External antennas on the likes of AC86U, is that the magic for range?

[L&LD]

In your post linked above, in the paragraph, "Executive summary: As expected, the newer RT-AX68U's excelled at throughput vs. the older design RT-AC86U's. Actually, surprisingly so. While the RT-AC86U could hit 150-210 Mbps download to the lowest level from the AiMesh node a floor and many walls above. The RT-AX86U could hit 350-430Mbps with a single router (before the AiMesh node was added) to this same device (current laptop, plugged into AC power) from two floors above and a few extra walls in between too." I presume you meant RT-AX68U that was highlighted in red?

Yes, I believe you're right. I should have said RT-AX68U in that paragraph after reading that many times right now (almost 3 years after I wrote it).

I won't edit that original post today (as I was also testing the RT-AX68U against my pair of RT-AX86Us too), but I can't make it make sense otherwise.

Btw, you're the only one that has 'caught' that in all this time! (Unless the posts right afterward have clarified that sentence, but I don't have time to check that today).

 

[Tech9]

Are you saying that there is only 5cm of actual antenna inside those three externals on an AC86 or AX86?

Yes, in most home routers, standard dual-band dipole for few bucks. The rest is non-functional "gaming plastic".

RT-AC86U and RT-AX86U (Pro) in particular have also one internal printed antenna (4-stream radio) similar to this:

High-end business class access points have much more complicated antenna arrays like this:

Or like this:

...but cost much more than a home router with all the sticking out antennas, color accents and RGB lights.

 

[sfx2000]

I just want to note that the major factor is not the AP, but the clients - they determine the range...

 

[snovvman]

Yes, in most home routers, standard dual-band dipole for few bucks. The rest is non-functional "gaming plastic".

View attachment 58866

RT-AC86U and RT-AX86U (Pro) in particular have also one internal printed antenna (4-stream radio) similar to this:

View attachment 58868

High-end business class access points have much more complicated antenna arrays like this:

View attachment 58869

Or like this:

View attachment 58870

...but cost much more than a home router with all the sticking out antennas, color accents and RGB lights.

"Gaming plastic" lol. I wish they had non-gaming plastic options for the AC86/88 lines that comes in white. Thanks for all the photos and explanations. They are very helpful!

 

[snovvman]

I just want to note that the major factor is not the AP, but the clients - they determine the range...

That is a good point. In my case, though, I had my 3-node AC86U set up and replaced it with a TP-Link Deco 6E setup in the exact same locations. I did an A-B-A-B-A-B test using my phone for the site survey. Each time, the AC86U provided coverage in the fringe areas while the Decos did not.

 

[coxhaus]

I would assume but I don't really know the better AP antennas would support more clients both active talkers and number of connections. Does ASUS publish how many active talkers and total connections they will support?

 

[L&LD]

No, they don't. But it is generally known that the maximum is around 32 to 50 clients per radio.

 

[coxhaus]

No, they don't. But it is generally known that the maximum is around 32 to 50 clients per radio.

But is that actual talking devices or connected devices? There is a big difference.

 

[L&LD]

Talking, from all conversations on these forums.

Connected can be in the hundreds.

 

[coxhaus]

Talking, from all conversations on these forums.

Connected can be in the hundreds.

So not published. I also read where people are losing connections because they have a lot of IoT devices, not hundreds. Down around 50 or 60.

 

[L&LD]

I don't put much stock, if any, in IoT devices. They're not necessary in the least.

I concentrate on having a robust network with client devices that are important (PCs, NAS, Printers, Scanners, etc.).

At that point, adding IoT devices (and possibly breaking the network, with those underdeveloped, unsecured, and offensively hackable intrusions into a home/business) is up to the network owner. I can't charge enough to 'fix' those issues.

 

[coxhaus]

I don't put much stock, if any, in IoT devices. They're not necessary in the least.

I concentrate on having a robust network with client devices that are important (PCs, NAS, Printers, Scanners, etc.).

At that point, adding IoT devices (and possibly breaking the network, with those underdeveloped, unsecured, and offensively hackable intrusions into a home/business) is up to the network owner. I can't charge enough to 'fix' those issues.

You are living in a fantasy world, not a real one.

 

[L&LD]

You are talking to yourself.

