M1 iPad Pro Dropping WiFi Constantly

[marclafountain]

I have a 12.9-inch M1 iPad Pro and two Asus ROG Rapture GT-AX11000 WiFi routers running Asuswrt Merlin 386.2_6 with AiMesh. I have an issue with the M1 iPad Pro dropping WiFi. I know the WiFi code is Asus closed source. I would still deeply appreciate any advice anyone can offer, especially if any Merlin settings changes might help.

The M1 iPad Pro works well in all rooms of my home except one bedroom. In that bedroom, the M1 iPad Pro drops off WiFi completely after 2-3 minutes of use. I can turn WiFi off and back on, use WiFi for 2-3 minutes, and the problem repeats.

When WiFi drops, the router logs always say something like:

Deauth_ind 4A:31:20:33:0D:3F, status: 0, reason: Deauthenticated because sending station is leaving (or has left) IBSS or ESS (3), rssi:0

I see from other forum posts that this error occurs incorrectly at times due to an Asus code bug, But it also occurs whenever the M1 iPad Pro drops WiFi.

I don’t think this is a coverage or interference issue. When the M1 iPad Pro works for the first 2-3 minutes, it shows strong signal strength and data transfer is peppy. Also, two older iPad Pros, an iPhone 11, and an iPhone 12 all work perfectly fine on WiFi in that bedroom. Only the M1 iPad Pro has the drop-off problem in that bedroom.

From watching the router dashboards, I think perhaps the M1 iPad Pro is trying to switch between the 2.4 and 5 bands and something fails. But I’m not certain of that.

I have tried:

- Rebooting the M1 iPad Pro
- Forgetting the WiFi network on the M1 iPad Pro and adding it again
- Rebooting both routers
- Resetting the Network Settings on the M1 iPad Pro and adding the WiFi network again
- Reducing sensitivity of the routers’ Roaming Assistant
- Disabling routers’ Roaming Assistant
- Enabling and disabling WiFi Agile Multiband on all the bands of the routers

Nothing is helping. Anyone have any ideas for me?

 

[L&LD]

Try a different Control Channel.

 

[Batnun]

I have the M1 iPad Pro 11 inch and the ASUS RT-AX86U with the latest Merlin firmware.
Never had any problem, but I'm using separate SSIDs for 5GHz and 2.4GHz, and making sure the iPad (and almost all the other devices) connects only to the 5GHz one...

 

[marclafountain]

I now realize that all WiFi devices in that bedroom have been using 2.4 rather than 5. My working theory is that the 12.9-inch M1 iPad Pro is the only device able to get 5 in that bedroom, but it is a very precarious connection that drops often.

I see that Asus makes a WiFi 6 AiMesh network extender, so I am going to order one and see if I can improve 5 coverage in that bedroom. I live in Indonesia so it will be a few weeks before I have the extender and know whether it solves my issue. I will update this thread then.

Thanks!

 

[sbsnb]

5 GHz is junk for range. Is your Internet speed (and the iPad's storage and processor) faster than your maximum 2.4 GHz speed? If not, there's no benefit to using 5 GHz unless you're in an apartment/condo with so many neighbors on 2.4 GHz that it's unusable.

Just having a lot of APs visible doesn't mean the 2.4 GHz spectrum is saturated. You have to check the actual bandwidth utilization.

 

[marclafountain]

5 GHz is junk for range. Is your Internet speed (and the iPad's storage and processor) faster than your maximum 2.4 GHz speed? If not, there's no benefit to using 5 GHz unless you're in an apartment/condo with so many neighbors on 2.4 GHz that it's unusable.

Just having a lot of APs visible doesn't mean the 2.4 GHz spectrum is saturated. You have to check the actual bandwidth utilization.

I have an 12.9-inch M1 iPad Pro, so it’s pretty beefy in terms of processor speed and supports WiFi 6. I’m also doing video chats with people half way around the globe, so I want any networking edge I can get.

I have run speed tests on the IPad on 2.4 in the bedroom and 5 standing outside the bedroom closer to the router. The 5 connection is much faster, especially on upload speed. So, I want to see if the AiMesh extender will make 5 in the bedroom viable or make 2.4 in the bedroom faster. Thanks for the suggestion!

