Hi,
The second Flint 2 Wi-Fi speed works perfect, and the first one works not very well even in front of the fan?
The first one works ok when in front of a fast fan. Without the fan its just 50-150 Mbit.
The second router does not need a fan for the full 5GHz speed.
Simply put the WIFI OFF and use an Unifi AP with PoE. Problem solved for ever. I did that, works like a charm.
Thanks for your reply.
What country/area do you live in?
I live in Germany. I'm already in contact with Cathy from the technical support.
1 Like
Question to @bruce:
Is there any chance of the power supply is creating (or not filtering) EMI, interfering to the radio device?
If this is the case, then this filter could work:
https://uk.farnell.com/schaffner/fn2010-3-06/filter-3a-chassis/dp/1191363
Sorry I cannot sure now. I will track this issue and this deivce.
Thanks for your reminding. 
Can EMI really influence 5GHz? I can try the device with a different 12V power supply, but both routers were using the same power supply.
EMI filters are usually designed to work between 10kHz to 10GHz.
It can also heat your devices unecessaryly.
If you have two Flint2 devices, you should have 2 power supplies. Try the another one.
Do not share the same power line of your microwave.
The second power adapter did not help.
With my Pixel 7 and AX210 adapter the speed is identical. I tested it this time with firmware 4.6.2, after a fresh reset, with no VPN. As always the upload is perfect but the download horrible (relatively seen).
The ping is too high...
Try to use the "tracert" to see where the highest latency is happening.
On Windows you can use pathping instead which will provide better overview. (Or at least tries to)
1 Like
Yes, the ping is too high while using WiFi under load on this device. On the 2nd Flint the load stats are fine.
Its also all fine via cable.
My internet connection is working as it should:
Source to Here This Node/Link
Hop RTT Lost/Sent = Pct Lost/Sent = Pct Address
0 192.168.8.183
0/ 100 = 0% |
1 0ms 0/ 100 = 0% 0/ 100 = 0% 192.168.8.1
0/ 100 = 0% |
2 1ms 0/ 100 = 0% 0/ 100 = 0% 192.168.179.1
0/ 100 = 0% |
3 3ms 0/ 100 = 0% 0/ 100 = 0% xxxxxxx.versatel.de [94.134.xx.xx]
0/ 100 = 0% |
4 --- 100/ 100 =100% 100/ 100 =100% 224-221-142-46.pool.kielnet.net [46.142.221.224]
0/ 100 = 0% |
5 7ms 0/ 100 = 0% 0/ 100 = 0% 80.150.171.166
0/ 100 = 0% |
6 10ms 0/ 100 = 0% 0/ 100 = 0% ffm-b11-link.ip.twelve99.net [213.155.129.188]
0/ 100 = 0% |
7 7ms 0/ 100 = 0% 0/ 100 = 0% ffm-bb1-link.ip.twelve99.net [62.115.124.116]
0/ 100 = 0% |
8 5ms 1/ 100 = 1% 1/ 100 = 1% ffm-b16-link.ip.twelve99.net [62.115.132.227]
0/ 100 = 0% |
9 8ms 0/ 100 = 0% 0/ 100 = 0% plusline-ic-323934.ip.twelve99-cust.net [62.115.153.153]
0/ 100 = 0% |
10 7ms 0/ 100 = 0% 0/ 100 = 0% 82.98.102.79
0/ 100 = 0% |
11 7ms 0/ 100 = 0% 0/ 100 = 0% 82.98.103.3
0/ 100 = 0% |
12 6ms 97/ 100 = 97% 97/ 100 = 97% 212.19.61.13
0/ 100 = 0% |
13 7ms 0/ 100 = 0% 0/ 100 = 0% redirector.heise.de [193.99.144.80]
Trace complete.
You need to run the pathping on the device where there is the problem...
Try the online tool:
https://pingtool.org/
Wanted to show you with the first one, that the latency of my internet connections in general is fine.
Here on the same device but via WIFi this time while downloading a file with only 50Mbit (of 950), 20cm away from the Flint 2.
Computing statistics for 250 seconds...
Source to Here This Node/Link
Hop RTT Lost/Sent = Pct Lost/Sent = Pct Address
0 [192.168.8.183]
2/ 100 = 2% |
1 543ms 8/ 100 = 8% 6/ 100 = 6% 192.168.8.1
0/ 100 = 0% |
2 536ms 5/ 100 = 5% 3/ 100 = 3% 192.168.179.1
0/ 100 = 0% |
3 542ms 5/ 100 = 5% 3/ 100 = 3% xxxxxx.versatel.de [94.134.xxx.xxx]
0/ 100 = 0% |
4 --- 100/ 100 =100% 98/ 100 = 98% 224-221-142-46.pool.kielnet.net [46.142.221.224]
0/ 100 = 0% |
5 560ms 7/ 100 = 7% 5/ 100 = 5% 80.150.171.166
0/ 100 = 0% |
6 541ms 4/ 100 = 4% 2/ 100 = 2% ffm-b11-link.ip.twelve99.net [213.155.129.188]
0/ 100 = 0% |
7 532ms 2/ 100 = 2% 0/ 100 = 0% google-ic-324085.ip.twelve99-cust.net [62.115.153.213]
2/ 100 = 2% |
8 --- 100/ 100 =100% 96/ 100 = 96% 108.170.236.175
0/ 100 = 0% |
9 --- 100/ 100 =100% 96/ 100 = 96% 142.250.213.213
0/ 100 = 0% |
10 544ms 4/ 100 = 4% 0/ 100 = 0% dns.google [8.8.8.8]
I also noticed that my WAN ports starts having packet loss. LAN1 as WAN works fine. Test were done with LAN1 as WAN.
If same network cable, same original WAN, same configuration of the 2 MT6000, I think it supposed to be same result, it is weird. 
1 Like
Between the flint 2 whos having packet loss and the flint 2 which doesn't.
Does the kernel log mentoin the same hardware naming when detecting the ethernet port?
It needs to be RTL8221B-VB-CG.
I remember a commit in openwrt branch which added the commit as official support from Linux, but it seems there are two revisions of this hardware component and according the commit only one was present in very early examplars.
Though i doubt it is true, because ive checked my router from the beta test period and the one as early bird they show both the same hardware name.
But that is the only possible component difference i can think off between the two.
Finding that commit is like finding a needle in a haystack, but i do remember to have readed such thing.
2 Likes
I will share this to R&D to try to find.
The sample of the user bonefire will be returned to our lab, I will follow it.
3 Likes
I can see that eth1 and lan1 on the faulty device use the RTL8221B-VB-CG.
Edit: Packetloss on WAN is solved, my second Flint 2 was using the same MAC as the first one because of the imported config (symptoms were immediately noticeable, like speedtest breaking up with connection errors, therefore i used LAN1 instead on my first device).
But this still doesn't explain the WiFi problems (btw. random BSSID is activated on both devices). Today the performance improved, but slowdowns to 50Mbit still happened.
Which leads me to another theory: WiFi interferences/overlapping or jamming
If someone would use the same channel as my device and use the the full bandwidth would it be equally shared among others on the same channel?
I noticed in the past that all my used WiFi APs change the channel back and forth a lot. My Fritz repeater even complained that the channel is jammed/faulty. How high is the chance that something like this would massively reduce the WiFi speed only for the downstream?
Edit2: Tested the router this early morning. I really doubt that there is any other traffic on the band. Instantly locked to <50Mbit download and full upload speed again.
2 Likes