Need advice on Very Slow Wireless G & N Speed on ASUS RT-AX89X Routers

[fbagnato]

Hi All,

I have a 4 router mesh system setup in my house that was comprising of 4x RT-AC68U routers.

Recently, I purchased 2x RT-AX89X routers to replace the main router and the furthest node from the main router.

We have a couple of older devices in my house which are Wireless G and Wireless N wifi adaptors.

I am getting slow wifi speeds on these older devices (of which 2 of the devices are our main use laptops).

Occasionally they do get decent speeds but most of the time do not get over 10Mbps throughput.

I initially thought the main router was faulty as I spent a month to-ing and fro-ing with ASUS;s (very poor) overseas support centre who suggested the basic suggestions like resetting, changing to default settings, different firmware versions etc. In the end they concluded it was a faulty router and had to be returned.

On the weekend I had the main router swapped with a new RT-AX89X router.

Initially it seemed good giving me around 50Mbps performance on my 50Mbps Internet connection but after a day it has returned to slow Wifi performance.

I read on https://www.snbforums.com/threads/asus-rt-ax89x-assus-merlin.75712/ that user Tech9 claims that this router is a failure of which "performance is not matching the price tag".

I am now debating whether this poor performace is normal for the RT-AX89X and that my best option is to swap the 2x new RT-AX89X routers for another model (maybe the GT-AX6000 suggested in that above-mentioned thread)?

I am surprised that a large company like ASUS would produce such a poor product for the high price paid for it but need some advice about the problems I am encountering.

Is the RT-AX89X a dud? Can I expect better wireless performance if I exchange the routers for GT-AX6000?

I noticed that RMerlin (Asuswrt-Merlin dev) does not support the GT-AX89X with Merlin and am curious as to why? Is it because he believes it is a failed product?

Looking forward to expert opinions on this matter before giving up on my RT-AX89X's.

I have performed all the basic troubleshooting, tried various things and the routers are running the latest firmware.

As an example the wifi connection that I am typing this on is connected at 300Mbps and the fast.com speedtest is getting 12Mbps. My partner's laptop is connected at 300Mbps and the fast.com speedtest is getting 41Mbps.

It seems that this router is not really compatible with all wireless G and N adapters?

Regards,

fbagnato

 

[OzarkEdge]

Hi All,

I have a 4 router mesh system setup in my house that was comprising of 4x RT-AC68U routers.

Recently, I purchased 2x RT-AX89X routers to replace the main router and the furthest node from the main router.

We have a couple of older devices in my house which are Wireless G and Wireless N wifi adaptors.

I am getting slow wifi speeds on these older devices (of which 2 of the devices are our main use laptops).

Occasionally they do get decent speeds but most of the time do not get over 10Mbps throughput.

I initially thought the main router was faulty as I spent a month to-ing and fro-ing with ASUS;s (very poor) overseas support centre who suggested the basic suggestions like resetting, changing to default settings, different firmware versions etc. In the end they concluded it was a faulty router and had to be returned.

On the weekend I had the main router swapped with a new RT-AX89X router.

Initially it seemed good giving me around 50Mbps performance on my 50Mbps Internet connection but after a day it has returned to slow Wifi performance.

I read on https://www.snbforums.com/threads/asus-rt-ax89x-assus-merlin.75712/ that user Tech9 claims that this router is a failure of which "performance is not matching the price tag".

I am now debating whether this poor performace is normal for the RT-AX89X and that my best option is to swap the 2x new RT-AX89X routers for another model (maybe the GT-AX6000 suggested in that above-mentioned thread)?

I am surprised that a large company like ASUS would produce such a poor product for the high price paid for it but need some advice about the problems I am encountering.

Is the RT-AX89X a dud? Can I expect better wireless performance if I exchange the routers for GT-AX6000?

I noticed that RMerlin (Asuswrt-Merlin dev) does not support the GT-AX89X with Merlin and am curious as to why? Is it because he believes it is a failed product?

Looking forward to expert opinions on this matter before giving up on my RT-AX89X's.

I have performed all the basic troubleshooting, tried various things and the routers are running the latest firmware.