 

[sbsnb]

I have an 12.9-inch M1 iPad Pro, so it’s pretty beefy in terms of processor speed and supports WiFi 6. I’m also doing video chats with people half way around the globe, so I want any networking edge I can get.

5 GHz or 2.4GHz isn't going to make a difference for video chat, or even HD video streaming. 2.4 GHz on the AX86U has enough speed to support more than 20 simultaneous 4K video streams.

If your Internet connection is slower than 500 Mbps, you will never see a speed difference between 5 GHz and 2.4 GHz no matter what you're doing, absent interference issues. What you will see is that 2.4 GHz will likely work up to 200 feet or more from the router and 5 GHz is lucky to get half that distance.

 

[L&LD]

For throughput, you'd be correct.

For latency, 5GHz will be noticeably better (lower). And depending on the video chat, can make a difference even there.

 

[sbsnb]

What's the latency difference, and what's the source of the difference. We've run 4 simultanous MS Teams and/or Zoom meetings off of 2.4 GHz on the RT-AC88U while a kid is watching Netflix with no issues whatever.

 

[marclafountain]

I take your point about 2.4 perhaps being all I need. But, I still wonder if the extender would help make 2.4 faster or more stable in the bedroom.

I would also like to get any benefits of 5 when the iPad is nearer the routers. I do see speed and ping improvements when testing on 2.4 in the bedroom versus 5 outside of the bedroom with the iPad. So I wonder if 5 could provide some value in the bedroom.

The extender is inexpensive and works with AiMesh, so I figure I have little to lose by giving it a try. Will let you all know what I find in a few weeks.

 

[sbsnb]

You inspired me to check.

On 5GHz about 50 feet from router:

Pinging yahoo.com [98.137.11.163] with 32 bytes of data:
Reply from 98.137.11.163: bytes=32 time=35ms TTL=51
Reply from 98.137.11.163: bytes=32 time=33ms TTL=51
Reply from 98.137.11.163: bytes=32 time=33ms TTL=51
Reply from 98.137.11.163: bytes=32 time=33ms TTL=51

Ping statistics for 98.137.11.163:
    Packets: Sent = 4, Received = 4, Lost = 0 (0% loss),
Approximate round trip times in milli-seconds:
    Minimum = 33ms, Maximum = 35ms, Average = 33ms

Pinging google.com [172.217.13.206] with 32 bytes of data:
Reply from 172.217.13.206: bytes=32 time=74ms TTL=112
Reply from 172.217.13.206: bytes=32 time=74ms TTL=112
Reply from 172.217.13.206: bytes=32 time=73ms TTL=112
Reply from 172.217.13.206: bytes=32 time=74ms TTL=112

Ping statistics for 172.217.13.206:
    Packets: Sent = 4, Received = 4, Lost = 0 (0% loss),
Approximate round trip times in milli-seconds:
    Minimum = 73ms, Maximum = 74ms, Average = 73ms

On 2.4GHz:

Pinging yahoo.com [98.137.11.163] with 32 bytes of data:
Reply from 98.137.11.163: bytes=32 time=32ms TTL=51
Reply from 98.137.11.163: bytes=32 time=33ms TTL=51
Reply from 98.137.11.163: bytes=32 time=38ms TTL=51
Reply from 98.137.11.163: bytes=32 time=35ms TTL=51

Ping statistics for 98.137.11.163:
    Packets: Sent = 4, Received = 4, Lost = 0 (0% loss),
Approximate round trip times in milli-seconds:
    Minimum = 32ms, Maximum = 38ms, Average = 34ms

Pinging google.com [172.217.13.206] with 32 bytes of data:
Reply from 172.217.13.206: bytes=32 time=74ms TTL=112
Reply from 172.217.13.206: bytes=32 time=72ms TTL=112
Reply from 172.217.13.206: bytes=32 time=72ms TTL=112
Reply from 172.217.13.206: bytes=32 time=72ms TTL=112

Ping statistics for 172.217.13.206:
    Packets: Sent = 4, Received = 4, Lost = 0 (0% loss),
Approximate round trip times in milli-seconds:
    Minimum = 72ms, Maximum = 74ms, Average = 72ms

 

[Tech9]

If your Internet connection is slower than 500 Mbps, you will never see a speed difference between 5 GHz and 2.4 GHz no matter what you're doing, absent interference issues.

