Release ASUS RT-AX86U Firmware version 3.0.0.4.386_49599 (2022/07/12)

[L&LD]

RF signals decrease by 1/(r squared), where r = the distance from the first point of measuring to the second point. That is outdoor.

Indoors, it can be up to 1/(r to the 6th power), or more.

 

[octopus]

I did explain it as best as I can. The reasoning is the post I made previously.

I don't have time to do searches for you (and validate the links brought up via that search).

Try YouTube, you may get lucky and find someone to explain it as you need.

I don't need to search I'm RF developer since many years. Therefore I thought your reasoning did not hold.

 

[L&LD]

So, why don't you explain it to us then?

Don't get stuck on the words used.

Try to get the gist of my (multi-language translated) attempts.

 

[Tech9]

Hope you are joking ...

I just noticed who did you ask for an explanation and decided to explain to you the new technology in simple words, so it makes sense to you. I’ve read about it before here in SNB Forums and it makes perfect sense to me as an electronics engineer.

 

[SAL9K]

RF signals decrease by 1/(r squared), where = the distance from the first point of measuring to the second point. That is outdoor.

Indoors, it can be up to 1/(r to the 6th power), or more.

I think you mean that spherical spreading in air (indoor or out) is proportional to 1/R^2, I.e. -20*log10(R) dB of propagation loss at distance R from antenna? It's not this simple with router antenna's as they aren't point sources but dipoles, and with things like beam-forming have extra directivity when multiple antenna's are used. I don't know where you got the 1/R^6 thing, though.

 

[neil0311]

Are you using a WiFi5 node?

OE

I have a mix of wifi 5 and wifi 6 clients in my house. I’ve connected to the web admin console on the router with both types.

The issue though is with the same clients, and the same router, and the same external interference, when I update the firmware the router will not hold 160MHz.

With the older firmware it works just fine. So it’s not the clients or some DFS problem, unless fundamentally the newer firmware works differently.

 

[Paliv]

I have a mix of wifi 5 and wifi 6 clients in my house. I’ve connected to the web admin console on the router with both types.

The issue though is with the same clients, and the same router, and the same external interference, when I update the firmware the router will not hold 160MHz.

With the older firmware it works just fine. So it’s not the clients or some DFS problem, unless fundamentally the newer firmware works differently.

It definitely could be DFS. There is speculation that the new wifi drivers are more sensitive to radar bounce. ASUS may have updated sensitivity to adhere better to DFS guidelines. But it could also be a quirk...

 

[neil0311]

It definitely could be DFS. There is speculation that the new wifi drivers are more sensitive to radar bounce. ASUS may have updated sensitivity to adhere better to DFS guidelines. But it could also be a quirk...

Well I just bought this router a couple of months ago, but I guess I’ll just have to stay on the older firmware, until and unless they fix the issue.

If they never do, I’ll be on Jan 2022 firmware until I buy a new router.

 

[Tech9]

If they never do, I’ll be on Jan 2022 firmware until I buy a new router.

You just have to decide what do you want - a stable Wi-Fi to all devices or faster speed to some of them, sometimes. What guarantees you 160MHz wide channel with your next router?

 

[bbunge]

I have a mix of wifi 5 and wifi 6 clients in my house. I’ve connected to the web admin console on the router with both types.

The issue though is with the same clients, and the same router, and the same external interference, when I update the firmware the router will not hold 160MHz.

With the older firmware it works just fine. So it’s not the clients or some DFS problem, unless fundamentally the newer firmware works differently.

My router holds 160 MHz as long as a 160 MHz client is attached and then a bit longer. I have AC and AX clients that do 160 MHz. Yes, it does drop back to 80 MHz but the N and AC clients that stay connected do not care.

I have read of some issues when using an AiMesh node that does not support 160 MHz or AC nodes. Those issues are usually related to connection drops when the router is bounced by RADAR. Not sure if there is a real time fix other than using WIFI settings of the lowest common denominator.

For single router use with the AX86U I advocate default WIFI settings and that means using Dual Band SmartConnect. If you have clients that do not like that set up a Guest WIFI for them.

 

[bbunge]

I just noticed who did you ask for an explanation and decided to explain to you the new technology in simple words, so it makes sense to you. I’ve read about it before here in SNB Forums and it makes perfect sense to me as an electronics engineer.

Wow! Two weeks ago I couldn't even spell engineer and now I are one!

(Sorry for the dig but as a former Tool and Die Maker we were always telling the design engineers that what they designed would not work and why. We always figured out another way...)

 

[Paliv]

Well I just bought this router a couple of months ago, but I guess I’ll just have to stay on the older firmware, until and unless they fix the issue.

If they never do, I’ll be on Jan 2022 firmware until I buy a new router.

How many 160Mhz clients do you have? I’d rather have security patches than more bandwidth for one or two devices. But, you may have more devices than we do. Honestly I found 160 more trouble than it’s worth.