As an example the wifi connection that I am typing this on is connected at 300Mbps and the fast.com speedtest is getting 12Mbps. My partner's laptop is connected at 300Mbps and the fast.com speedtest is getting 41Mbps.

It seems that this router is not really compatible with all wireless G and N adapters?

Regards,

fbagnato

Test one AX89X with default settings and all other routers powered OFF. Inspect the Wireless Log for client link rate and connection details.

OE

 

[fbagnato]

Hello OzarkEdge,

Thanks for your reply.

I did what you suggested and the results are as follows (*sensitive information removed):

=======================================================================================
OP Mode : AP
SSID : My2.4Ghz
Phy Mode : 11b/g/n/ax
Bit Rate : 1.1471 Gb/s, 40MHz
Channel : 11

Stations List
------------------------------------------------------------------------------------
idx MAC PhyMode RSSI TX_RATE RX_RATE Connect Time

=======================================================================================
OP Mode : AP
SSID : My5Ghz
Phy Mode : 11a/n/ac/ax
Bit Rate : 4.8039 Gb/s, 80MHz
Channel : 136

Stations List
------------------------------------------------------------------------------------
idx MAC PhyMode RSSI TX_RATE RX_RATE Connect Time
Main My Mac Address 11NA_HT40 -56 90M 450M 00:00:36
========================================================================================

I performed a Speedtest via fast.com and got 30Mbps.

The connection is connected according to Windows at 300Mbps.

The laptop was placed in the same room (but not too close) to the main router.

I have a 50Mbps Internet service so I did not expect the Speedtest to go over 50Mbps.

Also remembering that at the start of a full reset of settings the wifi works ok but then after a day or so becomes very slow.

Regards,

fbagnato

 

[OzarkEdge]

Hello OzarkEdge,

Thanks for your reply.

I did what you suggested and the results are as follows (*sensitive information removed):

=======================================================================================
OP Mode : AP
SSID : My2.4Ghz
Phy Mode : 11b/g/n/ax
Bit Rate : 1.1471 Gb/s, 40MHz
Channel : 11

Stations List
------------------------------------------------------------------------------------
idx MAC PhyMode RSSI TX_RATE RX_RATE Connect Time

=======================================================================================
OP Mode : AP
SSID : My5Ghz
Phy Mode : 11a/n/ac/ax
Bit Rate : 4.8039 Gb/s, 80MHz
Channel : 136

Stations List
------------------------------------------------------------------------------------
idx MAC PhyMode RSSI TX_RATE RX_RATE Connect Time
Main My Mac Address 11NA_HT40 -56 90M 450M 00:00:36
========================================================================================

I performed a Speedtest via fast.com and got 30Mbps.

The connection is connected according to Windows at 300Mbps.

The laptop was placed in the same room (but not too close) to the main router.

I have a 50Mbps Internet service so I did not expect the Speedtest to go over 50Mbps.

Also remembering that at the start of a full reset of settings the wifi works ok but then after a day or so becomes very slow.

Regards,

fbagnato

Is the test client the 5.0 40MHz wireless-n connection shown at 450Mbps... the one Windows says is 300Mbs? Is that Windows7?

Is the legacy wireless client 'slow speed problem' an issue with both old and new routers?

OE

 

[fbagnato]

Yes the test client is the 5.0 40MHz wireless-n connection shown at 450Mbps, the one Windows says is 300Mbps.

The client is running Windows 10.

For the first 3 years the client ran fine with our previous setup of 4x RT-AC68U routers in a mesh setup when I first set up the mesh system but the old RT-AC68U connections deteriorated over time as the mesh router was barely within distance of the main router. As time went on it seemed that the routers lost a bit of signal strength making the mesh connection too weak to reach the main router hence why I replaced the main router and furthest distanced router with the 2x RT-AX89X's.

The bad connection with the old router setup was a problem with all devices.

Our phones which are AC devices are connecting at higher rates but still don't seem great. Just checking my phone at the moment the connection speed is registering as 390Mbps and getting around 50Mbps on Speedtest app (our internet is a 50Mbps plan).

I just don't know if these model routers are up to scratch and I need to find out what to do, like swap them for another ASUS model.

