Hi,
I have been trying different settings to get a decent coverage and speed with my new router with the 5GHZ band but I cannot seem to find a solution.
Any help would be much appreciated.

With Asus RT-AC86U I get full coverage at points #2,3,4 with around 440mbits and get coverage even in my balcony at #5.
I have tried using different countries including mine, set "TxPower" to 30-40, and "Antenna Gain" to 0 but those even lessened the coverage somehow.

The following is my apartment plan and is around 95 square meters: And my config:

Why don't you use the lower channels 5160-5340 (UNII-1 and UNII-2A)?

The channels you use now, are subject to reduced TX power AFAIK.
Nothing Mikrotik can do about it, they are pretty strict in following regulations for that aspect.

Which channels does Asus use ?

They are being used by my neighbours.
Asus was fixed to use channel 100.
I just tested it again with frequencies 5250-5330 and got slightly worse performance. 436mbits next to the router and around 50 on the other locations, with no range improvement.

Stick with hoelv, if anyone can get your ax3 working well he is da man.

That seems very low for an AX class device. However, the AC86 was known for its range - even newer ASUS routers have a hard time keeping up. The ASUS can also automatically change the band used (Smart Connect) as required by the client.

1. How was the performance with the previous firmware, and are you sure you updated both the software AND the firmware ?
2. Did you reset the router ?
3. Did you do a survey scan on the clients and check for channel utilization ?
4. Was the AC86 turned off when you performed your tests ?
5. Did you perform your tests with more than one client (I had an old Intel card what was notoriously finicky on the hAP ac2 that I use as an AP) ?
6. Can you move the router to room 4 to see if your range improves ?
7. How is the range on 2.4 GHz ?

It's tedious, but when I had issues, I would manually select frequency/channel width according to my survey results and work from there with each client. Changing gain/power shouldn't be necessary and might cause more problems. Lastly, like previously mentioned, higher frequencies generally offer the worst range...

I am using an hAP AX3 and I am very happy (way better than other solutions).

You are using two different SSID for 5Ghz and 2.4 Ghz.
My experience in a concrete building with 80 m2 would be to se the same SSID and let the client choose the best transmission.

Also don't look at speed indicated by wireless, but calculate them using iPerf3.
You will see that speed depends on walls, doors, concrete, etc...

Your speed will be 1Gbit/s in front of the AX3 and maybe 200Mb/s on the balcony.
All this is normal.

The underlying difference may be elsewhere.

That Asus uses 4x4 MIMO on AC if I read the specs correctly.
AX3 uses 2x2 AX

So even if there is a difference between wifi5 (Asus) and Wifi6 (MT), the difference in MIMO-usage gives an advantage to Asus.
Not an equal comparison then.

Hi,

Thanks for all your replies.

That seems very low for an AX class device. However, the AC86 was known for its range - even newer ASUS routers have a hard time keeping up. The ASUS can also automatically change the band used (Smart Connect) as required by the client.

1. How was the performance with the previous firmware, and are you sure you updated both the software AND the firmware ?
- Unfortunately, I upgraded it from 7.8 as soon as I started it. On some Reddit posts, people were saying it got worse after ROS 7.6. I am thinking about reverting it back to 7.6.
2. Did you reset the router ?
- I did.
3. Did you do a survey scan on the clients and check for channel utilization ?
- I did it in the past when I bought the Asus. I got the best performance with channel "11" for 2.4ghz and channel "100" for 5ghz(I recall all the higher channels were the same; it didn't matter which).
4. Was the AC86 turned off when you performed your tests ?
- Yes
5. Did you perform your tests with more than one client (I had an old Intel card what was notoriously finicky on the hAP ac2 that I use as an AP) ?
- I used MacBook Pro 16 - M1 PRO, Dell Precision 3561, and iPhone 13 for the tests.
6. Can you move the router to room 4 to see if your range improves ?
- I can't. That is where my optics cable and keystone jacks for the rooms are.
7. How is the range on 2.4 GHz ?
The range seems to be okay for a standard router, but the connection appeared to be quite unstable after doing some additional tests today.
Sitting at #4 without moving, I got different results like D:70, U:25, D:0.50, U:7, D:92, U:58

It's tedious, but when I had issues, I would manually select frequency/channel width according to my survey results and work from there with each client. Changing gain/power shouldn't be necessary and might cause more problems. Lastly, like previously mentioned, higher frequencies generally offer the worst range...

===================================================================================

I am using an hAP AX3 and I am very happy (way better than other solutions).

You are using two different SSID for 5Ghz and 2.4 Ghz.
My experience in a concrete building with 80 m2 would be to se the same SSID and let the client choose the best transmission.

Also, don't look at speed indicated by wireless, but calculate them using iPerf3.
You will see that speed depends on walls, doors, concrete, etc...

Your speed will be 1Gbit/s in front of the AX3 and maybe 200Mb/s on the balcony.
All this is normal.