What technology gives you up to 500Mbps on the 2.4GHz band?

 

[marclafountain]

I suspect what is happening in my specific case is that the bedroom has a lot of obstacles between it and the WiFi router. So even 2.4 with an extender may have better speed and latency than the current 2.4 there. If the extender can also make 5 work well in the bedroom that’ll just be gravy.

 

[L&LD]

50' from the router is a whole house away and not what is optimal for 5GHz.

What are your base latency measurements over a wired 1GbE connection (or better) to the same servers from an AC powered device?

Now, what are your 5GHz measurements from 15, 20, 25, 30, and 35' away from the router? From an AC powered device or a non - handheld device running on battery but in Performance mode.

That will be a better test method/results.

And it will show you what your router and Wi-Fi environment is optimal, for your testing device(s).

 

[sbsnb]

What technology gives you up to 500Mbps on the 2.4GHz band?

4 streams @ 256-QAM
4 streams @ MCS 7
3 streams @ 1024-QAM
4 streams @ 1024-QAM

 

[sbsnb]

50' from the router is a whole house away and not what is optimal for 5GHz.

What are your base latency measurements over a wired 1GbE connection (or better) to the same servers from an AC powered device?

Now, what are your 5GHz measurements from 15, 20, 25, 30, and 35' away from the router? From an AC powered device or a non - handheld device running on battery but in Performance mode.

That will be a better test method/results.

And it will show you what your router and Wi-Fi environment is optimal, for your testing device(s).

50 feet isn't the full distance across my house. Corner to corner I have about 75 feet. The neighbor's house is 90 feet from the router.

The laptop connected by 5GHz on the desk about 12 feet from the router shows nearly identical pings:

Pinging yahoo.com [98.137.11.163] with 32 bytes of data:
Reply from 98.137.11.163: bytes=32 time=34ms TTL=51
Reply from 98.137.11.163: bytes=32 time=34ms TTL=51
Reply from 98.137.11.163: bytes=32 time=34ms TTL=51
Reply from 98.137.11.163: bytes=32 time=34ms TTL=51

Ping statistics for 98.137.11.163:
    Packets: Sent = 4, Received = 4, Lost = 0 (0% loss),
Approximate round trip times in milli-seconds:
    Minimum = 34ms, Maximum = 34ms, Average = 34ms

Pinging google.com [172.217.13.206] with 32 bytes of data:
Reply from 172.217.13.206: bytes=32 time=74ms TTL=112
Reply from 172.217.13.206: bytes=32 time=77ms TTL=112
Reply from 172.217.13.206: bytes=32 time=75ms TTL=112
Reply from 172.217.13.206: bytes=32 time=72ms TTL=112

Ping statistics for 172.217.13.206:
    Packets: Sent = 4, Received = 4, Lost = 0 (0% loss),
Approximate round trip times in milli-seconds:
    Minimum = 72ms, Maximum = 77ms, Average = 75ms

I can't do the plugged in test because it's in the closet where the wife is sleeping.

 

[Tech9]

4 streams @ 256-QAM
4 streams @ MCS 7
3 streams @ 1024-QAM
4 streams @ 1024-QAM

Most clients are 2-stream and most 2.4GHz networks operate @20Mhz. Turbo/NitroQAM are Broadcom non-standard specifications, used for marketing purposes. Real life 2.4GHz band throughput is up to 100Mbps, with some luck. I’m not sure if @thiggins can achieve 500Mbps throughput on 2.4GHz with his test equipment with RF shielded chamber.

 

[sbsnb]

I'm getting 157 Mbps right now on speedtest.net on 2.4 GHz, and I'm 50 feet away through 5 walls.

 

[Tech9]

What happened to up to 500Mbps? Can you get that next to the router 5 feet away?

 

[sbsnb]

What happened to up to 500Mbps?

50 feet and 5 walls. It's 178 Mbps this morning.

Can you get that next to the router 5 feet away?

Unfortunately, all three mobile devices I have appear to be limited to around 100 Mbps whether on 5 GHz or 2.4 GHz. They're all connecting with only 802.11n using 2 streams.

I can't move the desktop.