Regards,

fbagnato

 

[OzarkEdge]

Yes the test client is the 5.0 40MHz wireless-n connection shown at 450Mbps, the one Windows says is 300Mbps.

The client is running Windows 10.

For the first 3 years the client ran fine with our previous setup of 4x RT-AC68U routers in a mesh setup when I first set up the mesh system but the old RT-AC68U connections deteriorated over time as the mesh router was barely within distance of the main router. As time went on it seemed that the routers lost a bit of signal strength making the mesh connection too weak to reach the main router hence why I replaced the main router and furthest distanced router with the 2x RT-AX89X's.

The bad connection with the old router setup was a problem with all devices.

Our phones which are AC devices are connecting at higher rates but still don't seem great. Just checking my phone at the moment the connection speed is registering as 390Mbps and getting around 50Mbps on Speedtest app (our internet is a 50Mbps plan).

I just don't know if these model routers are up to scratch and I need to find out what to do, like swap them for another ASUS model.

Regards,

fbagnato

You have come full circle to replacing routers again without narrowing down the problem. Feel free to do that but I suspect money would be better spent replacing slow/legacy clients/adapters.

My observations/suspicions from a distance:

o too much WiFi/APs/radio interference?
o old slow/worn/defective? wireless adapters on a legacy? PC with a modern OS (dirty upgrade?)
o new router indicates a full link rate wireless-n test connection at 450Mbps... nothing wrong with that (except its half-duplex on a time-shared medium, WiFi)
o works, then slows down on both old and new networks... not a router issue?

I would change up your test approach/thinking and try to narrow down the issue. Accept that the new routers are ok and focus on the ISP service/equipment affecting throughput, and the radio environment affecting WiFi, and the client health. Look for too much interference from competing WiFi and any other 5GHz cordless/radio equipment... change 5.0 WiFi channels, try ch Auto and variable bandwidth 20-40-80MHz. Rule out the ISP service by testing with a PC wired directly to the ISP/modem Ethernet with a new/different Cat5e cable... it's probably ok, just slow at 50Mbps. And be sure to test with nothing else connected/wired/wireless to the network and powered ON to rule out extraneous affects... test with one wireless client and one router running... and try to also test with a PC with a newer wireless adapter to compare clients.

OE

 

[fbagnato]

Hello again OzarkEdge,

Thanks for your reply.

> You have come full circle to replacing routers again without narrowing down the problem.

The problems faced with previous router setups have been due to the signal being output from the routers getting weaker with time. It happened with our first setup which were cheap Tenda routers setup with WDS (2 routers in the network). They worked fine for about 5 years then Wifi connections to the furthest-most router became poor. I then upgraded to the 4x RT-AC68U routers in a mesh setup. Again the routers worked well for about 5 years then again the Wifi connections to the furthest-most router became poor. Even though I cannot prove it seemed that the strength of the signal from the router to the other router weakened a bit but as the distance from the furthest-most router to the main router was already nearly at its limits it just needed to weaken slightly to start losing its connection.

I feel confident that there are not other factors causing poor wifi connections.

> Feel free to do that but I suspect money would be better spent replacing slow/legacy clients/adapters.

Why should I do that when the previous routers worked well for the first 5 years with the same wifi adapters and the RT-AX89X routers are supposed to be compatible with Wireless G and N adapters?

>My observations/suspicions from a distance:

> o too much WiFi/APs/radio interference?

The older devices worked fine in the previous router setup until only recently. Nothing in our environment was changed. ie. same cordless phones, appliances, etc.

> o old slow/worn/defective? wireless adapters on a legacy? PC with a modern OS (dirty upgrade?)

As above; the older devices worked fine in the previous router setup until only recently. Nothing in our environment was changed. ie. same cordless phones, appliances, same OS (Windows 10) etc.

o new router indicates a full link rate wireless-n test connection at 450Mbps... nothing wrong with that (except its half-duplex on a time-shared medium, WiFi)

Yes, I do not have a problem with the 450Mbps speed connection the router reported when I did those tests, however in reality according to speedtest I was not even getting close to the 50Mbps that the speed should have been.

o works, then slows down on both old and new networks... not a router issue?

