Gig+ Wireless test
equipment;
intel ax210 ( Wireless Kartı) Wi-Fi 6E
MikroTik RB4011iGS+5HacQ2HnD-IN (QCA9984) 802.11ac wave2
Maximum 1.1Gbit Wireless One Directional Speedtest.
good performance QCA9884, RouterOS Version 6.46.8



Gig+ Wireless test
equipment;
intel ax210 ( Wireless Kartı) Wi-Fi 6E
MikroTik RB4011iGS+5HacQ2HnD-IN (QCA9984) 802.11ac wave2
Maximum 1.1Gbit Wireless One Directional Speedtest.
good performance QCA9884, RouterOS Version 6.46.8



I wonder if MikroTik will provide Wi-Fi 6E support after how many years. When can it be ready for the technologies of the future?
This is same chip as audience.
Audience however has 4 chains for it.
I can’t get however more than 480mbit data transfer (with 1300mbit link, 3 chains 80mhz) as cpu seems to be limiting factor, where it uses fully 2 cores, while for some funny reason 2 others are unused.
Rb4011 ha much better cpu.
Can you check your cpu usage?
Most importantly if all cores are used or just two as in my case?
audionce will easily reach 1Gbps traffic in almost any scenario.
I don’t think there is a problem with the CPU. Test conducted using Traffic Generator. The CPU generating to ip package bit high as they produce 4011.

Are you sure you’re connecting using 160mhz?
In your 1st post you have mentioned wave2, so I assumed you are using new beta 7.1b4 which uses wifiwave2 impl.
But from your other screenshot I can see you are on stable version.
My question was only regarding wifiwave2.
I am using it as it improves massively range and link stability. Otherwise audience with ros6 is worse than my 7 years old AirPort Extreme which I am looking to decommission- for the better rather than worse.
However under wifiwave2 I have this strange cpu issue which limits my bandwidth to about ~500mbit transfer, tested with iperf
Hi ib254254
Interesting.
I tried RB4011 Wifiwave2 and Intel ac 9560 160Mhz client in Windows laptop.
For me speed is extremely disappointing! around 500Mbps (TCP iperf3 from separate laptop).
I saw similar speed on Audience and RB4011 with normal wifi package.
Interesting is that I see the same "PHY " rates than you, 1733mbps etc. but the final throughput is half of yours…!
Can you share the actual TCP speed you get on the client? I am kind of not trusting your Tx and Rx data rate numbers from within Winbox ![]()
This is what I am getting too: max 500mb with iperf and it seems to use only half of the cores so might be some temporary beta limitation/bug I hope, which I hope they will fix soon.
The main reason why I am using wifiwave2 is because it has much better range and stability for mobile devices.
Rest I have wired using 10gig MikroTik switches.
When I saw your gig results I started to worry that this might be audience limitation, but if you get it with much more powerful rb4011, then it must be software.
The reported result are bogus…
1753/3 = 583Mpbs is what is more likely with direct LOS and no interfering obstacles.
At 1300/3 = 433Mbps speed can be expected.
Rough rule of thumb (take advertised two way speed, divide by two for one way and then subtract losses). 1/3 rule is a good starting number.
Intel AX200 connected at 1.2Gbit/s at Aruba AP-555 with 80 MHz channel == stable 800 Mbit/s up and down while copy a big file to and from a SMB file server.
Tell the rest of the class what you paid for the Aruba.
Hahah what a tard, anuser…
https://www.cdw.com/product/hpe-aruba-ap-555-us-campus-wireless-access-point/5520246
Maximum data rate (2.4 GHz): 1.150 Mbps, Maximum data rate (5 GHz): 4.8 Gbps, Aggregate bandwidth: 5.95 Gbps
Thus using my rule of thumb, this unit should achieve one way throughput of 4.8/3 = 1.6Mbps plus minus. To be connecting at 1.2 and only getting 800 is a bit of a disappointment, you got robbed LOL.
If nothing else this demonstrates/proves, that the original poster is full of something that is not facts, and has an odour!
The reported result are bogus…
1753/3 = 583Mpbs is what is more likely with direct LOS and no interfering obstacles.
At 1300/3 = 433Mbps speed can be expected.Rough rule of thumb (take advertised two way speed, divide by two for one way and then subtract losses). 1/3 rule is a good starting number.
seems like a devastating result for you.
If you review my MuM Presentation below you probably will elvis can leave the brain ![]()
https://mum.mikrotik.com/presentations/TR18/presentation_5628_1539936230.pdf

https://www.cdw.com/product/hpe-aruba-ap-555-us-campus-wireless-access-point/5520246
WoW, for that price, I will rather by 6 x RB4011s and place them all over where needed ![]()