This topic is temporarily closed for at least 4 hours due to a large number of community flags.
This topic was automatically opened after 4 hours.
Hello everyone Here comes our Close source WIFI driver version firmware. We are conduct the test at the same time. you can download and test if you want to try it ahead of time.
5 Likes
How scared should we be about this version? 
2 Likes
The test version is OpenWrt 21.02-SNAPSHOT r15812+1066-46b6ee7ffc.
It’s similar to GL.iNet’s GL-MT6000 snapshot version, but it’s been built from an older branch with a different set of changes applied e.g. an older DDNS version, OpenWrt feeds, different edits to various scripts, 1 or 2 different modules etc.
3 Likes
Thanks Alex.
Can you give an overview of the changes to this alpha build?
If this test proves positive for the wifi stability, does this mean you will regress the version of openwrt to an older version or merge the driver and code uodatws into the latest openwrt (23.05)?
Thanks
Are you able to change the WiFi transmit power in this version?
I haven’t installed it to check, since I don’t currently have the time to test firmwares. But I’d assume you can’t configure them settings without a LuCI app package, like this.
I just hope that this firmware is never released as a beta or stable build, since a downgrade to an old end of life version of OpenWrt isn’t an acceptable solution for new hardware. But as a test build, just to prove that the issues with the mt76 driver, it’s fine.
5 Likes
I’m unable to change the WiFi tranmission power.
According to the WiFiman app its 11dbm on both bands.
Otherwise in a quick test the WiFi seems to run faster, smoother and more responsive. Just the distance is lowered because of the TX power.
Guess this is because of the closed source driver - maybe the connection between OS and driver isn’t completely done yet.
No, it’s not 100 dBm. 
3 Likes
@alex_zheng @alzhao and other admins, thank you for allowing us to use alpha versions for testing. However, could you include a changelog of the changes you made? Otherwise, we will always question whether there have been changes in 2.4 GHz and other features.
3 Likes
OpenWrt Version: 21.02
Linux kernel: 5.4.238
@bonefire Have you change the TX power in the Web ? or just default ?
Can you SSH the teminal to see the tx power?
@admon maybe something wrong with luci. 
For a stable WIFI , We try a Close Source WIFI driver. And at the same time, OpenWRT need to downgrade to 21.02 and Linux kernel: 5.4.238.
based on the version @robotluo release. Also fix a eth disconnection issue.
Does driver code will be updated new version openWRT in the future?
1 Like
I changed the txpower in the web.
iwconfig ra0
does not show any tx power result.
also
iwconfig ra0 txpower 20
does not work
The WebUI shows under Current power mW = dBm
I’m not sure if WiFiman is correct with these information, but most of the time I used it, it was accurate.
I will try it later in a field test.
In Europe the DFS channels around #100 can use about 30dBM. I tried to manually edit the wireless config country but still could not adjust the txpower.
The country code settings seems to be stuck to World 00.
1 Like
I wonder if Qualcomm would have been a better choice for the CPU, the Flint 1 is super stable with its Qualcomm quad core processor, MediaTek needs to do something about this as fast as possible. Either way we will always see minor bugs when a new product is released, that’s something normal when it comes to technology, I can’t wait to see the future of the Flint 2 which is a beast
Here are my first test results.
4.5.7 provides a better range and speed on 80 MHz compared to the Flint 1. My record is 178m away from the router and around 100Mbit Download speed (in the middle of the a busy city).
The signal strength on both devices is often very similar but the Flint 2 a few dbM better.
One thing which I don’t understand is that the 5GHz range is better than the 2,4 GHz range on both devices.
My conclusion is that the Flint 2 sends with a higher power than reported in “WiFiman”, but the real value is still unknown.
I will setup the Flint 2 tonight as a public WiFi+Wireguard and let you know the results.
How can i get rid of the HT WiFi channels?
Edit:
It looks like I can only have one SSID per band
If i create a second SSID on a band, it shows up as a HT_Channel, which is not functional.
I hope this gets fixed soon.
Edit2: This applies to the firmware 4.5.7. from google drive.