It didn't have this problem in our previous router setups (Tenda then RT-AC68U routers) On the new router it only works initially (maybe less than 1 day) then becomes consistently poor. Why would this happen with all 3 laptops with older Wireless G and N adapters? This indicates to me that the routers are at fault.

> I would change up your test approach/thinking and try to narrow down the issue. Accept that the new routers are ok
> and focus on the ISP service/equipment affecting throughput, and the radio environment affecting WiFi, and the client
> health. Look for too much interference from competing WiFi and any other 5GHz cordless/radio equipment

I don't know how I can narrow down to identify the actual problem. Maybe I setup one of the RT-AC68U routers as the main router just as a test and test the laptops in question with the RT-AC68U test router? If the laptops get good speedtest results then that would prove it is the new RT-AX89X routers that are at fault?

> change 5.0 WiFi channels, try ch Auto and variable bandwidth 20-40-80MHz.

I will do this after using the RT-AC68U test router as mentioned above.

> Rule out the ISP service by testing with a PC wired directly to the ISP/modem Ethernet with a new/different Cat5e
> cable... it's probably ok, just slow at 50Mbps.

Again I will do this after using the RT-AC68U test router as mentioned above. Regarding the 50Mbps speed, it is the plan that we are paying for; 50Mbps Download/20Mbps Upload. I am not worried about that as the newer devices as mentioned previously are getting 50/20 speeds and the speeds are sufficients for our needs.

> And be sure to test with nothing else connected/wired/wireless to the network and powered ON to rule out extraneous
> affects... test with one wireless client and one router running...

Will do.

> and try to also test with a PC with a newer wireless adapter to compare clients.

My desktop is connecting at 1.2Gbps and is getting the 50Mbps speedtest. My phone is connecting at 433Mbps and on my partner's phone she is connected at 1.2Gbps. Both are Wireless AC devices. Haven't tested my daughter's iPhone and Macbook. We also have a lot of other devices that I haven't tested (Internet Radios, TVs, Blu-ray Recorders, etc. but feel that they are working ok. My concern is only with the older G & N devices.

I am getting 2.6Mbps on my laptop as I write this post and my partner is getting 17Mbps. As you would appreciate this is unacceptable and barely usable performance on what are our main 2 laptop that we use.

Would like your opinion/reply on these points.

Regards,

fbagnato

 

[OzarkEdge]

Would like your opinion/reply on these points.

I can't think of anything to add.

OE

 

[RTWZ80]

For the RT-AX89X, did you try switching Wireless Mode to Legacy and disable 802.11ax/Wifi 6 mode on both the 5Ghz and 2.4Ghz bands? Also the Authentication Method to WPA/WPA2-Personal? Perhaps that may help with older/legacy devices.

 

[oldguy]

On a quick scan of this thread, I did not see any mention as to how the various mesh nodes are connected to the primary router.
If they are using wireless connections the RT-AX-89X is only a dual band device which means that the backhaul will share the 5GHz channel with 5GHz clients
if one or more mesh nodes form a daisy chain, this is not an ideal situation as there will be backhaul traffic that could impact client performance.

On the Merlin Firmware point, the RT AX89X is based on a Qualcomm platform which is not covered by Merlin release’s

I have found two irritants with my 2 * RT AX89X mesh system, one being devices frequently disconnecting which was fixed by looking at the signals strengths and adjusting the handoff threshold for both 5G and 2.4G bands as the system log showed hand offs between the two nodes. (I used the Apple Airport Utility WiFi scan as my android tablet used previously has died given there are lots of wireless scanning tools available)

the second performance issue is noticeable after 15 to 20 days, when the DRAM usage goes to about 620MB, a system reboot from the AIMESH ”System Options” tab seems to fix the problem

 

[RTWZ80]

On a quick scan of this thread, I did not see any mention as to how the various mesh nodes are connected to the primary router.
If they are using wireless connections the RT-AX-89X is only a dual band device which means that the backhaul will share the 5GHz channel with 5GHz clients
if one or more mesh nodes form a daisy chain, this is not an ideal situation as there will be backhaul traffic that could impact client performance.

