123Next
Back New
Author: Greg Hook

TP-LINK Deco M9 Mesh Wi-Fi System Review & Comments

[Copy link]

11610K

Threads

12810K

Posts

37310K

Credits

Administrators

Rank: 9Rank: 9Rank: 9

Credits
3732793
 Author| 2-12-2019 04:53:38 Mobile | Show all posts
Thanks for the post and the suggestions.
I will look into Ekahau and see if they have an option for reviewers.
Reply Support Not support

Use magic Report

11610K

Threads

12810K

Posts

37310K

Credits

Administrators

Rank: 9Rank: 9Rank: 9

Credits
3732793
2-12-2019 04:53:38 Mobile | Show all posts
Maybe it's just me

If no one else cares about that sort of info... I am happy to admit I am wrong about this sort of review on this site.
Reply Support Not support

Use magic Report

11610K

Threads

12810K

Posts

37310K

Credits

Administrators

Rank: 9Rank: 9Rank: 9

Credits
3732793
2-12-2019 04:53:39 Mobile | Show all posts
It's useful to have this sort of comment and information
Reply Support Not support

Use magic Report

11610K

Threads

12810K

Posts

37310K

Credits

Administrators

Rank: 9Rank: 9Rank: 9

Credits
3732793
2-12-2019 04:53:40 Mobile | Show all posts
Always worth having a pre-determined set of measurements for product comparisons in reviews, especially if the same house and locations within it are used.
Reply Support Not support

Use magic Report

11610K

Threads

12810K

Posts

37310K

Credits

Administrators

Rank: 9Rank: 9Rank: 9

Credits
3732793
 Author| 2-12-2019 04:53:40 Mobile | Show all posts
Yes, that's what I have at the moment. Always happy to add additional tests if they are free of course!
Reply Support Not support

Use magic Report

11610K

Threads

12810K

Posts

37310K

Credits

Administrators

Rank: 9Rank: 9Rank: 9

Credits
3732793
2-12-2019 04:53:41 Mobile | Show all posts
I've been using four of these M9 Deco units for about 2 months now, and for the most part I'm happy with them - some things I love, other things I find hugely irritating .

I have a bit of an irregular setup. The unit by the modem (at the front door) and the unit farthest away (in the office at the opposite end of the flat) are connected over 1Gbps ethernet for the backhaul, and the other two units are purely wireless - typically anywhere in the flat I'll get 300-400Mbps with no noticeable dead spots. Just doing a quick test transferring a large file from the PC in the living room to my office Macbook gave me a solid 520Mbps for most of the transfer. So speed-wise, no complaints. (Although my flat is all on one floor without too many thick walls, which no doubt helps.)

As for gripes, I have quite a few although no huge deal breakers. I'm a home automation nut, and personally find all the automation support in the Deco completely useless. As you'd probably expect, you have no control over any of the router's automation functionality except through what the app gives you access to - and that boils down to limited support for a small selection of devices, with a very narrow set of tools to construct behaviours based on whatever devices you've managed to add. (I'm sitting here with a pile of various zigbee sensors and I haven't been able to get a single one working correctly with the Deco app.) With the right kit, it might be straightforward to do some simple stuff like making a light switch on when you return home - but I can imagine anything more advanced is likely to be a struggle. It's just frustrating when you know all the tech is there but there's nothing useful you can do with it. I use Home Assistant for my automation setup, and that brought another issue to light - IP address reservation. The M9 only supports 16 reserved IP addresses (which seems like an absurdly needless limitation), which is a problem when a lot of my automation devices require a static IP to function correctly. (The Hue hub, Google Mini speakers, etc.)

That brings me onto another annoyance - IFTTT support. Given that this router is really billed as a home automation platform, to me that kind of suggests that it's going to open up all your devices to endless possibilities with IFTTT to create routines to do anything you can imagine with lights, sensors, and IoT gadgets around the house. Unless I'm doing something wrong or have massively missed the point, that isn't the case at all - you can create triggers based on devices joining the network, leaving the network, or when a new device connects for the first time. And that's it. My mind isn't exactly fizzing with ideas for exciting functionality that I could build off the back of that, but maybe I expected too much.