Sitting at #4, my MacBook Pro cannot connect to the 5ghz band, saying I should get closer to the router.
I returned the 5 GHz band to its default settings and got 20 Mbits at #4. I saw that it selected channel "5745/ax/Ceee".
I get mixed results with the 2.4 GHz band, just with better coverage.
All the testing has been a frustrating experience.
I don't expect a similar performance as Asus, but this performance seems below an entry-level router.
My neighbor with the cheapest ISP router below my apartment says he gets 200 MB out of 300 at #4.
I could use the ASUS as an access point or buy another access point, but I don't want to.
I expected this device to have decent coverage, at least.
I saw people complaining on Reddit, specifically for hAP AX3, that Wi-Fi became unusable after upgrading from ROS 7.6 to ROS 7.8, and they downgraded it to ROS 7.6.

Upgrade to at least 7.11.2.

I use ax3 and ax2 to cover a HOUSE with 3 floors, thick brick walls and concrete floors.

Upgrade to at least 7.11.2.

I use ax3 and ax2 to cover a HOUSE with 3 floors, thick brick walls and concrete floors.

I am already on 7.11.2. I don't get how you get coverage in a three-floor house, and I get such a bad performance in an apartment.
Please let me know if you think I am missing something and need me to share additional configuration information.
Shall I open a ticket to Mikrotik as well? Do they provide support to all users?

Unless the unit is defective (possible, but unlikely), there's surely a setting/firmware issue in a space with probably lots of interference. You should downgrade back to the factory firmware just in case. Even if I had an ax3, my results would be different from yours. Anyway, it would also be good to get the signal strengths readings during the survey from a client. And I would finally test the router somewhere else to make sure it's defective (work, a relative's or a friend's house).

I am out of ideas. MT support isn't the best, from the comments I see online - you can try. You might have better luck going back to the dealer it was purchased from. Paying a consultant to fix the issues you are having doesn't sound reasonable for a low-cost router...

The underlying difference may be elsewhere.

That Asus uses 4x4 MIMO on AC if I read the specs correctly.
AX3 uses 2x2 AX

So even if there is a difference between wifi5 (Asus) and Wifi6 (MT), the difference in MIMO-usage gives an advantage to Asus.
Not an equal comparison then.

MT wireless cannot be fairly compared to ASUS even if they could be on par feature wise. While they do have duds and other issues, they use optimized closed source drivers tied to specific kernel versions (in the last decade at least). Hence, they don't support anything for a very long time and move on to new gen features quickly (they already have Wi-Fi 7 routers announced). Merlin has been limiting support because of this. They also sell a huge amount of units, and thus, they have a very close relationship with chip suppliers like Broadcom.

This is a strange problem you have here, I can confirm that with ax3 i had no problem with maxing out my internet connection in my house (two stories, and floor is made from reinforced concrete, and I don't have drywall walls, all made from bricks.)

Did you check if your antennas are screw tightly ? Is there any source of interference near router ?

Set your frequency to 5180 for eg. Also you should perform a scan to see what channel are less crowded.

I would suggest to start again from scratch.
You made quite some changes w.r.t. TX power etc.

Build from there.
First suggestion: set ssid for both radios the same as already mentioned.
Let the client decide which one to connect to.

Setting Antenna Gain to 0 and boosting the Tx Power as you have done will seriously distort the signal and make it a noisy garbled mess with lower speeds and shorter range, not longer!

Set the Antenna Gain to 5-6 as it should be (the datasheet specify the antenna gain 5.5) and experiment with lowering the Tx Power. For reference i use a gain of 6 and tx power 22 with much improved performance from what i previously had.

Also try different channels. Don't worry too much about the neighbours and instead focus on avoiding the frequencys that have limitations on transmit power in your country.

Well I just configured antenna gain to 6 and TX to 22 and it boosted speed a lot! Thanks.

Where to find such paramaters.......gain tx power?

Where to find such paramaters.......gain tx power?

In the documentation:
https://help.mikrotik.com/docs/display/ROS/WifiWave2

You can find it in Wireless -> WiFi Wave2 -> then when you select interface you want to configure you will get this:


Here you have TX power and antenna gain. I leave antenna gain default and change tx power if needed.

But where does it show what the actual gain and tx power settings currently are??

In status tab you can see TX power, antenna gain probably only in CLI.

Also in webfig (don't know for CAP)
https://IP/webfig/#Wireless.Configuration

I didn't check the configuration details of OP.

But I have a hAP ax2 and hAP ax3, identical config, but ax3 has wireless coverage and performance issues.

Maybe it's the way I aligned the antennas or what? Possibly MikroTik can share antenna guide for this model, the blueprints of how the rotational feature of the antenna affects signal directionality?

Not really my expertise but one can assume those antennas on AX3 provide a doughnut shaped radiation pattern perpendicular to the axis of the antenna.
For AX2 I think it will be more spherical.
Measurement data (from certification ??) might be helpful indeed.

I also have AX3 and AX2 and find AX3 performing better (but I did pay attention placing one of the anntennas in such a way the radiation really should go where I think (hope) it should go.

Not really my expertise but one can assume those antennas on AX3 provide a doughnut shaped radiation pattern perpendicular to the axis of the antenna.
For AX2 I think it will be more spherical.

