) is better spent than in a single use new frequency - and if you don't have an equipment area, you soon will - not sure ha would be the straw that breaks the camel's back but you know this stuff better and can foresee the problems - which was the point of my discussions

ok getting lost here - what devices would need routing capability?

i think uds gets a bad rap - i tell people 'it offends my sense of completeness' but for the right application it works remarkably well - voip and such - another question - does streaming video use udp?  if so, it would explain the pixilating 

didn't know that

the only wifi troubleshooting i have much experience with was with cisco's level 2 - when their 8821 wifi phones were new - had about 150 or more and got to fight the birthing pains of testing new firmware - as best i recall, i placed one in a diagnostic mode and located the phone close to an ap - not sure what all they used but wireshark was part of the process - but i was cisco end-to-end - all new equipment wired and wifi - added the be6kt hat was gifted from corporate 

i know they are not morons - which is why i was asking why they did not choose wifi as their packet transport - again - thanks for taking the time to educate me

time will tell - but the people behind matter are not morons either - and some pretty smart people with lots of money and market influence have bought in - (obviously no guarantee of success)

I agree wired is always better. TCP/IP has 4 layers. However, most network engineers use the OSI model for education which is 7 layers, so I will do the same.

1. Physical layer.

2. Data layer.

3. Network layer.

4. Transport layer.

5. Session layer.

6. Presentation layer.

7. Application layer.

The only device that doesn't have "routing tables" is hubs. Hubs do not process any switching, they simply forward all packets to all ports. Hubs are known as a layer 1 device.

Switchs in the strict sense are layer 2 devices and hold a table of MAC addresses. This is how a switch knows which ports need to talk to each other. 

Routers and "smart switches" are layer 3. They are both aware of IP addresses and MAC addresses.

So you see unless you are using a hub(which would be a huge cluster f) each device is at the very least routing MAC addresses.

Now even at the top of the mark consumer grade wireless routers are often quadcore arm processors or less with very small memory footprints. Think raspberry pi 3 or lower for processing. 

I have seen some "top" consumer grade routers that can only handle 120 entries. Nowadays doesn't take long to get there.

The biggest issue with attempting to use UDP for anything really, is lack of data verification. UDP doesn't have any error check, so if data is corrupt, too bad. 

Now maybe wifi maybe the way of the future, but I am not going to hold my breath. More and more people are g upgrading to gigabit connections, in my neck of the woods gigabit is the second slowest connection one can buy, and maybe the slowest connection available within 2 years. The additional spectrum is more likely to be saturated quickly as more people move towards 10gigbit connections.

Intel has proposed IOT in wifi 7, the question remains if consumer grade routers will upgrade CPUs and Memory enough to really see an advantage.

It all can work smoothly enough right now with professional grade equipment. However,we have to accept that the reality is a home diyer buying a wifi swich off shelf and home depot and is at best running a linksys router they bought at Walmart.

IMHO wifi is never going to be the best route for Home Automation. IP and network aware, yes. People will just keep saturating that bandwidth with video and games.

[ase]

As for troubleshooting Insteon. I have always found it to be very easy. If you are using Insteon Hub, Insteon itself runs in a single process and the suite they built around it has several tools to help troubleshoot issues. Simply ssh session run some tools and bingo you know what the problem is.

As for with a PLM or USB stick you can simply plug it into a computer and use your choice of several dozen insteon command line tools. Several have options to find bad links, issue commands and so on. 

Very early on Insteon was very open and willing to work with other. I think the closing down of that communication and openness is what put the final nail in the coffin.

Had they been open with us(the Insteon community) about what was going on and how component shortages and free acess to hub had been draining the money, I am positive they could've raised capital. Hell I would have invested.

Either way the earlier openness spawned a number of great tools to troubleshoot issues.

[Brian H]

I found the same things in my Insteon use.

I was an early Developers Group member and paid for the documentation with an NDA agreement. One day the group just went away (I did not see any warning messages) and the documentation,  I had an NDA on just got released to public domain.

Then we have the lack of information on the Alert and Siren modules. Causing UDI to reverse engineer the modules features and setting  to the best of their abilities.

Edited June 26, 2022 by Brian H

[ryarber]

If I may pose a question… Admittedly, I am speaking with people who know much more about this than I. I understand what ase is saying even though I don’t have the depth of knowledge of routing, switches, layers, etc….

But it seems to me the way things will shake out is that, while Wi-Fi will be useful in IoT, it will not be the predominant mode of information delivery/control. I see Wi-Fi as a means to provide hubs and other bandwidth hungry IoT devices an interface with our smarter devices. And then thread as the primary interface with most IoT devices that don’t need the large packets and huge bandwidth. That’s been my understanding from the start. That while Wi-Fi is a necessary element, it will be a relatively minor player going forward. And that where possible, hardwired connections will still be preferable to wireless. 

I think I am understanding people here saying that each light switch, LED bulb, door lock, motion sensor, etc. would have independent wifi capability and I do not believe that is where we are headed. Ideally, I believe these “dumber” devices will have their own communication protocols while their controllers will act as a bridge to “smarter” devices and the internet. 

[lilyoyo1]

5 hours ago, ryarber said:

If I may pose a question… Admittedly, I am speaking with people who know much more about this than I. I understand what ase is saying even though I don’t have the depth of knowledge of routing, switches, layers, etc….

But it seems to me the way things will shake out is that, while Wi-Fi will be useful in IoT, it will not be the predominant mode of information delivery/control. I see Wi-Fi as a means to provide hubs and other bandwidth hungry IoT devices an interface with our smarter devices. And then thread as the primary interface with most IoT devices that don’t need the large packets and huge bandwidth. That’s been my understanding from the start. That while Wi-Fi is a necessary element, it will be a relatively minor player going forward. And that where possible, hardwired connections will still be preferable to wireless. 

I think I am understanding people here saying that each light switch, LED bulb, door lock, motion sensor, etc. would have independent wifi capability and I do not believe that is where we are headed. Ideally, I believe these “dumber” devices will have their own communication protocols while their controllers will act as a bridge to “smarter” devices and the internet. 

When it comes to current diy "protocols", there's insteon (whenever they start releasing new devices), zwave, ZigBee, Lutron clear connect, wifi, and Bluetooth. All with their own pros and cons. What will exist in the future, we do not know. 

Only wifi  does not require a physical bridge in order to work. (Technically insteon doesn't but you wouldn't have timers or remote operation). Instead, it requires an app on your phone to set up and configure. 

Each device/mfg. can still use their own chosen protocol (insteon, ZigBee, zwave, clear connect) since those with hubs will still be connected to your network as well. 

Wifi devices would still be connected independently to your wifi network as is now. However, if it's a Matter connected device, thread would be able to connect it and other devices on your network under a common language. 

I've linked an article that explains things better than I can. In short, wifi will be there and used. Thread is just an additional.layer tying things together

Edited June 26, 2022 by lilyoyo1