On the Merlin Firmware point, the RT AX89X is based on a Qualcomm platform which is not covered by Merlin release’s

I have found two irritants with my 2 * RT AX89X mesh system, one being devices frequently disconnecting which was fixed by looking at the signals strengths and adjusting the handoff threshold for both 5G and 2.4G bands as the system log showed hand offs between the two nodes. (I used the Apple Airport Utility WiFi scan as my android tablet used previously has died given there are lots of wireless scanning tools available)

the second performance issue is noticeable after 15 to 20 days, when the DRAM usage goes to about 620MB, a system reboot from the AIMESH ”System Options” tab seems to fix the problem.

On your first irritant, I think it's a problem across most newer Asus routers. I experienced the same 'devices frequently disconnecting' issue on my pair of AXE11000s in AiMesh mode and fixed it the same way you did. On the two AXE16000s which I recently got, the issue was compounded by the new RangeBoost Plus feature which upped the handoff threshold quite a bit between the main router and AiMesh node. I've since returned the AXE16000s because NAT/CPU limitations prevented the routers from reaching near ISP-advertised wired speeds. (I'm on 10Gbps broadband) without turning off nearly all the features that I buy Asus routers for (AiMesh, VPN, QoS etc.) as per Asus Tech Support's recommendation.

In contrast, I experienced little to no impact on wired speeds with the RT-AX89X with AiProtection enabled.

Ditto on the second irritant. I scheduled daily reboots for the RT-AX89X in the wee hours of the morning so no one else's Wifi gets cutoff.

 

[fbagnato]

I can't think of anything to add.

OE

LOL......

I have to put off the testing of the RT-AC68U test router for a couple of days at least as I am snowed under with other matters, but I will report my findings in case anyone else is in the same boat as me.

Regards,

fbagnato

 

[fbagnato]

For the RT-AX89X, did you try switching Wireless Mode to Legacy and disable 802.11ax/Wifi 6 mode on both the 5Ghz and 2.4Ghz bands? Also the Authentication Method to WPA/WPA2-Personal? Perhaps that may help with older/legacy devices.

Hello RTWZ80,

Thanks for your reply.

I just tried changing the Wireless Mode to Legacy, Disabled 802.11ax/Wifi 6 mode and changed the Authentication Method from WPA2-Personal to WPA/WPA2-Personal and did a Speedtest via fast.com. My laptop got 5.6Mbps and my partner got 6.9Mbps.

Regards,

fbagnato

 

[fbagnato]

On a quick scan of this thread, I did not see any mention as to how the various mesh nodes are connected to the primary router.
If they are using wireless connections the RT-AX-89X is only a dual band device which means that the backhaul will share the 5GHz channel with 5GHz clients
if one or more mesh nodes form a daisy chain, this is not an ideal situation as there will be backhaul traffic that could impact client performance.

On the Merlin Firmware point, the RT AX89X is based on a Qualcomm platform which is not covered by Merlin release’s

I have found two irritants with my 2 * RT AX89X mesh system, one being devices frequently disconnecting which was fixed by looking at the signals strengths and adjusting the handoff threshold for both 5G and 2.4G bands as the system log showed hand offs between the two nodes. (I used the Apple Airport Utility WiFi scan as my android tablet used previously has died given there are lots of wireless scanning tools available)

the second performance issue is noticeable after 15 to 20 days, when the DRAM usage goes to about 620MB, a system reboot from the AIMESH ”System Options” tab seems to fix the problem

Hello oldguy,

Thanks for your reply.

The mesh nodes are connected to the primary router wirelessly (I just added Mesh Nodes through the configuration utility.

> If they are using wireless connections the RT-AX-89X is only a dual band device which means that the backhaul will share the 5GHz channel with 5GHz clients

Should this produce such poor throughput on wifi clients like I am experiencing?

> if one or more mesh nodes form a daisy chain, this is not an ideal situation as there will be backhaul traffic that could impact client performance.

I don't really understand but if it is that the path from one mesh node links to the primary router via another mesh node then there is a mesh node in the master bedroom that links to the primary router via my daughter's bedroom's mesh node. The master bedroom node barely gets used though. My daughter's bedroom's mesh node is heavily used. The Living Room RT-AX89X mesh node links directly to the primary router. I see the paths described above in the AiMesh tab of the router configuration. Forgive me if this is not what you mean by daisy-chained.