A few other miscellaneous notes:
- The USB ports, while unused, are powered on - handy for powering my Raspberry Pi.
- I've never had an issue with intermittent disconnection or rebooting issues, but I have noticed that sometimes when I power off a unit at the mains, I have to wait a good five minutes before I plug it back in otherwise it appears dead. No idea what the reason could be for this, but all my units have this problem. (Although not an issue unless I'm constantly unplugging and moving them around.)
- I only briefly looked at the Alexa support, but it seems pretty worthless to me (and I'd imagine to most people). If I recall correctly it's mostly things like toggling the priority mode for different services/devices, toggling guest WiFi, and things like that.
Reply Support Not support

Use magic Report

11610K

Threads

12810K

Posts

37310K

Credits

Administrators

Rank: 9Rank: 9Rank: 9

Credits
3732793
2-12-2019 04:53:42 Mobile | Show all posts
But the data has to be meaningful and provable. A speed test means very little without the other relevant data to back it up.

Most corps with medium to large Wifi deployments will not be running at 80 or 40MHz they will set the APs to 20MHz to provide a larger number of channels (providing its a microcell architecture) to limit co-channel when you compare the speed test of an SSID in 20MHz to 80MHz you could then argue the speed of this £600 AP is rubbish compared to a home AP. however, you are not comparing apples to apples.

Lots of home AP's cannot be changed (can the TP-Link in the review?)

You could argue that there is a big case to make all wifi 20MHz if you have neighbours... but this goes against the selling of speed with wifi...and the reviewing
Reply Support Not support

Use magic Report

11610K

Threads

12810K

Posts

37310K

Credits

Administrators

Rank: 9Rank: 9Rank: 9

Credits
3732793
2-12-2019 04:53:42 Mobile | Show all posts
@JJW how about using the Pi with PiHole and using the DHCP server on that to control static IP addresses
Reply Support Not support

Use magic Report

11610K

Threads

12810K

Posts

37310K

Credits

Administrators

Rank: 9Rank: 9Rank: 9

Credits
3732793
2-12-2019 04:53:43 Mobile | Show all posts
Yeah, if I hit the limit I'll use another DHCP server, it's just an extra step of hassle that I'd rather avoid. Currently using 8 out of the 16 available IPs so hopefully I'll be ok.
Reply Support Not support

Use magic Report

11610K

Threads

12810K

Posts

37310K

Credits

Administrators

Rank: 9Rank: 9Rank: 9

Credits
3732793
2-12-2019 04:53:44 Mobile | Show all posts
Not to mention we tune the transmit power down - yes down - to create smaller cells and encourage clients to roam more readily and reduce the co-channel interference (meaning we can cram in more cells with fewer devices in each and thereby lessen contention which can actually increase the throughput ("speed") of any given device. It's complicated.)

I've opined in the past, and suffered the flames for so doing, that unless you have a proper RF shielded test facility with calibrated test equipment (not to mention break open the devices so you can see the antenna and orient them correctly for testing,) one cannot give any kind of objective opinion on the radio signals from any device. Such facilities are expensive, which is why no-one - including the trade - does it. (I weep internally every time I pick up one of the trades and find "this month we've group tested the latest batch of Wi-Fi AP's - Dave took five leading models home and had a play with them last weekend." At about that point I stop reading.)

Lay people (and I'm sad to say far too many IT professionals) do tend to obsess over "Wi-Fi signal" (ironic, since there's actually no such thing,) more specifically "Wi-Fi Signal FROM the router/AP" as if it is the only thing that matters. Of course, the marketing hype feeds that narrative (though one suspects the ad. copy writers don't know what they are talking about.)

I'll drop in my usual boiler plate that, Wi-Fi transmit power is limited by law and most kit is, and always has been, at of very close to the permitted maximum. What differences there are, are not worth worrying about. (Unless maybe you are a radio ham or electrical engineer.) If you can get hold of a datasheet, I'll bet most of them cite Tx power of 20/23 dBm - ie the legal max. I suspect most professionals network managers buying AP's spend all of 10 seconds worrying about Tx power as we know it's all much of a muchness and every vendor claims their's are better everyone else's. By definition, they cannot all be right.

In any case, (more of my boiler plate) Wi-Fi is a two-way radio "conversation" like walkie-talkies, not a one way "lecture" like television. It's like sound, for two communicating peers to be able to converse, they both need to be able to hear each other. If they cannot, and they are already as "loud" as permitted, then all you can do is move closer together, which generally means putting up more AP's closer to where the clients are. It's not for the fun of it that on big site we put up hundreds.
Reply Support Not support

Use magic Report

123Next
Back New
You have to log in before you can reply Login | register

Points Rules

返回顶部