Measurement data (from certification ??) might be helpful indeed.

I don't know, but a crystal clear documentation from MikroTik on hAP ax3 antenna alignment guide would likely solve the wireless issues for 90% of the users in this forum.

Not really my expertise but one can assume those antennas on AX3 provide a doughnut shaped radiation pattern perpendicular to the axis of the antenna.
For AX2 I think it will be more spherical.

Measurement data (from certification ??) might be helpful indeed.

I don't know, but a crystal clear documentation from MikroTik on hAP ax3 antenna alignment guide would likely solve the wireless issues for 90% of the users in this forum.

Definitely would be useful to have documentation for signal coverage pattern of those antennas, and hAP ax² product. It would allow pick appropriate product for different property layouts (cube, wide flat,...). And for hAP ax³, why would sticks rotate around second axis, if it's doughnut shape?

And for hAP ax³, why would sticks rotate around second axis, if it's doughnut shape?

Exactly. I don't know how to properly align the antennas on hAP ax3, MikroTik made sure to keep this a secret for reasons I cannot understand.

I've had the hap ax3 for over 2 months and I'm deeply disappointed.
I tried different settings but the WIFI signal is very bad. This is the worst WIFI device I have ever tested in my life.
I hope I can sell and buy a Ubiquiti.
I no longer intend to use Mikrotik devices as Access Points.

I've had the hap ax3 for over 2 months and I'm deeply disappointed.
I tried different settings but the WIFI signal is very bad. This is the worst WIFI device I have ever tested in my life.

This is quite highly negative claim against hAP ax³. What are the details of yours deployment?

In my deployment, the hAP ax³ can penetrate two reinforced-concrete floors, or one floor and one thick (30cm wide) wall, or 30cm wall and 25m of air distance. The speeds aren't great after this, but 20~50Mbps is good enough non-data-intensive applications.

When it penetrated just one floor, with maybe 5m of air distance, I had speeds roughly of 500Mbps. More than my WAN bandwidth. Tested with a laptop with Intel AX210 wifi card, using iperf3, against NAS connected via gigabit ethernet.

It can't do two floors, and diagonally the 30cm thick load-bearing wall, that cuts the house in two halfs. I covered this dead spot with hAP ax².

Given the price of hAP ax³ and routing performance. It is quite a good bargain. I only don't like, that it's so ugly big.

Almost the same situation here at home !
AX3 hidden in small closet, AX2 for other corner of house.

Acceptable speeds for normal use.

My ax3 is upstairs and im testing downstairs and i get full connection speed on my phone. I don't see any problems...

Not sure what peoples expectations are with ax products, 5g doesn't do walls as well as 2.4g I would move that router to the bottom of #4 or put a cap where #4 is on your plan. Here are my speed results local and to a speedtest site from a hap ax2 and cap ax, pretty much in line with what I expect tbh. 4 bedroom house here, I think for me a hAP ax3 and cAP ax would have been a bad choice. But then i got my hAP ax2 before the hAP ax3 was released.
speedtest downstairs

speedtest upstairs


And for those wanting QOS figures, I'll only bother posting the upstairs cAP ax 75ms is max.

Agree with others that performance I’m with hap ax3 in my case is not what I expected… crappy isp router gives much more performance (could not follow the regulations, I do not know) after a couple of days os messing around with the wifi configuration I am close to isp speeds in 5Gh(ISP 550Mbps MT 450 Mbps) and half of the mark in 2.4Ghz (isp 120Mbps MT 60 Mnps) I assume I can’t get any better than this I am not an expert here and don’t know what else to change, the main advantage I got was to use channels mentioned above in this post. I think MT should improve this by default with out touching much stuffs in the configuration for their Home Routers and many of their potential users does not know about many things inside.

Since posting in this thread, I have been surprised by the very good performance of the wave2 drivers under 7.12 with my aging RB4011 with default values (not much to tweak anyway). I am reaching 900 Mbps with an ac class device using ax clients and 550 Mbps with ac clients. I have almost no interference at my house, but I would expect ax class devices to perform better - the numbers I have seen are at least close to 900 Mbps on MT. That is slightly lower than other brands, but you get a mostly stable OS (early be adopters are having lots of issues right now) with enterprise features that aren't available on most consumer devices.

However, interference will degrade the signal of any device and without very good knowledge of the components/software used under the hood (the closed-source stuff is always off limits), it is very hard to compare brands fairly. Then there's the actual location of the installation - I have seen reviews from YouTubers (with too much money apparently) where they managed to botch the installation of top of the line Ruckus/Meraki systems vs the low-cost solutions they were using before. I am not making excuses for MT, but I can attest that they have been improving the performance of their WiFi devices over time on my unit. Many versions back, I couldn't get more than 500–600 Mbps. Anyway, maybe try adding an AP from another brand if you really need that extra speed/range right now or wait for WiFi 7 to become mainstream.

The question is simple. The hAP ax3 is AX1800, it does not work with 160MHz. To pass 1GB via WIFI, it needs to be AX3000 standard. You can make any configuration that won't solve it. The issue is Hardware.

And your point is ... ?

Most likely spam