Remembering that when I tested the connection earlier with a third wireless G Windows 10 laptop I did that in the same room as the primary router with all nodes turned off and no other devices (except a wired desktop) connected to the router and I still experienced poor speeds.

>On the Merlin Firmware point, the RT AX89X is based on a Qualcomm platform which is not covered by Merlin release’s

Ah, I capish. So it is not because the developer believes the RT-AX89X is not worthy of being included as a supported device. Thanks for clearing that up.

> I have found two irritants with my 2 * RT AX89X mesh system, one being devices frequently disconnecting which was
> fixed by looking at the signals strengths and adjusting the handoff threshold for both 5G and 2.4G bands as the system
> log showed hand offs between the two nodes. (I used the Apple Airport Utility WiFi scan as my android tablet used
> previously has died given there are lots of wireless scanning tools available)
>
> the second performance issue is noticeable after 15 to 20 days, when the DRAM usage goes to about 620MB, a system
> reboot from the AIMESH ”System Options” tab seems to fix the problem

I am not experiencing disconnections (at least that is not a problem!). Are you on the latest firmware 3.0.0.4.386_47191-g9ddcc31?

Just checked my RAM status and it is at 536Mb used. I performed the reboot from the AIMESH ”System Options” tab but the RAM is now 545Mb used so the reboot didn't make much difference. Did a Speedtest via fast.com. My laptop got 6.1Mbps and my partner got 10Mbps.

Do you have any older Wireless G or Wireless N clients connecting to your RT-AX89X routers? If so are you getting acceptable speeds?

Regards,

fbagnato

 

[oldguy]

Yup I am on latest firmware
I have 22 Wemo Switch’s and a Printer that are 802.11N, all of my computers are 802.11AC or 802.11AX or better, so not the same setup as you have

looking at the picture below this shows all of the devices connected to the primary mesh node, a daisy chain would be were the first down stream device is connected to the primary mesh node, the second down stream device is connected to the first down stream device Etc….

The primary problem with wireless backhaul in an AI mesh platform with a dual band router is that the 5G device data and all 2.4Ghz data is routed to the 5GHz channel to connect upstream, there is also overhead on the backhaul channel that consumes bandwidth.
If you have a second mesh node connected to the first wireless mesh node the same data aggregation applies, and will also consume bandwidth on the first mesh node 5G channel
The other constraint is then is whether the 5G is using 40Mhz, 80Mhz or 160Mhz bandwidth, which also limits the through put of the system

Thats why a Tri band ASUS router may be a better solution as this has a dedicated 5GHz channel for backhaul

I have a 10GBit Ethernet backhaul between my two RT AX89X systems which works for me

When time permits i will add my old tri band router to the system with a gigabit Ethernet backhaul so I can improve connection to devices on one side of my home when i have visitors stay over

I recall seeing a great explanation of the impact of daisy chained access points that I will try to find and post the linkl

 

[sfx2000]

changed the Authentication Method from WPA2-Personal to WPA/WPA2-Personal

Change it back - 802.11n requires WPA2 (or none) for frame aggregation.

Also consider using 20MHz channels vs. 40MHz channel - using wide channels cuts the radio link budge by 3 dB (up and downlink) - that's enough to drop an MCS level

 

[fbagnato]

Yup I am on latest firmware
I have 22 Wemo Switch’s and a Printer that are 802.11N, all of my computers are 802.11AC or 802.11AX or better, so not the same setup as you have

looking at the picture below this shows all of the devices connected to the primary mesh node, a daisy chain would be were the first down stream device is connected to the primary mesh node, the second down stream device is connected to the first down stream device Etc….
View attachment 43300
The primary problem with wireless backhaul in an AI mesh platform with a dual band router is that the 5G device data and all 2.4Ghz data is routed to the 5GHz channel to connect upstream, there is also overhead on the backhaul channel that consumes bandwidth.
If you have a second mesh node connected to the first wireless mesh node the same data aggregation applies, and will also consume bandwidth on the first mesh node 5G channel
The other constraint is then is whether the 5G is using 40Mhz, 80Mhz or 160Mhz bandwidth, which also limits the through put of the system

Thats why a Tri band ASUS router may be a better solution as this has a dedicated 5GHz channel for backhaul

I have a 10GBit Ethernet backhaul between my two RT AX89X systems which works for me

When time permits i will add my old tri band router to the system with a gigabit Ethernet backhaul so I can improve connection to devices on one side of my home when i have visitors stay over

I recall seeing a great explanation of the impact of daisy chained access points that I will try to find and post the linkl

Hello again,

That all makes sense but I can't see how my performance would be affected as bad as it is.