 

[neil0311]

You just have to decide what do you want - a stable Wi-Fi to all devices or faster speed to some of them, sometimes. What guarantees you 160MHz wide channel with your next router?

I have both. WiFi is very stable.

 

[neil0311]

How many 160Mhz clients do you have? I’d rather have security patches than more bandwidth for one or two devices. But, you may have more devices than we do. Honestly I found 160 more trouble than it’s worth.

I have 5 or 6 AX clients and a couple of AC clients that do 160MHz.

Been zero trouble until the updates came out.

 

[Paliv]

I have 5 or 6 AX clients and a couple of AC clients that do 160MHz.

Been zero trouble until the updates came out.

Maybe it was for the planes flying by?

 

[neil0311]

Maybe it was for the planes flying by?

Again, would mean that the new firmware and new drivers operate differently from the previous AX86U firmware, as well as the multiple versions of firmware for my RT-AX88U that I used before getting the AX86U.

Either way…new functionality or bug, it’s not working as desired and expected. So I guess I’ll stay put for now.

 

[L&LD]

I think you mean that spherical spreading in air (indoor or out) is proportional to 1/R^2, I.e. -20*log10(R) dB of propagation loss at distance R from antenna? It's not this simple with router antenna's as they aren't point sources but dipoles, and with things like beam-forming have extra directivity when multiple antenna's are used. I don't know where you got the 1/R^6 thing, though.

Yes, it is that simple, but I don't know what you mean by spherical spreading in air?

It doesn't matter what the signal strength is for a given antennae array, the formula I stated holds.

RF Propagation - YouTube

 

[SAL9K]

Yes, it is that simple, but I don't know what you mean by spherical spreading in air?

It doesn't matter what the signal strength is for a given antennae array, the formula I stated holds.

RF Propagation - YouTube

This is not a correct statement. Antenna's generally have directivity gain, i.e. point more strongly in a particular direction, so their free-space loss will no longer be directly proportional to just 1/r^2.

That video is terrible, btw. He seems to just be saying that with walls and other obstacles, indoor RF loss is going to be greater than 1/r^2, which is probably correct.

 

[RogerSC]

Jul 14 20:56:56 kernel: CPU: 1 PID: 2652 Comm: dcd Tainted: P O 4.1.52 #2
Jul 14 20:56:56 kernel: Hardware name: Broadcom-v8A (DT)
Jul 14 20:56:56 kernel: task: ffffffc02b56cb00 ti: ffffffc029998000 task.ti: ffffffc029998000
Jul 14 20:56:56 kernel: PC is at 0x29d34
Jul 14 20:56:56 kernel: LR is at 0x29fb4
Jul 14 20:56:56 kernel: pc : [<0000000000029d34>] lr : [<0000000000029fb4>] pstate: 20030010
Jul 14 20:56:56 kernel: sp : 00000000ffa21900
Jul 14 20:56:56 kernel: x12: 00000000000a211c
Jul 14 20:56:56 kernel: x11: 0000000000081d6c x10: 0000000000000042
Jul 14 20:56:56 kernel: x9 : 0000000000000003 x8 : 0000000000000042
Jul 14 20:56:56 kernel: x7 : 00000000f60fe77c x6 : 0000000000000038
Jul 14 20:56:56 kernel: x5 : 00000000000003bb x4 : 00000000f60fe7dc
Jul 14 20:56:56 kernel: x3 : 0000000000000000 x2 : 000000000000003f
Jul 14 20:56:56 kernel: x1 : 000000000000000f x0 : 0000000000000000

Makes me feel right at home *smile*. Probably won't see it again for a while...seems to be the way that it goes. Don't have a kernel symbol table, of course, so I don't know what the router was doing when this happened.

Actually, the interesting thing was the that the first thing that I DVR'd on my Tablo after flashing this firmware was the 6pm NBC News. That was chopped up into several odds and sods. A real mess, I had 4 DVR chunks of different sizes. Didn't go through them all...but my theory is that since I had the 5GHz. channel set to channel 36 with 160MHz. channel width, that the Tablo disconnected several times to deal with DFS stuff. Although the wireless log doesn't show that, it shows continuous connection for my 4-stream 160MHz. channel width client following the initial reboot.

Anyways, this was shortly after flashing, resetting, and rebooting the router. So I'll see in time what's going on, but if I see more of this I'll try going back to 80MHz. channel and see what happens there. But what I'm seeing looks a lot like what I just mentioned, DFS stuff going on. Didn't see this problem with the GT-AX6000 using 160MHz. channel width, but we'll see in the fullness of time what's going on. Ended up watching the news on "Very Local" app.

Everything else seems to work well, the wireless is surprisingly fast...happy with it as long as i can get the Tablo DVR stuff sorted out.

 

[L&LD]

@SAL9K, do you have any links to support your statement? We may be (I may be) misinterpreting what you're saying to each other.

If you hold everything else the same, and just change the distance, then 1/r^2 must be true.