The few customers I have who insist on having IoT devices are always having issues, because of those insecure toys. The rest of my customers have a network that they don't have to think about because it just works.

I only offer help for the latter group, when they need to change.

When IoT devices grow up and can be used dependably and securely, I may reconsider. I don't follow lemmings off cliffs.

 

[snovvman]

@L&LD what do you tell your customers when they want to implement thermostats, cameras, or light switches? Just curious. What would you classify as IoTs? Smart watches, for example, are commonly regarded as IoTs.

Notwithstanding various attacks, which as DoS radio frequency attacks on Z-Wave or WiFi, IoTs can be implemented more safety. With managed switches, VLAN, and discreet SSID, I firewall off the IoTs from my data network.

 

[L&LD]

They implement anything they want. I just don't support what I can't control.

More safely than what? All IoTs are insecure.

Teslas can still be stolen with a cheap radio hack despite new keyless tech

Tesla owners should enable PIN-to-drive protection to thwart relay attacks.

arstechnica.com

Don't believe the false marketing, or get lazy with the promises of easy with less effort...

 

[snovvman]

More safely than what? All IoTs are insecure.

More safely than commingling IoT traffic with your secure data network.

In the *one* example you cited with Tesla, it can be mitigated with PIN to drive, so is it still insecure [unsecure]?

I do believe that a sweeping statement like "All IoTs are insecure" cannot be accurate, especially when tech is constantly changing. If you consider a RPi an IoT and the distro is properly hardened, I would say that it is more secure than many other "robust", in your words, systems.

Going off of your "important client devices" list, there have been plenty of computers and their OS, printers, scanners, NAS (how can we forget) that are equally "insecure", and continue to be so.

 

[Tech Junky]

Why are you bickering about IOT devices when you asked for signal distance?

2.4 - longest distance / penetration
5/6 - faster speeds but less distance

Antennas... yes, they're minimal in functional elements inside unless you're going with higher end antennas or swapping to a MIMO/Yagi/Panel/Dish type setup. Antennas can be cheap for as little as $5 from CN directly for the PCB / postage stamp types or upwards of thousands if you have the real estate for them. Typically though it's more of a limitation imposed by the FCC for indoor use/ The outdoor setups that require a license will broadcast at higher strength though.

Anything in the consumer market though is junk for the most part if you're serious about RF.

 

[snovvman]

Why are you bickering about IOT devices when you asked for signal distance?

There is a voice of reason lol! (I was even advised about it.) Thanks for the information!

 

[L&LD]

More safely than commingling IoT traffic with your secure data network.

In the *one* example you cited with Tesla, it can be mitigated with PIN to drive, so is it still insecure [unsecure]?

I do believe that a sweeping statement like "All IoTs are insecure" cannot be accurate, especially when tech is constantly changing. If you consider a RPi an IoT and the distro is properly hardened, I would say that it is more secure than many other "robust", in your words, systems.

Going off of your "important client devices" list, there have been plenty of computers and their OS, printers, scanners, NAS (how can we forget) that are equally "insecure", and continue to be so.

No, you're assuming it's more safe. I won't give you more examples (millions of them exist), but the point is that people have written articles and whole books about how unsafe IoTs are. Search is your friend.

No matter what you do within your network, if you allow IoT devices Internet access, your network is effectively compromised, if not now, in the future.

And, that single article is to show how ineffective IoT connections are, even on a tech-laden, heavily hardware-laden, and effectively infinite resource-rich, and mega-dollar vehicle like a Tesla is. And it's still insecure. It they could make it secure, they would. They can't. Good luck allowing them on your network(s).

No, computers, OSes, NAS and similar may be found insecure, but they get issued fixes. IoT devices don't. Ever. And for printers/scanners that don't get updates, they're replaced (or, should be, IMO).

Dig a little on your own and see why you can buy cameras for $15 or less. You're the product, not the camera.

 

[sfx2000]

No, they don't. But it is generally known that the maximum is around 32 to 50 clients per radio.

Yes, I agree - I've seen some AP's limit to 32, others limit to 50... once they hit the limit, they can either block new connections, or start shedding connections - depends on the implementation.

WiFi is a shared medium, and those limits are reasonable.