Here is a photo of my Imesh topology:

As you can see there are 0 clients currently using Master Bedroom and Tia's bedroom.

A speedtest at the moment returned 14Mbps on my laptop and 13Mbps on my partner's laptop.

I am planning to connect a RT-AC68U router leftover from my previous setup as a test router but cannot do that for a couple of days. I will then test the performance of that router for my older wifi adapters. That should tell me a bit more about the performance of the current setup.

I also disconnected the cordless phone that is in the same room as the primary router but that didn't fix the poor wifi performance.

Regards,

fbagnato

 

[oldguy]

Hello again,

That all makes sense but I can't see how my performance would be affected as bad as it is.

Here is a photo of my Imesh topology:

View attachment 43312

As you can see there are 0 clients currently using Master Bedroom and Tia's bedroom.

A speedtest at the moment returned 14Mbps on my laptop and 13Mbps on my partner's laptop.

I am planning to connect a RT-AC68U router leftover from my previous setup as a test router but cannot do that for a couple of days. I will then test the performance of that router for my older wifi adapters. That should tell me a bit more about the performance of the current setup.

I also disconnected the cordless phone that is in the same room as the primary router but that didn't fix the poor wifi performance.

Regards,

fbagnato

When you select Mesh node in your MAP above, have you looked at the management tab to get any information about the link status?
Also on the AIMesh page there are two tabs “Topology” which shows the layout you have and “optimization”
The latter is a mechanism that is supposed to optimize the AIMESH topology to give the best performance, if you have not run this “Optimization” process it might help.

on the wireless phone, i have a cordless phone which has no impact on my network as it is operating between 1921 and 1948 MHz

 

[oldguy]

Looking at an AIMESH node management TAB, I see that there is an option for auto wireless backhaul selection.

do you see any sign of this type switching in the log?

the other interesting feature in the Adaptive QoS screen, is a TAB for internet speed, which allows the primary router attached to the modem to run an Ookla speednet Test that will allow you to verify that you are getting the speed that you pay for.

 

[fbagnato]

Change it back - 802.11n requires WPA2 (or none) for frame aggregation.

Also consider using 20MHz channels vs. 40MHz channel - using wide channels cuts the radio link budge by 3 dB (up and downlink) - that's enough to drop an MCS level

Hello sfx2000,

Thanks for your reply.

I changed the Authentication Method back to WPA2-Personal.

I then changed the Channel Bandwidth on both the 2.4Ghz and 5Ghz bands to 20 Mhz and got the following results:

2.4Ghz Connection - Speedtest - My Laptop 16Mbps, Partner's Laptop 47Mbps
5Ghz Connection- Speedtest - My Laptop 7.5Mbps, Partner's Laptop 47Mbps

I then changed the Channel Bandwidth on the 5Ghz band to 40 Mhz and got the following results:

5Ghz Connection- Speedtest - My Laptop 3.3Mbps, Partner's Laptop 42Mbps

Changed the bandwidths back to 2.4Ghz = 20/40 Mhz and 5Ghz = 20/40/80/160 Mhz and got the following results:

2.4Ghz Connection - Speedtest - My Laptop 3.6Mbps, Partner's Laptop 27Mbps
5Ghz Connection- Speedtest - My Laptop 9.6Mbps, Partner's Laptop 23Mbps

When I changed Channel Bandwidth on both the 2.4Ghz and 5Ghz bands to 20 Mhz I don't know if I was imagining it but the quality on the TV when streaming was poorer than normal. Wouldn't the performance of the connection be lowered when set to 20 Mhz as opposed to 20/40/80/160 Mhz if interference is not a problem??? I was concerned that the throughput on the TV was lowered when set to 20 Mhz.

Regards,

fbagnato